Home

Snowmelt Runoff Model (SRM) User`s Manual

1. lt i LS Snowmelt Runoff Model SRM Version 4 0 Program Options Model Run Number 17 Start Date MMDD 0401 Units 0 Metrac f Engiash d End Date MMDD 0831 Model Mode 0 Temperature Average F C 0 O Simulation 1 9 Updating O Daily Mean 1 Max Min Temperature Input 1 Temperature Lapse Rates 0 O basin wide 1 by zone O basin wide 1 by zone Brecrprrtation Enput Runoff Coefficients 1 O basin wide 1 by zone O basin wide 1 by zone Critical Temperature 1 Actual Runoff Available 1 O basin wide 1 by zone 0 No 1 Yes Comments Beginning with Version 3 10 three comment lines are provided that allow a user to describe the current data making it easier to keep track of Individual data frles 25 65 bbe a a Geb bee he Os Os Oe Ge EscQuit FlHelp F2Summary F3FilelO F5Plot F6Print F7Compute Program Options Data Entry Screen 156 G 3 5 Program options The Program Options data entry screen displays run specific data values These items are initialized by the model on program entry G 3 6 Basin definition The Basin Definition data entry screen is used to enter basin wide values e g zone area and hypsometric mean elevation which identify and physically define the subject basin G 3 7 Basin variables parameters The Basin Variable Parameter data entry screen manages the process of providing daily values for each required variable and parameter for the entire snowmelt period and for each
2. Place your comments here Help Wildy e ona sereen element Tor help Click on the question mark to hide these help comments Window 16 Climate Change Scenario Definition Window 100 A simple example is to add 4 to each daily temperature value in a rough attempt to examine the effect of global warming Each scenario is stored in the database as a separate uniquely named table 10 3 5 3 Data entry and selection Select a scenario from the Existing Scenarios for combo box in the upper left corner of the window The scenario rule information is load into the window To create a new scenario click the button A naming dialog will display Provide a unique scenario name in response Enter a Winter End Date if the default provided by WinSRM is unsatisfactory The default winter end date is the date halfway between the simulation starting and ending dates A maximum of 500 rules per scenario may be defined one per line in the Climate Definition grid Each rule consists of the following component parts e Variable Parameter Each grid cell is a combo box containing the variables and parameters for which WinSRM will accept climate changes e Date the rule must specify the time period for which this change applies e Edit Action a combo box of the v D Add to each day Multiply each day Shift backward forward Normalize temperature and precipitation only LI UU e Edit Factor the
3. UI April May June July ALC Sept Fig 41 Transformation of runoff in the Rio Grande basin near Del Norte from runoff in 1979 to runoff in 1977 using temperature and precipitation of 1977 as new climate for 1979 71 8 6 WinSRM to improve real time runoff forecasts With WinSRM it is no longer necessary to assemble a set of modified depletion curves MDC s as shown in Figure 21 in order to forecast the future course of the conventional depletion curves CDC s In a new basin where these graphs are not yet available it is sufficient to have one historical year with good satellite data on snow covered areas and the usual daily values of temperature and precipitation Based on this data set runoff starting on 1 April can be forecasted as follows 1 Runoff is simulated for the available historical year in order to verify the preselected model parameters The climate program is run for the winter half of the year with the known values of precipitation and temperature T and P of the forecast year F Snow water equivalent Hw on 1 April of the historical year A is evaluated using the modified depletion curves MDCexcr a Snow water equivalent on 1 April of the forecast year is evaluated from the winter deficit or excess as shown in Equation 24 and the curve MDCexcr rf 18 derived accordingly The runoff in the summer half of the year is composed of a guaranteed runoff volume from MDCexcr F taking into account th
4. 10 3 1 Welcome to WinSRM Window 10 3 1 1 Purpose This windows displays introductory information to the user To skip this introduction in the future click in the entry box in the lower left corner The user may read additional help information or go directly to WinSRM The window can be subsequently displayed from the main program window by clicking the Help Show WinSRM Welcome menu item Welcome to ORM TATUNG Snowmelt Runoff Model for Windows Version 1 11 USDA an BiS Disclaimer Although WinSRM has been tested by its developers NO warranty expressed or implied ts made as to the accuracy and functioning of the program and reloted program material nor shall the fact of distribution constitute any such warranty and NO responsibility is assumed by the developers in connection therewith Check hereto 2kip this introduction in ihe future Exit program Fin User E Star Window 1 Welcome to WinSRM Window 10 3 1 2 Buttons Click to begin using the WinSRM software Mew User Click New User to display introductory help text Exit program Click to exit WinSRM This button is only visible at program startup 88 10 3 2 WinSRM main window 10 3 2 1 Purpose The WinSRM Main Window controls the processes and functions that make up the snowmelt runoff modeling environment The window allows the manual input of the basic information required by the model to define a basin for which modeli
5. 30 5 3 2 Degree day factor a The degree day factor a cm C d converts the number of degree days T C d into the daily snowmelt depth M cm M a T gt Degree day ratios can be evaluated by comparing degree day values with the daily decrease of the snow water equivalent which is measured by radioactive snow gauge snow pillow or a snow lysimeter Such measurements Martinec 1960 have shown a considerable variability of degree day ratios from day to day This is understandable because the degree day method does not take specifically into account other components of the energy balance notably the solar radiation wind speed and the latent heat of condensation If averaged for 3 5 days however the degree day factor is more consistent and can represent melting conditions The effect of daily fluctuations of the degree day values on the runoff from a basin as computed by SRM is greatly reduced because the daily meltwater input is superimposed on the more constant recession flow Equation 1 The degree day method requires several precautions l The degree day factor is not a constant It changes according to the changing snow g y g g gmg properties during the snowmelt season 2 If point values are applied to areal computations the degree day values must be determined for the hypsometric mean elevation of the snow cover in question and not for example for the altitude of the snow line 3 If the snow cover is sc
6. 14 cm P 16 cm Po 18 cm Rainfall contributing area option 0 Zone A 100 km P rain 14 cm S 0 4 Zone B 100 km P rain 16 cm Sp 0 6 Zone C 100 km P rain 18 cm Sa 05 14cm 1 S 100km 16cm 1 S 100km 18 cm 1 S 100 km P rain SS 300 km 6 1cm gt k x 40 on 5 consecutive days The threshold value can be changed in order to activate the rainfall peak program for smaller rainfall amounts or to delay activation until higher rainfall amounts are reached By putting the threshold to 0 cm the program will be activated on each day with rainfall in an earlier version of the computer program the threshold 0 cm resulted in automatic switching to a threshold of 6 cm With a 0 cm threshold the recession coefficient will be continuously decreased and SRM is likely to overestimate the runoff By putting the threshold higher than the highest daily precipitation Equation 15 is eliminated and SRM will probably underestimate the sharp runoff peaks from heavy rainfall 40 5 3 7 Time Lag L The characteristic daily fluctuations of snowmelt runoff enable the time lag to be determined directly from the hydrographs of the past years If for example the discharge starts rising each day around noon it lags behind the rise of temperature by about 6 hours Consequently temperatures measured on the nth day correspond to discharge between 1200 hrs on the nth day and 1200 hrs on the n 1 day Dischar
7. 35 4 33 50 53 54 24 33 15 18 62 68 Publ 1987 2002 1998 2002 1980 1981 2002 1989 1990 1992 1986 1991 1981 1987 2002 1994 1998 1985 2000 2000 2002 1991 1988 1994 2000 1997 1991 1992 1991 2000 1987 Y ear appl 1982 1985 1999 1990 1993 94 1999 1973 79 1999 1981 1987 1987 1975 1973 1979 1982 1985 1999 1984 1986 lt 1998 1985 lt 1999 lt 1999 1986 199 1976 1981 1983 84 1996 1985 1991 1987 1989 1994 1983 75 76 Ge 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 Country Switzerland Argentina Switzerland Argentina USA France USA Uzbekistan Morocco Austria USA USA Chile USA Switzerland Switzerland USA USA Chile India USA Uzbekistan Canada Argentina Canada Turkey Uzbekistan Kyrgystan Tadjikistan Basin Inn at Tarasp Alps Tupungato Andes Inn Martina Alps Chico Tierra del Fuego Boise Rocky Mts Durance Alps Madison Montana Pskem Tillouguit Atlas Salzach St Johann Alps Henry s Fork Idaho Cache la Poudre Colorado Aconcagua en Chababuquito Andes Sevier River Utah Rhine Felsberg
8. 46 48 63 66 67 37 75 68 3 13 76 58 78 21 26 78 23 21 57 Publ 2002 2002 1995 1997 1984 1989 1995 2000 2000 1996 1985 1986 1997 1979 1999 1995 1980 1996 1987 1983 1998 1992 1992 1999 2002 1985 1997 2002 2002 1981 Year appl 1999 1999 1992 1993 1979 1980 1985 1993 1996 1998 1996 1998 1989 1994 1983 1992 1979 1985 1985 1993 1974 1976 1984 1991 1993 1991 1991 1996 1998 1999 1984 1985 1999 1999 1976 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 Country Switzerland Spain USA Spain USA Spain Argentina USA Chile Poland India USA Switzerland Spain China Austria Argentina China China Uzbekistan India Canada Spain India Chile New Zealand Switzerland USA Basin Tiefencastel Alps Valira en Seo d Urgel Pyrenees Towanda Creek Applachians Garona en Pont de Rei Pyrenees South Fork Colorado Noguera Ribagorzana en Pont de Suert Pyrenees Las Cuevas en Los Almendros Andes Independence R Adirondacks Mapocho Andes Dunajec High Tatra Saing Himalayas Conejos
9. The name of the simulation that owns the SimulationFile Text scenario The name of the scenario table that contains ScenarioFile Text the definition rules for the climate change scenario 3 WinterEndDate The last day of the winter half year CreationDate Date the scenario was last changed Comments Description of the climate change ClimateScenario Table 147 The ClimateScenario table is the structure that contains the climate change scenario for a specific simulation in the database It occurs 1 n times within a database Each table contains 1 n records Each record contains a rule 9 3 5 3 for temporarily altering a variable parameter to formulate the new climate for the simulation See Example 3 Field Name ata Type 1 VarParm Integer for more information on climate scenarios Description The number of the variable parameter affected by the rule see Table D 2 The starting date of the period for which the rule applies The ending date of the period for which the rule applies EditAction The edit action to be performed see Table D 2 5 EditFactor Single Value used for the edit action Integer The elevation zone s to be modified by the rule women DREES eener egene Ds ran tcs Table D 1 Numeric equivalents to variables and parameters ES Add edit factor to each day Multiply daily value s by edit factor Shift values forward edit factor number of day
10. and the Clim run in summer is the effect of the climate change on the summer runoff In theory identical Prom Tnorm and Snorm Should be derived be derived from any selected year of present times Such a result cannot be expected in view of some uncertainties involved in the procedure However a normalized data set of T P S may be considered preferable to represent the present climate in comparison with the data set from a year which happens to be available but deviates from normal conditions of the present climate Years with little snow are not suitable for derivation of a normalized year if there is no snow in an elevation zone the normalized depletion curve of the snow coverage cannot be derived The normalized precipitation and temperatures are treated as a changed climate In a climate run a shift of parameters is sometimes advisable as mentioned in Section 8 4 It should be noted that the shift becomes automatically part of the climate scenario For example if the seasonably increasing degree day factors are shifted to earlier months then higher melt rates result but at the same time the CDCcLim curves are decreased If the users want to adjust the normalized runoff by shifting the degree day factor they must cancel this automation as described in chapter 10 By taking this precaution a parameter can be changed by shifting and the resulting runoff will be adjusted accordingly 72 Example using actual n
11. current for the reader one can search on SRM home or WinSRM to locate a current site So far four SRM workshops in 1992 1994 1996 and 1998 have been organized at the University of Bern Switzerland with about 130 participants from 20 countries taking part A fifth SRM workshop was organized in 2005 at New Mexico State University In addition the authors are available to assist users in overcoming special problems which may be encountered 2 INTRODUCTION The Snowmelt Runoff Model SRM is designed to simulate and forecast daily streamflow in mountain basins where snowmelt is a major runoff factor Most recently it has also been applied to evaluate the effect of a changed climate on seasonal snow cover and runoff SRM was developed by Martinec 1975 in small European basins Thanks to the progress of satellite remote sensing of snow cover SRM has been applied to larger and larger basins Recently the runoff was modelled in the basin of the Ganges River which has an area of 917 444 km and an elevation range from 0 to 8 840 m a s l Contrary to the original assumptions there appear to be no limits for application with regard to the basin size and the elevation range Also a dominant role of snowmelt does not seem to be a necessary condition It is however advisable to carefully evaluate the formula for the recession coefficient Runoff computations by SRM appear to be relatively easily understood To date t
12. winter zonal meltuormaL 2winter zonal meltc im from Steps 1 2 above Reduce enhance effect of temperature induced deficit by adding any net change to seasonal precipitation resulting from the climate change scenario In the case of Illecillewaet 1984 zones C and D the scenario T 4 C P 1 2 results in a positive Winter Change increases in P overwhelm the winter deficit in melt In that case MDCexcy is stretched rather than cut off as explained in Step 5c The Climate Change Progress Screen allows a user to over ride a calculated zonal Winter Change value at this point in the processing sequence See Section 8 3 for a description of when such intervention might be advisable 5b For each zone develop zonal melt curve AZMexcy X a T S vs Time for the normal climate Find the date along AZMegxcL where zonal Winter Change is equaled or exceeded At this point the Climate Change Statistics Report is produced This printout disk file if no printer is available details the calculated values used by SRM to compute zonal Winter Change and the cutoff points or gain factors used in Step 5c to derive MDCexcr wa Sc Create data for a new curve MDCexcy wa To understand the methodology for modeling CDC im it is important to understand the MDCexc curve Each point along the curve is a daily intersection of snow water equivalent independent of melt season snowfalls a T newmelt and daily snow covered area
13. 20546 U S A 1992 Updated Edition Version 3 2 Hydrology Laboratory Technical Report HL 17 USDA Hydrology Laboratory Beltsville MD 20705 U S A 1994 Updated Edition Version 3 2 Geographica Bernensia P29 Department of Geography University of Berne Switzerland 1998 Updated Edition Version 4 0 Geographica Bernensia P35 Department of Geography University of Berne Switzerland 1998 Russian Edition Version 4 0 Deoartment of Geography University of Berne Switzerland 1999 Spanish Edition Version 4 0 USDA Hydrology Laboratory Agricultural Research Service Beltsville MD 20705 U S A 2008 Updated Edition for Windows WinSRM Version 1 11 USDA Jornada Experimental Range New Mexico State University Las Cruces NM 88003 U S A ations Around the World SNOWMELT RUNOFF MODEL SRM USER S MANUAL Updated Edition 2008 Windows Version 1 11 J Martinec Consulting Hydrologist Davos Switzerland A Rango Jornada Experimental Range USDA Agricultural Research Service ARS Las Cruces New Mexico USA R Roberts Hydrology amp Remote Sensing Laboratory USDA ARS Beltsville Maryland USA Edited by E Gomez Landesa M P Bleiweiss New Mexico State University Las Cruces New Mexico USA Contents Ms EE 13 Z INTRODUCTION dada As 13 3 RANGE OF CONDITIONS FOR MODEL APPLICATION ccceceececeeceeeseeteeeeeeseeeeeeeeeseeeaeeeeasaetaeeeeegs 14 A MODEL STRUCTURES sia 19 5 NECESSARY DATA FOR RUNNIN
14. 34 71 73 4 5 4 8 Publ 2000 1989 1995 1991 1985 1986 1984 2002 1989 1984 1984 1992 1987 1998 2000 1997 2000 1982 2000 1992 199 1990 2002 1982 1991 2002 2001 2002 1996 2002 Year appl 1996 1981 1990 N A 1976 1978 1975 1979 1976 1978 lt 2002 1979 gt 1991 1976 1979 1983 1987 1983 1982 85 88 89 92 94 96 1985 1976 77 1979 1973 75 82 88 1982 1988 1986 87 1989 1967 N A N A 1997 1999 1996 98 1999 103 104 105 106 107 108 109 110 111 112 Country Uzbekistan Kazakhstan USA USA Tadjikistan Kyrgystan Pakistan Tadjikistan Afghanistan China India Bangladesh India Bangladesh Basin nee Kafinirgan 12369 Sevier River Utah 13380 Snake River Idaho 14897 Vakhsh 37759 Naryn 33237 Kabul River Himalayas 63657 a Pamirs and Hindu 120534 een Shan GE Brahmaputra Himalayas 547346 Ganges Himalayas 917444 18 Elevation Range m a s l 505 3005 1506 3719 1524 4196 1791 5291 800 5000 305 7690 2141 5564 2500 5224 0 8848 0 8848 0 57 0 93 0 90 0 63 0 96 0 66 0 65 N A 0 75 0 94 8 6 4 0 0 4 2 8 1 0 6 0 5 6 N A 8 3 Years seasons Zones 7 Ref 8 68 31 8 4 11 8
15. If the increase in k with size appears to be too large the exponent may be replaced with A Ay Even if the envelope line in Figure 10 can be reliably derived in a basin it is possible that the resulting k values may be too low to represent average conditions during the snowmelt season especially in large basins In such cases the SRM model will react too quickly to any change of the daily input The simulated peaks would be too high and the simulated recession too fast A quick improvement is possible by deriving a new x and y not from the envelope line but from an intermediate or medium line between the envelope and the 1 1 line This modification may especially be needed if the runoff simulation is extended to the whole year Recession coefficients which may be right for the snowmelt period are usually too low for the winter months so that the simulated flows drop below the measured minimum values ee E AA a e 0 1 e Da U5 1 0 2 345 10 20 30 40 50 100 200 300 500 1000 Q ms Fig 11 Range of recession coefficients k related to discharge Q resulting from various evaluations In Dischma the range results from using either the envelope line or the medium line in Figure 10 In Modry Dul the relation varies in different years In Felsberg the relations 1 and 2 are derived from the size of the basin by two alternative formulas Martinec amp Rango 1986 In very small basins noticeable differences in the recession flow
16. J Martinec A Rango R Roberts Snowmelt Runoff Model SRM User s Manual Edited by Enrique Gomez Landesa amp Max P Bleiweiss SRM Applications Around the World OR RAM F s PROG casts w RUNOFF D CHANGE wee cT oF CLIMA AND E Acknowledgment It has taken the efforts of many people and the support of their organization during the last several years to allow us to reach this new milestone in snowmelt runoff modeling The following organizations and people were particularly helpful and supportive USDA ARS Jornada Experimental Range Las Cruces NM USDA ARS Hydrology and Remote Sensing Laboratory Beltsville MD Bureau of Reclamation El Paso TX Mr Michael Landis Rio Grande Basin Initiative New Mexico State University Las Cruces NM NMSU Water Task Force New Mexico State University Las Cruces NM Mr Craig Runyan US Army Corps of Engineers Albuquerque NM Ms Gail Stockton US Army Research Laboratory WSMR NM McElyea Family Endowment for Water Research SRM SNOWMELT RUNOFF MODEL USER S MANUAL UPDATED EX TIONFOR WINDOWS WinSRM Version 1 11 February 2008 Jaroslav Martinec Albert Rango amp Ralph Roberts Edited by Enrique Gomez Landesa amp Max P Bleiweiss This publication is an updated edition of the Snowmelt Runoff Model SRM User s Manual featuring the new computer program WinSRM Version 1 11 Printing history 1983 NASA Reference Publication 1100 Washington D C
17. 10 x Output Definition Available Reports Check Box any combination of active choices may be checked Temperature yale Inactive choices a grayed out Current data description Degree Day Factors Runoff Coefficients m Lo Option button used when Zone only one of two or more F Melt Di choices is appropriate ntibuting Area Lag Time n Snow Covered Area 5 1 Contributing Fain Cpr Command button Pri o Printer File Report Viewer Help All print bone Fig 46 Additional GUI objects used by the WinSRM interface Feature Checkbox Used to display multiple choices from which the user can select one or more Used to display multiple choices from which only one Option button choice may be selected Begin interrupt or end a process When chosen a command button appears pressed in Access the button functions using the keyboard by pressing the Alt key then the corresponding hot underlined key Command button 87 10 3 WinSRM Window descriptions As mentioned previously the WinSRM computer program interacts with the user through a collection of specialized windows and window objects Each window used by the model provides specific functionality designed to address the tasks inherent in computer modeling The window descriptions that follow are presented in more or less the order that a user might see them upon initial inspection of the program
18. AZMexc with new snow zonal melt excluded and AZMexcy wa derived from AZMexcy by cutting it off on 27 April and transferring it to 1 April 100 J N 15 L d i 50 25 Fig 31 Modified depletion curve adjusted for the winter deficit and including new snow of the changed climate MDCcrm wa derived from MDCexcr wa for zone A 67 Fig 32 Effect of a changed climate T 4 C on snow covered areas of 1979 in elevation zones A B and C of the Rio Grande basin near Del Norte Colorado CDCcrm wa 1s shifted from the original CDC due to a reduced snow cover on April and due to increased temperatures in the snowmelt season Fig 33 Climate affected runoff T 4 C in the Rio Grande basin near Del Norte Colorado compared with the runoff simulated by data of 1979 as shown in Figure 27 for April September 68 250 200 a A A a AAA ta A me 150 Ki DEA m s 100 eee 1 50 A X XI H um NW V Vi Vit MI IX Fig 34 Simulated runoff in the Rio Grande basin near Del Norte Colorado in the hydrological year 1979 and climate affected runoff computed by increased temperatures T 4 C and correspondingly changed snow conditions Figure 33 shows the climate affected runoff computed by original precipitation temperature T 4 C and snow cove
19. C per 100 m critical temperature Terr 0 75 to 3 C and time lag L 9 hours Components of Equation 28 were computed for a hypothetical temperature increase Ton T 4 C In the interest of a simple demonstration we used an average 79 runoff coefficient with the same value for snowmelt and for rain Cs Cr 0 7 for the period from July to September and cs Cr 0 8 for October The precipitation remained unchanged Figure 43 shows the depletion curves CDC and CDC cy for the glacier zones C and D It appears that the glaciers were covered with snow until the end of September 1984 With T 4 C they became exposed in July and glacier melt took place Computed hydrographs for temperatures of 1984 and for T 4 C are shown in Figure 44 In basins without glaciers for example the Rio Grande near Del Norte Figure 34 runoff is redistributed by a temperature increase but the runoff total volume remains practically the same 1 212 10 m in 1979 and 1 193 10 mr for T 4 C In the Illecillewaet basin Figure 44 these total volumes are 1 665 10 m in 1984 and 1 807 10 m for T 4 C This runoff increase can be roughly attributed to glacier melt For a more accurate result components of the water balance in Equation 28 can be computed as follows Roum R1984 156 44 cm 144 14 cm 12 30 cm Boun P 0 storage of unmelted snow 1984 T 4 C zone A 0 0 zone B 0 0 zone C 1 79 cm 0 zone D 10 86 cm 2 03 cm For the
20. Critical Temperature Lag Time Parameters for the current simulation This report is for one zone or all zones depending on the parameters distribution modifiers A table of the physical data for degree day temperature precipitation and Zone Degree Days DD snow cover for each elevation zone DD values are calculated during a Observed Precipitation simulation If changes are made to the input data afterwards the table Snow Covered Area S will display for each DD value and add a footnote to the table requesting a new simulation run Melt Depth M S Melt New Snow M 1 S Contributing Rain Cpr A table displaying three important zonal values derived by the model during the simulation A table comparing daily basin runoff derived by the model with observed daily basin runoff if available Run statistics are included in this report Measured vs Computed Snowmelt Runoff A summary of bimonthly values 1st 16th for 7 model parameters is nput Summary Repor printed for each zone in the basin Run statistics are for the most current model simulation Climate Change Deficit The Climate Change Deficit Gain Summary presents the computed Gain Summary values used to derive climate affected snow cover Run Statistics 113 10 3 13 File display window 10 3 13 1 Method of access The File Display window also referred to as the Report Viewer is displayed from the Output Definition window Window 29
21. Department of Geography and Department of Earth and Space Sciences University of California at Los Angeles 1255 Bunche Hall P O Box 951524 2002 Sommerfeld R A Musselman R C Wooldridge G L amp Conrad M A 1991 The performance of a simple degree day estimate of snow accumulation to an Alpine watershed In XX General Assembly IUGG in Vienna 1991 IAHS IUFRO Symposium of Snow Hydrology and Forests in High Alpine Areas IAHS Publication No 205 pp 221 228 127 73 Sorman U Uzunoglu E Kaya H 1 2001 Application of the SRM and SLURP Models in Eastern Turkey using Remote Sensing and Geographical Information Systems Remote Sensing and Hydrology Proceedings of the Santa Fe Sysmposium April 2000 AHS Publ No 267 pp 81 86 74 Span L 1978 Schneeschmelzabflusse in einem randalpinen Einzugsgebiet Snowmelt runoff in an alpine basin Technical Report Technical University Munich Germany 1978 Zulassungsarbeit fur das Lehramt an Gymnasien 75 Swamy A N amp Brivio P A 1996 Hydrological modelling of snowmelt in the Italian Alps using visible and infrared remote sensing Int Jour Remote Sensing pages 3169 3188 76 Viskum Jorgensen P 1992 Application of the SRM Model in a high mountain basin in Norway using NOAA AVHRR data Technical Report NN Norway 1992 Presentation at the 2nd SRM Workshop at the University of Berne Switzerland 77 Wang Ch 1994 Snowmelt Runoff
22. Durance Basin Basin Variables Zone A Precipitation B Function Keys Apr Value May Value May Value F5 Prior zone F6 Next zone E Duplicate an existing zone F8 Repeat prior value 1 time F9 Repeat prior value n times F10 Adjust day s Esc Exit process Fl Help PQUD Eilon montas PgDn Next 2 months Variable Precipitation Data Entry Screen G 3 8 Climate change processing control screens Beginning with Version 4 0 the Snowmelt Runoff Model supports an expanded climate change modeling component that supports year round climate change modeling see Section 8 for a complete description of SRM s climate change approach Two additional screens have be added to the computer program to manage the additional complexity required to support this new capability G 3 8 1 Climate change control screen The Climate Change Control data entry screen provides the user with a template for defining the winter and summer periods of a hydrologic year and the seasonal climate change scenario associated with each During climate change processing steps SRM modifies the affected variables parameters as described by the scenarios detailed on this screen The two variables and five parameters listed on the screen can be modified for a new climate in two basic ways Shifting The model user may want to shift parameters most affected by climate to an earlier time period in the hydrologic year to reflect the new climate The shift f
23. Martinec J Rango A 1986 Parameter values for snowmelt runoff modelling Hydrol 84 197 219 Martinec J Rango A 1989 Merits of statistical criteria for the performance of hydrological models Wat Resour Bull 25 20 421 432 Martinec J Rango A amp Major E 1983 The Snowmelt Runoff Model SRM User s Manual NASA Reference Pub 1100 Washington D C USA 120 Martinec J Rango A amp Roberts R 1994 The Snowmelt Runoff Model SRM User s Manual ed by M F Baumgartner Geographica Bernensia P 29 Department of Geography Univ of Berne Berne Switzerland Martinec J amp Rango A 1995 Seasonal runoff forecasts for hydropower based on remote sensing Proc Western Snow Conf Reno Sparks Nevada USA 10 20 Nash L L Gleick J A 1991 Sensitivity of streamflow in the Colorado basin to climatic changes Hydrol 125 221 241 Rango A Martinec J 1988 Results from international intercomparisons of snowmelt runoff model performance Proc 45th Annual Eastern Snow Conference Lake Placid New York USA 121 128 Rango A amp Martinec J 1994 Areal extent of seasonal snow cover in a changed climate Nordic Hydrol 25 Munksgaard Copenhagen Denmark 233 246 Rango A Martinec J 1995 Revisiting the degree day method for snowmelt computation Wat Resour Bull 31 4 657 669 Rango A amp Martinec J 1997 Water storage in mountain basins from sa
24. Massa Blatten Alps Kulang Himalayas Tavanasa Alps Dinwoody Wind River Cordevole Alps Salt Creek Utah Beas Manali Himalayas El Yeso Laerdalselven Lo Bru Viveli Hardangervidda Scofield Dam Price River Utah Cardos en Tirvia Pyrenees Okutadami Mikuni Joes Valley Dam Cottonwood Creek Utah Garona en Bossost Pyrenees Noguera Pallaresa en Escal Pyrenees Bull Lake Creek Rocky Mts Size km 84 85 98 106 108 116 121 125 130 165 183 196 205 215 228 248 248 345 350 375 386 401 417 422 435 449 450 484 15 Elevation Range m a s l 1440 2940 1140 3360 1890 3771 350 1265 1840 3210 879 3062 1790 3510 1032 2062 1820 3580 1850 3771 1500 3146 1447 4191 2350 5000 1277 3210 1981 4202 980 3250 1564 3620 1900 6000 2475 6550 530 1720 880 1613 2323 3109 1720 3240 782 2346 2131 3353 1620 3080 1860 2960 1790 4185 0 68 0 83 0 88 0 88 0 79 0 44 0 86 0 7 0 90 0 82 N A 0 91 N A 0 82 0 85 0 89 N A 0 68 0 91 0 86 0 73 0 8 0 8 0 83 0 83 0 75 0 87 0 82 Dy 8 1 1 5 2 4 N A 1 9 12 7 8 7 6 6 1 7 5 4 N A N A Sel 2 8 4 6 2 6 12 2 6 I2 5 0 2 6 5 4 3 0 See 4 8 Years seasons Zones 4 8 3 4 4 Ref 21 21 46 60 17 65 66 59 59 20 44 26
25. S on the x axis and y axis respectively The data are saved internally by SRM in ascending order by date SRM uses two different methodologies for creating MDCexcL wa e Methodology for winter deficits Beginning on the date following that identified by Step 5b above 1 e the cutoff date all remaining daily values for MDCexc S and a T newmelt are shifted backwards in time to the beginning of the melt season Additionally as the curve s x values are shifted each daily value is reduced by a constant equal to the Ist x value following the cutoff date reestablishing an x origin of 0 0 For example if a cutoff date of 15 April is identified 15 days into the melt season then all succeeding days for the data constituting the curve are shifted to the corresponding day 15 days earlier with each daily 2 a T newmelt being reduced by X a T newmelt day 16 173 e Methodology for winter surpluses increased P overwhelms temperature increase for the zone stretch the MDC Compute the proportional increase in winter ending snow water equivalent using values obtained in Steps 1 2 above 2 Pam 2Zonal_ Melt aiw 2P 2 Zonal Melt Multiply each x value of MDCexcy by this factor to stretch MDCexcz creating MDCexct wa 5d Derive a new curve MDCcrm wa by adding to MDCexcy wa the corresponding daily melt depths of new snow surviving in the warmer climate 2 newmelt c im from Step 4 Se For each daily value n
26. WinSRM Main Window Run Menu Pull down Item Choice Explanation Run a SRM simulation using the current simulation definition A melt season simulation activates several graphic plots that are not meaningful for year round simulations Simulation Melt Season Runs an SRM simulation for an entire year using the current simulation definition Plots not applicable for a year round simulation are temporarily disabled Simulation Year Round Run a SRM simulation in forecast mode using the current simulation definition A separate dialog is presented that allows the user to specify forecast update frequency see Appendix F Message 5 Plot choices are identical to a melt season simulation Forecast Melt Season Runs an SRM forecast for an entire year using the current simulation definition Plot choices are identical to a year round simulation Forecast Year Round l Open the Run a Climate Change Window which manages a Climate Change i subsequent climate change model simulation Tools The Tools menu pull down organizes the additional support processing options available to the model user When the Calculate option is selected the Long Term Monthly Averages window Window 22 is displayed allowing the user to calculate and display the long term monthly averages for temperature and precipitation for the current basin The Convert from choice is currently in development
27. file or a text txt file Beginning with version 1 00 10 a third output format flx is supported This option saves the flow data in a form compatible with the SLURP model Flow values saved in the Slurp format are always metric cubic meters per second The model automatically converts English format flow values to metric arenero OOOO Shows hides Quickhelp short descriptions of the window elements Redraw at revised scale only displays when scale factor changes see below Done Close the window and return to the Main window SAGA e ARAORRA NET Plots are drawn using default scaling derived from the data been depicted To alter the scale for instance to enhance comparison of similar plots there are two text boxes located on the button bar one for maximum x value one for a maximum y value When a value in either box is changed the Redraw button is added to the button bar see example below When it is clicked the current plot is redrawn using the modified scaling factors 110 Button bar default scale Button bar modified scale Each plot includes a legend box that displays an identified sample of each line shown on the plot Hydrograph plots also include an additional legend box displaying the statistics resulting from the simulation that generated the hydrograph Both legend boxes can be repositioned on the plot by dragging them to another unused position The mouse cursor changes from its nor
28. only in selected test basins with good data or in specially equipped experimental basins where a particular calibration model was developed but also in basins where such forecasts are required by the user SRM has relatively modest requirements for input variables temperature precipitation and snow covered area and therefore it was easily possible perform runoff simulations for the basins delivering water for hydroelectric projects as required by an electric company Examples of such simulations are shown in Figures 17 and 18 As mentioned in Section 3 SRM can be used for short term for example weekly forecasts of daily flows as well as for longer time period forecasts such as monthly runoff volumes or seasonal runoff volumes For short term forecasts temperature precipitation and snow covered area must be forecasted or predetermined for the coming days and substituted into the model Temperature and even precipitation forecasts are becoming increasingly available from meteorological services but the snow covered areas must be extrapolated by the model user The forecasts of input variables is still an important challenge for all snowmelt runoff models 7 1 Extrapolation of the snow coverage The future course of the depletion curves of the snow coverage can be evaluated from the so called modified depletion curves MDC These curves are automatically derived by SRM from the conventional curves CDC by replacing the time scale with c
29. report see section on reports 10 3 6 3 Buttons See Window 18 above Access the button functions using the keyboard by pressing the Alt key then the corresponding hot underlined key 10 3 7 Run a climate change window 10 3 7 1 Method of access After a current simulation has been identified by clicking on a row in the Basin Simulations data grid climate change processing is initiated by clicking the Main Window menu bar Run Climate Change Window 19 WINSRM Snowmelt Runolf Model for Windo Fie Options Data Run Tools Graphics Report E 8 z D Basin Definition simulation Forecast Name Rio Gro Window 19 Initiate Climate Change Processing 10 3 7 2 Purpose The Run a Climate Change window Window 20 guides the user through the complex process of simulating climate change in a mountain basin for an entire hydrologic year The functionality supported by this window can be subdivided into three areas selection control and reporting Click the check box prior to Step 1 to use the most recent climate change iw Run a Climate Change Identify the scenario here Scenaro Definitions for ROWY Tplus4 E AZ Use the climate changes defined in Results climate Change Processing Steps Measured Runoff Volume 10 6 ne 86 531 Average Measured Runoff m s 5 503 Computed Runoff Volume 10 6 nm 91 6677 Average Computed Runoff m Zelt 5 842 Stepi Winter Pr
30. see below The remainder of the record contains specific control information the model uses to control processing Data SE Name assigned to the simulation used to index the simulation s parameter table 1 50 characters SimTemp Distribution 0 basinwide lapsing kel L separately by zone SimP_ Distribution Integer 0 basinwide zone 1 P for all 1 P for each zone SimLR_ Distribution 0 basinwide zone 1 LR for all 1 LR for each zone SimCoeff Distribution Integer 0 basinwide zone 1 Cs Cr for all 1 Cs Cr for each Sum ct Distribution 0 basinwide use zone 1 Tc for all 1 Tc for each zone SimXY_ Distribution Integer 0 basinwide zone 1 X Y for all 1 X Y for each zone Sim nitialRunoff Runoff from day preceding day1 of the simulation f Is m s Precipitation value indicative of a significant event requiring special processing in cm SimComments 1 255 character description of the simulation SimName Text SimPrecipThreshold Single 1 required 142 PhysicalData Table This table occurs once per database It is the repository for the physical variables representing the basin s period of record POR The POR consists of all the unique time periods referenced by the simulations stored in the database Once entered records representing physical time periods are never removed even if the simulation that originally stipulated the time period is deleted Physi
31. where Quin the minimum discharge in the given basin 5 3 6 1 Adjustment of the recession coefficient for heavy rainfalls The formula 7 for computing the recession coefficient reflects the usual conditions characterizing the snowmelt runoff in the given basin When a heavy rainfall occurs the input is concentrated in a short time interval creating an abrupt rise and subsequent decline of the hydrograph In order to simulate such events the computer program automatically adjusts the recession coefficient whenever the daily rainfall averaged over the whole basin equals or exceeds 6 cm If P rain 26cm gt k x 4Q ka 5X 4 Qu kn 3 X 4 Qu ka Ia kis TKQ 15 after which it returns to the normal formula 7 In this way k gets lower so that the basin response to input becomes faster If the precipitation is recognized by Terry as snow and not rain the mechanism will not be activated If there is partly rainfall and partly snowfall in the respective elevation zones according 38 to Term the rainfall value is determined as a total of rainfall volumes from the zones with rainfall divided by the entire basin area gt P rain A rain P rain P 16 Note that in the following examples the type and amount of precipitation varies with zone from example to example Example a Precipitation input option 0 P 10cm Rainfall contributing area option 1 Zone A 100 km P rain 10cm Zone B 100
32. 25 This water equivalent can also be computed by accumulating the daily zonal melt depths right side of Figure 25 In this hypothetical example there is an agreement of the water equivalent determined either way In natural conditions discrepancies are to be expected mainly due to difficulties in evaluating the areal precipitation value for mountainous regions In such cases it is recommended that the value from the accumulated zonal melt be considered as more reliable The accumulated zonal melt value might even be used to correct the winter precipitation data and to estimate the altitude precipitation gradient Another advantage of this method is that it takes into account a possible redeposition of snow by wind during the accumulation season PRESENT CLIMATE E O T LLI 1 et 0 y 5 4 E Y A 10 j 5 lt SEASONAL SNOW COVER 15 ki A Al A I E TW WV V VE MI MI IX Fig 25 Illustration of the snow accumulation in the winter and snowmelt in the summer in the present climate hypothetical example On the other hand no losses are normally considered losses indicated by the runoff coefficient are assumed to take place after meltwater has left the snow cover This may lead to underestimation of the retrospectively computed water equivalents if significant evaporation from the snow surface takes place However if degree day ratios are used which have been derived from lysimeter measurements under similar eva
33. 3 8 Climate change processing Control screen 157 G 3 8 1 Climate change control ecreen 157 G 3 8 2 Climate change progress screen 158 GA Keyboard 0 un e Le EE 161 G47 del denotar as 161 6 4 2 Cursor MOVEMENT EEN 162 GAS Held Camden S S EE E EN 162 GAA FUNTION HE 163 EE Ee eg leien EE 166 GS MICO SRM OUTDUL BEGGEN 167 G 5 1 Simulation forecast statistics ocoococncccccicnccnncnncnononnncnononnonnncnnrornnrnnnnnnonnonnnns 167 Go SUMM aliada 167 Gora Re RTE 167 GG PIOCOISDIAY S EEN 167 G 5 4 1 Plot displays climate change 168 E e POMEO TEPON oreroraa enee dadas 168 G 5 6 Printed reports climate change 169 GO USING MICOS EE 170 G 7 Using Micro SRM to simulate a year round climate change 171 G 8 Using Micro SRM trace file OPtions cccccecececeeeeeeseseeeauaueueueetetereeaeaeaeaeeeauaueueererererstaeas 174 G9 MIGO SRM avala DIY ra 175 List of Figures Figure 1 Selected locations where SRM has been tested ANEN 13 Figure 2 Elevation zones and areas of the South Fork of the Rio Grande basin Colorado USA 21 Figure 3 Determination of zonal mean hypsometric elevations h using an area elevation curve for the SOUEM FOLK Ol th amp Ro Grande DAS ni il 22 Figure 4 Sequence of snow cover maps from Landsat 5 MSS Upper Rhine River at Felsberg 3250 km 560 3614 Ma 1 Baumgartner L987 iaa os 21 Figure 5 Example of a possible distortion of a depletion curve due to a tem
34. 36 Normalized runoff 9979 derived from the runoff of 1979 by normalized temperatures and precipitation Rio Grande basin near Del Norte Colorado coococococccconococoncnnororonnnroronnnnororonnos 73 Figure 37 Effect of a changed climate T 42C on the normalized snow covered areas in elevation zones A B and C of the Rio Grande basin near Del Norte Colorado o ooccocccccccccccncnnnoncnanananos 74 Figure 38 Effect of a changed climate T 42C on the normalized runoff 9979 in the Rio Grande basin neariDel Norte Colorado ai cds 74 Figure 39 Conventional depletion curves of the snow coverage CDC s in the Rio Grande basin near Del Norte Colorado in the years 1977 and E RA 75 Figure 40 Depletion curves of the snow coverage CDC s in the Rio Grande basin near Del Norte Colorado measured in 1979 and derived for 1977 by the climate program ooccccccccccnnccccnnnnnos 75 Figure 41 Transformation of runoff in the Rio Grande basin near Del Norte Colorado from runoff in 1979 to runoff in 1977 using temperature and precipitation of 1977 as new climate for 1979 76 Figure 42 Measured and computed runoff in the lllecillewaet basin in the year 1984 oeenn 80 Figure 43 Measured and climate affected T 42C depletion curves of the snow coverage in all elevation zones including the glacier zones C and D of the Illecillewaet basin oooooooomoooo 80 Figure 44 Computed runoff in the lllecillewaet basin in the year
35. 4 6 refer to Figure 15 Table 4 Results of model performances in the WMO project 10 years snowmelt season ent LEE CPE E onam ms ome ow sso ml zm iras mn ox osm m2 om zs y mus ra ozo aase ma oso 7 43 45 E AA Fig 14 Combined representation of model performance using three criteria R DG and D The volumes of the prisms indicate the average inaccuracies of the tested models from all results for snowmelt seasons reported in the WMO project Rango amp Martinec 1988 Fig 15 Combined representation of model performance using three criteria R DG and D The volumes of the prisms indicate the maximum inaccuracies of the tested models from all results as listed for snowmelt seasons and individual years in the WMO tables Martinec amp Rango 1989 46 6 1 2 Model accuracy outside the snowmelt season SRM has been designed to compute runoff during the snowmelt season but it can be run for the whole year if required According to the mentioned WMO intercomparison test WMO 1986 about the same accuracy as for the snowmelt season can be achieved for the entire year in mountain basins with a low winter runoff An example in Figure 16 shows that SRM can even be run without updating for 10 years Measured ssassn COM puted Discharge m s Discharge m s Discharge m s Fig 16 Runoff simulations in the basin Dischma 43 3 km 1668 3146 m a s 1 Compu
36. 92 WINSRM Snowmelt Runoff Model for Windows Fil Options Data Run Tools Graphics Reports Help Basin Definition Calculate Long term Monthly Averages i Define a custom time lag distribution For the basin Hame kio Grande a Window 7 WinSRM Main Window Tools Menu Pull down Choice Explanation Displays the Long term Monthly Averages window Window 23 which calculates and displays LTMA for several types of physical data for the current basin database Tools Displays the Define a Custom Lag Distribution window in E which a non standard lag time distribution may be defined A custom lag time supersedes all simulation lag time definitions for the life of the model or until the window is reopened and unpopulated Window 25 Calculate Long term Monthly Averages Define a custom lag time distribution Graphics The Graphics menu choice transfers control to the Graphics window Window 27 where all the available graphical outputs are selected and produced Because there is only one option for this menu choice there is no drop down menu WINSRM Snowmelt Runoff Model for Windows Fie Options Data Run Tools Graphics Reports Help E Bazin Definition Window 8 WinSRM Main Window Graphics Menu Explanation Displays the WinSRM Graphics window the window that Graphics controls the selection and display of the available graphical outputs Reports The Reports menu choice tr
37. Alps Rh ne Sion Alps Rio Grande Colorado Kings River California Maipo en el Manzano Andes Beas Thalot Himalayas Upper Yakima Cascades Chatkal Sturgeon Ontario Grande Tierra del Fuego Iskut Coast Karasu Upper Euphrates Karadarya Zerafshan Size km 1700 1800 1943 2000 2150 2170 2344 2448 2544 2600 2694 2732 2900 2929 3249 3371 3414 4000 4960 5144 5517 6591 7000 9050 9350 10216 12056 12214 17 Elevation Range m a s l 1235 4005 2500 6000 1030 4049 N A 983 3124 786 4105 1965 3234 1000 4200 1050 3411 570 3666 1553 3125 1596 4133 900 6100 1923 3260 562 3425 491 4634 2432 4215 171 4341 850 5600 1100 6400 366 2121 1000 4000 N A N A 200 2556 1125 3487 1100 4568 410 5500 0 77 0 63 0 82 N A 0 84 0 85 0 89 0 97 0 84 N A 0 91 N A 0 91 0 75 0 70 0 97 0 84 0 82 0 77 0 80 0 92 0 81 N A N A N A 0 95 0 87 N A Dy 8 0 6 4 4 3 N A 39 2 6 1 5 1 0 0 5 N A 1 5 N A 0 9 5 1 T2 3 8 3 2 0 9 1 5 2 8 1 6 N A N A N A 0 25 4 0 N A Years seasons 13 N A N A N A Zones 3 7 N A N A N A Ref 59 41 34 55 79 56 4 2 56 27 15 68 16 65 61 50 52 50 36 15 32 9 4 10
38. Cold Regions Hydrology Symposium Mitchell K M amp DeWalle D R 1998 Application of the Snowmelt Runoff Model using Multiple Parameter Landscape Zones on the Towanda Creek Basin Pennsylvania AWRA Water Resources Bulletin Vol 34 2 pp 335 346 46 Nagler T amp Rott H 1997 The application of ERS 1 SAR for snowmelt runoff modelling In M F Baumgartner G A Schultz and A Johnson editors 5th Scientific Assembly of the International Association of Hydrological Sciences Rabat Morocco pp 119 126 IAHS April 1997 47 NN 1982 Abflusssimulation Tiefencastel Semesterarbeit ET HZ Zurich Switzerland Institut for 48 Hydromechanics ET HZ 1982 NN 1979 Application of a snowmelt runoff model for improved water power generation Technical Report Institute of Hydromechanics and Water management Swiss Federal Institute of Technology ET HZ Z rich Switzerland 1979 125 49 Pfirter N 1980 Abflusssimulation Rh ne Gletsch Master s Thesis Swiss Federal Institute of Technology ET HZ 1980 50 Rango A amp Martinec J 2000 Hydrological effects of a changed climate in humid and arid mountain regions World Resources Review Vol 12 3 pp 493 508 51 Rango A amp Martinec J 1999 Modeling snow cover and runoff response to global warming for varying hydrological years World Resources Review Vol 11 1 pp 76 91 52 Rango A amp Martinec J 1997 Wa
39. Data selection data entry Click on the up down arrows at the bottom of the window to switch between elevation zones To view and edit days columns that are hidden use the grid scroll bars to reposition the grid s contents to the desired location different day and or column The keyboard s cursor arrow keys can also be used to reposition the grid display 98 To change a cell s value using the keyboard first identify the cell by moving the focus rectangle to the desired cell using the mouse or keyboard cursors To typeover replace the existing value simply begin typing the new value into the cell Press the Backspace key to edit modify the existing cell value The change is completed when focus 1s move to another cell The width of each grid column imposes a limit on the number of characters comprising the data entry value when that data value is manually entered using the keyboard This is overcome by using the mouse to resize the column To resize any grid column simply point at a column title line boundary until the mouse pointer changes to double lines Then press and hold the left mouse button while dragging the column boundary to its new location Data can be copied to or from the data entry grid on this window using one of several common editing techniques for grid spreadsheet data as follows e Identify a complete column to edit by clicking the column header The object of the subsequent edit in thi
40. FILE MENU 89 WINDOW 4 WINSRM MAIN WINDOW OPTIONS MENU 89 WINDOW 5 WINSRM MAIN WINDOW DATA MENU oi adds 90 WINDOW 60 WINSRM MAIN WINDOW RUN A ee eebe 91 WINDOW 7 WINSRM MAIN WINDOW TOOLS MENU 92 WINDOWS WINSRM MAIN WINDOW GRAPHICS EE 92 WINDOW 9 WINSRM MAIN WINDOW IREPORTS ee eebe A e 92 WINDOW 10 WINSRM MAIN WINDOW HELP MENU 93 WINDOW 11 EDIT SIMULATION CONTROL INFORMATION DIALOG cccccccescccecccesccccscecescsceecsseeceesecsesecesenseeeness 95 WINDOW 12 NAMING DIALOG FOR A NEW SIMULATION 95 WINDOW 13 DISPLAY THE EDIT BASIN VARIABLES DALOO 96 WINDOW 14 EDIT BASIN VARIABLES DIALOG breede A A 97 WINDOW 15 DISPLAY THE CLIMATE CHANGE SCENARIO DEFINITION WINDOW ccccsscccscccssccesccesccesccusccessceuesens 99 WINDOW 16 CLIMATE CHANGE SCENARIO DEFINITION WINDOW cooccnnoccnonccnnnccnnnoconnncononccnnnocnnnoronnnrcnnnccnnnccnanaccnnaros 99 WINDOW 1 2 RONA SIMULA TION a a si n 101 WINDOW 18 SIMULATION STATISTICS WINDOW cccccccccccsecccsscccescccescceescseesccceesseecsseecsseesseesseeecssenseseneseeeecenes 101 WINDOW 19 INITIATE CLIMATE CHANGE PROCESSING ccccccsscccsssccesccecscsccsccceseceeuecsseeceseusssesceeuecsseusseensseeecenes 102 WINDOW 20 RUN AC LIMATE CHANGE A laste dasiee eda bontcnets 102 WINDOW 21 WINTER CHANGE FOR ZONE DIALOG occonoccnnoccnnnccnnnoccnnnrononocnnnncnonarononccnnnrcnnnorcnnaronnnrononoccnnaronanicnns 104 WINDOW 22 ACCESS T
41. Forecasting in Toutunhe Basin in China Technical Report Dept of Water Resources Institute of Water Conservancy and Hydroelectric Power Research Al Fuxinglu P O Box 366 Beijing 100038 P R China Presentation at the 2nd SRM Workshop at the University of Berne Switzerland 78 Winterton S M 1999 Flow Forecasting Using Snowmelt Runoff Model Scofield and J oes Valley Reservoirs Technical Report College of Engineering and Technology Bringham Young University Provo UTAH USA 1999 79 WMO 1986 Intercomparison of models of snowmelt runoff Operational Hydrology Report 23 World Meteorological Organization WMO Geneva Switzerland 1986 WMO No 646 80 Zeng Qunzhu 1994 Runoff China Personal Communication 1994 128 Appendix A Examples Example 1 Create a new basin database by importing a DOS SRM data file 1 Start new session of WinSRM or click File New in an existing session 2 Click File Import In the Import a DOS SRM data file dialog window click on SAMPLE1 SRM then click Open 3 Provide a simulation name that will uniquely identify this simulation in the dialog depicted below Click Ok Name the Imported Simulation RE xj Enter a unique name to associate with this data Cancel 4 The Edit Simulation Control dialog appears Click Accept to accept the parameters exactly as defined in the DOS data set Be aware that if you attempt to edit any of the values in
42. J Wartena L Brugman A J M Zeilmaker D A 1966 Water balance assessment 40 41 42 of Lago Mar in the Pyrenees AHS Symposium on Hydrology of Lakes and Reservoirs Garda Italia Vol 70 71 pp 59 76 Martinec J 1963 Forecasting streamflow from snow storage in an experimental watershed Surface Waters AHS Publ No 63 pp 127 134 Maza J Roby H O Fernandez P C Fornero L Tarantola D 1990 Modelling Snowmelt and Runoff with Remote Sensing and GIS in Andean Watersheds Remote Sensing and Water Resources pp 191 200 The International Association of Hydrogeologists AH The Netherlands Society for Remote Sensing Mellander M K amp Eschner A R 1990 Snowmelt Runoff Modeling using GIS Parameter Estimation in a Western Adirondack Watershed 47 Annual Eastern Snow Conference Bangor Maine USA No 90 in 44 CRREL 43 Menajovsky S A 1985 Importancia del uso de reas nevadas como datos para los pron sticos 44 45 de derrames de los r os en las cuencas del Limay y Neugnen Argentina Importance of the use of snow covered areas for discharge runoff forecasting in the basins Limay and Neugnen Argentina Technical report NN 1985 Report presented at the XII National Congress on Water Mendoza Argentina Miller W 1986 Applying a Snowmelt Runoff Model which Utilizes Landsat Data in Utah s Wasatch Mountains AWRA Water Resources Bulletin pp 541 546
43. JUN JUL AUG Fig 35 Normalized depletion curves of the snow coverage 9979 in the Rio Grande basin near Del Norte Colorado dashed lines derived from the measured curves of 1979 solid lines zones A B and C m3 s WINTER Computed 1979 ses Hormalized 9979 Comp Clim 74 66 10 m Fig 36 Normalized runoff 9979 derived from the runoff of 1979 by normalized temperatures and precipitation Rio Grande basin near Del Norte Colorado 74 Snow cover Fig 37 Effect of a changed climate T 4 C on the normalized snow covered areas in elevations zones A B and C in the Rio Grande basin near Del Norte Colorado mtrs 8 EE e WINTER SUMMER 100 CL WMA SARRAN ANNE y AS ASS MA Alls UI IW E MA MA YIN IX Fig 38 Effect ofa Bei climate T 4 C on the normalized runoff 9979 in the Rio Grande basin near Del Norte Colorado 75 Dh SNOW cover 1 0 8 0 6 0 4 0 2 April May June July Aug Fig 39 Conventional depletion curves of the snow coverage CDC s in the Rio Grande basin near Del Norte in the years 1977 and 1979 Dh snow cover 1 0 8 0 6 0 4 0 2 April May June July Aug Fig 40 Depletion curves of the snow coverage CDC s in the Rio Grande basin near Del Norte Colorado measured in 1979 and derived for 1977 by the climate program He 210 180 present climate 130 la K Li LL Li daily discharge m s BU 30
44. Rocky Mts Ilanz Alps Cinca en Laspu a Pyrenees Toutunhe Otztaler Ache Alps Lago Alumin Andes Gongnisi Tien Shan Urumqi Tien Shan Angren Parbati Himalayas Illecillewaet Rocky Mts Segre en Seo d Urgel Pyrenees Buntar Himalayas Tinguiririca Bajo Briones Andes Hawea S Alps Ticino Bellinzona Alps Spanish Fork Utah Size km 529 545 550 558 559 373 600 618 630 700 705 730 776 798 840 893 911 939 950 1082 1154 1155 1217 1370 1460 1500 1515 1665 16 Elevation Range m a s l 837 3418 1740 3080 240 733 1420 3080 2506 3914 920 3380 2500 7000 261 702 1024 4450 577 2301 1400 5500 2521 4017 693 3614 1120 3380 1430 4450 670 3774 1145 2496 1776 4200 1880 4200 1200 3800 1500 6400 509 3150 360 2900 1200 5000 520 4500 300 2500 220 3402 1484 3277 0 55 0 92 0 78 0 72 0 89 0 91 N A 0 81 0 42 0 73 N A 0 87 0 53 0 78 0 81 0 84 N A 0 62 0 63 0 73 0 86 N A N A 0 88 N A 0 86 0 85 0 9 8 3 2l 1 8 N A 5 0 29 9 3 8 N A 1 1 8 6 5 6 2 0 9 18 N A 0 97 2 78 23 TS 7 0 N A N A Years seasons N A N A Zones 3 6 N A AK 5 N A N A Ref 6 21 45 23 69 70 22 41 42 15 19 33 69 6 21 77 1 43 35
45. SRM is completely contained on one distribution diskette It may be executed directly from that diskette The following steps describe installation on a hard disk and assume C and A are the hard disk and floppy disk drives respectively 1 Put Micro SRM diskette in your floppy drive 2 At the C gt prompt enter the following command lines ending each line by pressing Enter lt A Trademarks used in this document ep IBM PC IBM CORP MS DOS QuickBASIC Microsoft Corp QuickPak Professional Crescent Software are used solely for the purpose of providing specific information Mention of a trade name does not constitute a guarantee or warranty for the product by the U S Department of Agriculture or an endorsement by the Department over other products not mentioned 151 C gt cd Change to C s root directory C gt md srm create a model subdirectory C gt copy a c srm copy all files from floppy A to new subdirectory C SRM G 2 3 Configuring Micro SRM The model uses an external configuration file to define the variables that Micro SRM uses to control the interface between model and user The configuration file 1s specified on the command line invoking Micro SRM If no filename is specified the model uses the configuration file residing on the default directory named SRM CFG The file initializes program variables that define available disk devices printer control codes screen display color characteristics
46. The results can be compared with the measured runoff in order to assess the performance of the model and to verify the values of the model parameters Simulations can also serve to evaluate runoff patterns in ungauged basins using satellite monitoring of snow covered areas and extrapolation of temperatures and precipitation from nearby stations 2 Short term and seasonal runoff forecasts The computer program WinSRM includes a derivation of modified depletion curves which relate the snow covered areas to the cumulative snowmelt depths as computed by SRM These curves enable the snow coverage to be extrapolated manually by the user several days ahead by temperature forecasts so that this input variable is available for discharge forecasts The modified depletion curves can also be used to evaluate the snow reserves for seasonal runoff forecasts The model performance may deteriorate if the forecasted air temperature and precipitation deviate from the observed values but the inaccuracies can be reduced by periodic updating 3 In recent years SRM was applied to the new task of evaluating the potential effect of climate change on the seasonal snow cover and runoff as explained in Chapter 8 The microcomputer program has been modified and supplemented accordingly 4 MODEL STRUCTURE Each day the water produced from snowmelt and from rainfall is computed superimposed on the calculated recession flow and transformed into daily discharge from the
47. a 0 5 cm C d S 80 and EM 30 cm indicate that the curve for Hw 60 cm is applicable The snow coverage will drop to 64 in 7 days Extrapolated conventional depletion curve indicates values for day to day discharge computations Example 2 As above but the cumulative snowmelt depth to date is only 10 cm Consequently the curve for Hw 20 cm is applicable and the snow coverage will drop to 33 in 7 days which leads to a different extrapolation of the conventional depletion curve and to a different weekly total of forecasted daily runoff volumes If the initial water equivalent is known for example from SNOTEL a system of data transmission using meteor paths for reflecting the signals and operated in the USA the appropriate modified depletion curve can be selected at the start of the snowmelt season Otherwise the average curve dashed line in Figure 19 is used until the correct curve can be identified by satellite data Figure 20 illustrates the effect of new snow prior to the date of forecast In this example S 1 May 74 percent and XM 35 cm seem to indicate that the maximum modified depletion curve Hw 60 cm Figure 19 should be used The subtraction of the melt depth of the new snow reveals however that the seasonal snow cover is only average corresponding to the dashed curve in Figure 19 By this curve if the forecasted snowmelt depth M 15 cm is added to XM 35 cm S drops from 60 to 33 These auxiliary va
48. again taken over from the original CDC In less frequent cases no effect of temperature increase in a high elevation zone increased precipitation Scum wa gt S The program then determines AS Scum wa S and extrapolates CDCcrim wa as CDC AS on the missing days In order to evaluate the effect of a temperature increase on the Rio Grande basin the following steps were taken 1 Runoff in the whole hydrological year is simulated Figure 27 in order to verify the preselected parameters and the estimated snow coverage in winter 2 Conventional depletion curves of the snow coverage CDC used as input variable in the summer are plotted Figure 28 3 Winter runoff is simulated separately for T and T 4 C in order to obtain the respective runoff volumes hydrographs are printed 4 The decrease of the snow water equivalent on 1 April due to the increased snowmelt in winter winter deficit or negative winter adjustment is computed as explained in Section 8 2 5 Summer runoff is simulated separately for T in order to obtain the runoff volume hydrographs are printed At this point the climate affected conventional depletion curves CDCceim which are needed as an input variable for computing the summer part since they do not exist in the accumulation period in the winter of the climate affected runoff are derived as follows 64 6 The modified depletion curve MDCner is derived from the CDC This curve
49. by selecting a report or reports clicking the Print to Report Viewer radio button then clicking the View button 10 3 13 2 Purpose The File Display window Window 30 is a tool provided by WinSRM to preview one or more printed tabular reports available for use at any given point in the modeling process Reports that include calculated values may be viewed only when a model simulation has been successfully executed BB File Display f ioj xj Friot Edit SRM Reports Help File Display SEH Snovwmelt Runoff Model Version EetaD0l Basin Fio Grande Basin Zone A 01 29 2002 Period 10 1 1978 to 9 30 1979 Run 10 File C Program Files Microsoft Visual StudioYEJEAWinSEM Setup Files RioGrand mib Simulation EG Lapse Bate Critical Temperature Bain Contributing rea Lag Time Cf100m C Hours Oct Nov Dec Jan Feb Mar Lr Tc ECA LT Lr Tc ECA LT Lr Tc ECA LT Lr Tc ECA LT Lr Tc ECA LT Lr Tc ECA LT 1 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 2 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 3 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 4 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 5 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 6 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1
50. conditions and the k values may occur from year to year Figure 11 illustrates the range of k values for varying basin size and for the mentioned 37 different years Martinec 1970 For a larger alpine basin the limits refer to the envelope line and to the medium line in Figure 10 For the largest basin the higher limit 1 of the indicated range was derived by substituting the constants x 0 085 y 0 086 derived for the Dischma basin into Equation 12 The lower limit 2 was obtained by replacing JA y Ay in Equation 12 with 5 Am An Figure 11 shows that k can theoretically exceed 1 for very small discharges in large basins This does not really happen because such small discharges do not occur there Such a situation could be produced however by the user inadvertently taking over the x and y values derived in a large basin and using them for a small basin without modification In this case 1f the daily snowmelt input exceeds the previous day s runoff SRM computes a runoff decrease instead of increase In order to avoid this error the computer program prevents k from exceeding 0 99 However it is advisable to avoid approaching this limiting situation by checking the x and y values with regard to the lowest flow to be expected Recalling Equation 11 it follows for n because y gt 0 and 1 y lt 1 lt 1 1 1 Qo L y 13 x Therefore the values x and y should fulfill the condition 1 Quin gt X 14
51. data field s three associated data validation attributes maximum length context and content range Maximum Length each data entry field displayed on the screen has a predefined physical dimension that is displayed via the field s shadow Any attempt by the user to exceed a field s shadow is ignored and a warning message is displayed Field Context each field has a predefined context that dictates the type of characters which it may receive Ifthe context rule for the field is violated the illegal character is ignored and a warning message is displayed There are 3 basic contexts Alphanumeric the default context Any character is valid Binary a value of 0 or 1 Numeric valid numeric character 1 e 0 9 Field Content Content validation differs from the first two validation forms in that 1t occurs after all the characters for a given field have been entered The information may then be validated as a distinct entity It is at this point that field values are checked for valid range consistency with previously entered data etc Violation of a field s range or consistency checks causes a error message to be generated and locks the cursor on the invalid field until the condition is corrected These techniques help insure that data entry errors are minimized and are caught before the fact that is before the data are introduced to the actual model algorithms There are three data entry screens used within Micro SRM
52. depletion curve indicates the water volume stored in the snow cover if the y axis scale is in km If the y axis scale is in percent snow coverage it indicates the water equivalent of the snow cover as an areal average Therefore each curve in Figure 21 can be labeled by the water equivalent which it indicates The rectangular shaded area means 0 1 x 10 cm 1 cm Because the area below the highest curve is 83 times larger this curve indicates that at the beginning of computations of the cumulative snowmelt depth usually on April the snow accumulation corresponded to the water equivalent of 83 cm The values for each curve are automatically determined by the computer program The nomograph is used for real time forecasts as follows In a current year the snow covered area is monitored from the start of the snowmelt season and simultaneously the cumulative snowmelt depth is computed The snow covered area must be evaluated as quickly as possible after each satellite overflight The degree days necessary for melting the temporary snow cover from intermittent snowfalls are disregarded If after some time for example on 15 May the snow coverage amounts to 80 and the cumulative snowmelt depth amounts to 15 cm the modified depletion curve labeled by 37 cm is identified to be valid for that year This curve can be used for extrapolating the snow covered area For example if another 15 cm will be melted in the next week according to tempera
53. example If Example 3 was skipped the climate change scenario Tplus4 can be substituted for Ex3 Skip steps 1 and 2 1f you have proceeded to this example directly from completion of Example 3 l Start the WinSRM program Load the Rio Grande database that is included with the software by clicking File Open In the Open an SRM database dialog double click the entry RioGrand mdb Control will return to the main window populated with the new database information 2 Click on RGWY79 in the Basin Simulations grid making it the current simulation 3 Click Run Climate Change to begin the climate change simulation process The Run a Climate Change window appears 4 Select Ex3 from the Scenario Definition s for RGWY79 drop down list This list contains all the scenarios in the database that have been created to use with simulation RGWY79 5 Click to begin the climate change simulation As with any normal simulation the model first initializes the SimResults table with a copy of variables and parameters taken from the database that are required to drive the simulation Next using the rules contained in 132 the Ex3 scenario the model creates a set of changed climate data modifying variables and or parameters as directed by the rules contained in the climate change scenario Ex3 Finally the model runs a simulation for winter present climate 6 After the step completes results of the simulation are displayed in the Statistics gr
54. for updating simulated unott with observed values Cancel o Message 5 Physical data already present in database see Page 86 150 Appendix G SRM Computer Program Version 4 G 1 Background The current version of the model Micro SRM Version 4 0 was developed using Microsoft QuickBASIC 4 5 and contains several subroutines from QuickPak Professional a BASIC toolbox developed by Crescent Software Inc G 2 Getting started G 2 1 System requirements Micro SRM requires the following personal computer resources e Any 100 IBM PC compatible machine e 1 floppy disk drive The model is distributed on a 3 floppy disk By special request 5 media are available The latest version release and documentation are also available via the Internet at http www ars usda gov Services docs htm docid 8872 e A minimum of 600K available random access memory RAM e A PC DOS operating system Version 3 0 or later SRM will run in a MS DOS window under Microsoft Windows 3 1 95 e If graphics are desired a graphics adapter is required Micro SRM supports the following graphics modes CGA Color Graphics EGA Enhanced Graphics HGA Hercules Graphics VGA Virtual Graphics e For printed output a dot matrix or other printer that supports a 132 character print line either physically or using compressed mode print characters e A Microsoft compatible mouse optional G 2 2 Installing Micro SRM Micro
55. initiate logical functions which are available for use at any point in the modeling process Where defined the corresponding mouse click equivalent is presented in parentheses at the end of the definition ESCape The use of the ESCape key in Micro SRM is consistent with most typical PC applications Basically the ESCape key suspends the current activity inspecting help screens performing variable parameter data entry loading a data file etc and returns control to the point in the program where the suspended activity was originally invoked It backs the user up one logical processing step From the Program Option Basin Definition or Variable Parameter Menu screens the ESCape will terminate Micro SRM and return control to DOS LeftClick on Esc or RightClick to simulate ESCape PgUp PgDn The PgUp and PgDn keyboard keys control which screen display is presented to the user of Micro SRM They are functionally equivalent to turning forward or back one page in a multi page form The exception to this usage rule occurs while entering or editing data values on the variable parameter data entry screen display During this type of data entry the PgUp and PgDn keys are used to superimpose the next two or prior two month s data values for the specific variable parameter onto the data entry screen enabling the user to logically step forward or backward in time through the snowmelt period LeftClick on lt 4 L gt PgUp PgDn 162
56. is on the cursor movement key equivalents on the number pad are deactivated Micro SRM assumes you are entering numeric data not trying to relocate the cursor or change the display page G 4 4 Function keys Micro SRM uses the microcomputer s dedicated keyboard function keys F1 F10 to provide the user with the ability to interrupt the current model activity typically data entry with a single keypress mouse click and transfer control to a specific predefined function process On function termination control returns to the point of interruption Each Micro SRM screen display includes information on the function keys that are hot 1 e active at that point in the program The functions described below are Micro SRM primary function key definitions These functions are hot while the Program Options Basin Definition and Basin Variable Parameter Menu screens are displayed FI Help The Help function is invoked by pressing clicking function key F1 at any point in the modeling process When help is requested from a primary screen display Program Option Basin Definition Variable Parameter Menu Variable Parameter Data Entry Micro SRM replaces that screen display with the Help Screen most logically related to the user s position within the model While help is active the user may PgUp or PgDn to view all of Micro SRM s available help screens The function is terminated with an ESCape and returns the model to the logical state e
57. km P snow 10 cm Zone C 100 km P snow 10 cm 10 cm 100 km P rain aa 33cm gt k xQ Example b Precipitation input option 0 P 10 cm Rainfall contribution area option 1 Zone A 100km P rain 10cm Zone B 100 km P rain 10cm Zone C 100 km P snow 10cm 10 cm 200 km P rain E 6 7cm gt k x 4Q on 5 consecutive days Example c Precipitation input option 1 P 7 cm Pg 10 cm Po 13 cm Rainfall contributing area option 1 Zone A 100 km P rain 7cm Zone B 100 km P rain 10cm Zone C 100 km P snow 13 cm 7 em 100 km 10 cm 100 km P rain 300 km 57cm gt k x0Q 39 These examples refer to the rainfall contributing area option 1 that is to say rainfall contributes to runoff from snow free as well as snow covered parts of the basin For rainfall contributing area option 0 with rain retained in the snowpack the adjustment of the recession coefficient is less likely to be activated Example d Precipitation input option 0 P 10 cm Rainfall contributing area option 0 Zone A 100 km P rain 10cm S 0 4 Zone B 100 km P rain 10cm S 0 6 Zone C 100 km P snow 10 cm Sc 0 8 10cm 1 S 100km 10cm 1 S 100 km 3 33cem gt k xQ 300 km P rain but it can still take place if a very heavy rain falls over the whole basin Example e Example e Precipitation input option 1 P
58. may duplicate one or more days already present in the database When this occurs the user is asked to decide which set of duplicate value s to use Appendix F Message 4 10 4 2 Data base data entry At this point in the database creation process the data storage structures tables required by the model have been created and initialized However before any meaningful modeling can be performed real data and parameters must be provided replacing the default values assigned during the table creation steps just described Data entry is performed using the Edit Variable Parameters dialog Window 14 Though the values required may be entered using the computer keyboard the recommended method for moving data into the WinSRM database is through use of cut copy paste functionality found in WinSRM and in most Windows software The initial step in this process would be to import the digital data into a spreadsheet program Microsoft Excel Corel Quattro Pro etc Once the data are in such a form it becomes a simple task to copy columns of values to corresponding positions in the WinSRM edit dialog Be aware that the scope of any data entry editing varies depending upon whether the data are basin variables in the PhysicalData table or simulation parameters in a simulation table Basin variables belong to the basin not the individual simulation Changes to basin variables are global in scope They will affect ALL basin simulat
59. mn mn eB mm mm mm mm mm wm wm wm ell we ze rm vm vm vm Moo el ve D E Al 1 a oe ee es bie Myo ee eve a es a Pl Zong see rm vm mm mm mm mm mm mm ml e 1 NW 1 s Aug Jul of the snow coverage in all elevation C depletion curves affected T 4 Fig A3 Measured and climate zones including the glacier zones C and D of the Illecillewaet basin 81 350 4 i 300 T 4 P pane A Original Simulation 3 250 8 200 O ON 3 150 E A 100 50 0 Fig 44 Computed runoff in the Illecillewaet basin in the year 1984 and climate affected runoff increased by glacier melt Rearranging Equation 28 and recalling that Me 0 it follows M cum Rerm R Scum ST c ACOcum ACO c 12 30 cm 1 15 cm 0 47 cm 11 62 cm By adding up daily melt depths starting date indicated by CDCcy 1n Figure 43 a comparable amount is obtained Zone C E A 2 294 cm 0 7 1 606 cm E 8 61 cm 0 7 6 027 cm No A oE 3 66 cm 0 7 2 562 cm Total 10 195 cm Zone D 6531 leede 37 719 cm 0 7 26 403 cm August 52 84 cm 0 7 36 988 cm September 12 62cm 0 7 8 834 cm Total 72 225 cm Total basin 468 9 km 92 4 km _ __ AS 9 Eeer oo disso SEN Me cLim 10 195 cm 82 In terms of volume the net glacier contribution to runoff is 134 2 10 m acc
60. of the initial snow cover in a changed climate MDCcum wa winter adjusted curve It is derived from MDCexcy wa by taking into account snowfalls surviving in the new climate CDCcum wa conventional depletion curve of snow covered area in a changed climate The curve is derived from MDC cy iy wa CDC cum wa ma Curve adjusted for the model input model adjusted If the derivation from MDCc m wa results in more than one S value on a date the first highest value is used If there is no new S value on a date the previous day s S value is repeated until there is a new S value With this adjustment CDCcum wa ma can be used as input to SRM which requires one S value on each day as 1s provided by the original CDC This curve is used to compute the year round climate affected runoff 63 300 250 py JW 200 m s 150 100 OU A Al All ii UU IW V MI VE Wl IX Fig 27 Measured and simulated runoff in the Rio Grande basin near Del Norte Colorado in the hydrological year 1979 Due to the time shift by the changed climate the derivation of Scum wa values from MDCc um wa may stop before the end of the computation period In case of Scum wa lt S the program decreases the depletion curve CDCcim wa ma to the last S value of the original CDC which is typically zero and repeats it for the missing days If an elevation zone contains glaciers or permanent snow cover CDCcrm wa ma drops to the residual snow or ice coverage
61. per 100 m where y h altitude of the temperature station m h Hhypsometric mean elevation of a zone m Whenever the degree day numbers T AT in Equation 1 become negative they are automatically set to zero so that no negative snowmelt is computed The values of the temperature lapse rate are dealt with in Section 5 3 3 Temperature input Program options 0 basin wide 1 by zone 24 The program accepts either temperature data from a single station option 0 basin wide or from several stations option 1 by zone With option 0 the altitude of the station is entered and temperature data are extrapolated to the hypsometric mean elevations of all zones using the lapse rate If more stations are available the user can prepare a single synthetic station and still use option 0 or alternatively use option 1 With option 1 the user may use separate stations for each elevation zone however the temperatures entered for each zone must have already been lapsed to the mean hypsometric elevation of the zone Although SRM will take separate stations for each zone in this way it is only optional The measurement of correct air temperatures is difficult and therefore one good temperature station even if located outside the basin may be preferable to several less reliable stations In the forecast mode of the model it is necessary to obtain representative temperature forecasts for the given region and altitude in order to extrapo
62. physical data is queried from the database and combined with the current simulation s parameter data table to form a temporary table named Results There is one Results table per WinSRM database Each simulation overwrites the existing table After the Results table 1s created companion columns to the variables and parameters that affect climate change are initialized using the associated present climate values and then modified under control of the rules found in the selected climate change scenario These companion columns identified with the CC postfix in the column heading represent the variables and parameters that define the proposed changed climate The exception noted above occurs when the Use the Climate Changes Defined in Results check box 1s checked When checked the process of building the Results table is bypassed The Results table built 104 during the most recent simulation is re used See Appendix A Example 1 for a discussion of re using a Results table After the Results table has been established a simulation for the winter half year is run to determine total zonal melt and total precipitation occurring during winter under present climate conditions Step 2 Winter Changed Climate A simulation for the winter half year is run for the changed climate Changes in total zonal melt and total precipitation are calculated Step 3 Summer Present Climate Run a summer simulation for the present cli
63. popular Version 4 is also preserved in the Appendix because it is still in demand to be used within its limits The Windows version adds new capabilities it accepts more detailed climate scenarios for example different daily changes of temperature and precipitation It makes possible to substitute a data set of temperatures and precipitation of a selected year as a climate scenario for any available existing year and evaluate the resulting snow conditions and runoff A normalized year including normalized Conventional Depletion Curves CDC s from long term temperature and precipitation data can be derived to represent today s climate It is now possible to divide a basin into as many as 16 elevation or other zones in order to refine the modeling while Version 4 only allowed 8 These improvements facilitate new developments in SRM applications which are already taking place runoff modeling by using different land use zones separating satellite mapping of snow and glaciers runoff modeling in very large basins with an extreme elevation range and others The specific features of WinSRM Version 1 11 are explained in detail in this document in Sections 8 5 8 6 9 and 10 WinSRM Version 1 11 has been developed without sacrificing the advantages of the SRM Version 4 in particular the speed of getting results Both versions are available on the Internet by accessing http www ars usda gov Services docs htm docid 8872 Should this link not be
64. results generated for the climate change run from Step 5 a Click the header of the column of the average temperatures that were modified normalized by the above scenario The column title is TavgCC you may need to resize the column to see the complete column name The column should be highlighted after the column title is clicked see below w Simulation results l Ioj x Edit 1 1 B asin 0701779 0702 78 0703 78 0704 78 0 05 78 0706 78 07 07 78 0 08 78 0709 78 01078 0111478 0712778 0713 78 0 14 78 0715779 0716778 Rio Grande Basin Simulation RGW Y739 ActualQ TMax TMaxC Tmin TminC TAva META Precip PrecipCC CDC CDE 5 320 3 720 0 000 5 320 9 240 53 040 4 870 4 790 4 870 4 700 4 700 4 700 4 870 4 790 4 790 3 410 14 720 16 930 19 510 A0 RAN 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 n nnn 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 D nnn 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 n Ann 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 D nnn 10 730 2 950 3 640 2 950 8 090 2 250 2 670 3 060 4 340 10 590 10 730 2 3390 2 670 3 360 3 090 A Ann 0 000 0 000 0 000 n n
65. runoff in an alpine basin Hydrological Processes Vol 12 pp 1659 1669 Seidel K Brusch W Steinmeier C Martinec J Wiedemeier J 1995 Real time runoff forecasts for two hydroelectric stations based on satellite snow cover monitoring Jan Askne editor Sensors and Environmental Applications of Remote Sensing 14th EARSel Symposium 1994 in Goteborg Sweden pp 253 258 A A Balkema Rotterdam Brookfield 67 Seidel K Burkart U Baumann R Martinec J Haefner H amp Itten K 1 1989 Satellite data for evaluation of snow reserves and runoff forecasts Hydrology and Water Resources Symposium Christchurch N Z pp 24 27 November 1989 68 Sereno D J 1987 Applying the Martinec Rango Snowmelt Runoff Model with Landsat Satellite Data Imaging Techniques An Alternative Method of Streamflow Prediction in Utah s Wasatch Mountains Master s Thesis Dept of Civil Engineering Brigham Young University Provo Utah USA 69 Shafer B A Jones E B amp Frick D M 1981 Snowmelt runoff simulations using the Martinec 70 71 72 Rango model on the South Fork Rio Grande and Conejos river in Colorado AgRISTARS Report CP G1 04072 Goddard Space Flight Center Greenbelt MD Shafer B A 1980 Report on the Martinec model project Technical Report USDA Soil Conservation Service Denver Colorado 1980 Smith L C 2002 Runoff Iscut Coast Canada Personal Communication
66. simulation as shown in Figure 24b Even without updating however the initial discrepancy is soon eliminated automatically Further possibilities of updating will be made available to users when more experience in real time situations is accumulated For example it should be possible to adjust some parameters e g the runoff coefficient in the progress of the forecast but only within hydrologically and physically acceptable limits In any case false forecasts of temperature and precipitation should be updated whenever a correction by new data is indicated What is generally called updating can be thus divided into 3 categories 1 Updating the computed discharge by the measured discharge when it becomes known Le checking with the measured discharge to avoid carry over of errors when the next forecast is issued 2 Adjustment of model parameters in the process of forecast 3 Correction of temperature precipitation and snow cover forecasts according to actual observations Short term discharge forecasts can be updated as frequently as each day Baumann et al 1990 57 8 YEAR ROUND RUNOFF SIMULATION FOR A CHANGED CLIMATE SRM uses a real snow cover from satellite monitoring in the present climate in order to produce a snow cover and runoff in a changed climate This requires a rather detailed procedure but uncertainties arising from a fictitious snow cover simulated from precipitation and arbitrary threshold temperatures
67. the cursor as it proceeds The repeat count can vary from 1 to 365 F10 Insert Delete Day This function lets the user modify an existing array of daily values without reentering the entire array When invoked the user may insert a blank day at the point marked by the cursor shifting existing day values one day forward in time or delete the day marked by the cursor shifting existing daily values one day back in time G 5 Micro SRM output products Several different forms of output have been included in Micro SRM designed to support a user s specific requirements for information provided to and generated by the model Model output is user selectable and for several of the forms interactively controlled The available output products are described in the following paragraphs G 5 1 Simulation forecast statistics Each time an F7 Compute is successfully executed the model suspends further processing while it displays several important statistics which quantify the accuracy of the calculations Any keypress then returns the model to its precompute state G 5 2 Summary display Function Key F2 invokes the model summary screen See previous sections for a complete description of this function It is designed to give the user a quick overview of the state of seven critical model parameters Summary is useful when making iterative runs each with slightly different parameter configurations The basin name and model run number are included
68. this dialog each zone in the simulation s parameter table will be initialized using the values shown 5 At this point the main window regains control The DOS file has been imported into a new database Before continuing you should save the new database using File SaveAs 6 Run the imported simulation click Run Simulation Melt Season A warning message dialog appears as shown below This occurs since there is no initial runoff value present in the physical data table for the date March 31 1975 The model will use the value contained in the DOS simulation 12 0 for that missing value Click to proceed with the simulation Inconsistent data a x The Initial unoffY alue For this simulation 4 12 is different than that Found in the basin database Ob Continue Cancel 7 Because no latitude longitude is available from the DOS SRM file another warning dialog appears as shown below Click to continue Missing latitude E 0 xj 4 positive or blank latitude indicates a basin located in the northern hemisphere a negative latitude indicates a basin located in the southern hemisphere Continue processing 129 8 Upon completion of the simulation the Results dialog appears as shown below After examining the statistics click Done to return to the main window w Simulation Statistics SE Ioj xj Basin Hame Durance Basin Simulation SAMPLE Run Date 1271772002 Measured Runo
69. to runoff e Displays basic statistics compiled from the computed results of the simulation see Window 18 118 11 REFERENCES 11 1 General references Bahr D B M F Meier S D Peckham 1997 The Physical Basis of Glacier Volume Area Scaling Journal of Geophysical Research Vol 102 No BY pp 20355 20362 Baumgartner M F 1987 Schneeschmelz Abflusssimulationen basierend auf Schneeflachenbes timmungen mit digitalen Landsat MSS and NOAA AVHRR Daten Snowmelt runoff simulations based on snow cover mapping using digital Landsat MSS and NOAA AVHRR data German version Remote Sensing Serie 11 Department of Geography Univ of Zurich Zurich Switzerland English summary Tech Report HL 16 USDA Agricultural Research Service Hydrol Laboratory Beltsville MD USA Baumgartner M F Rango A 1995 A microcomputer based alpine snow cover analysis system Photogramm Engng 61 12 1475 1486 Baumann R Burkart U amp Seidel K 1990 Runoff forecasts in an Alpine catchment of satellite snow cover monitoring Proc International Symp on Remote Sensing and Water Resources International Association of Hydrogeologists Enschede The Netherlands 181 190 Becker A amp Serban P 1990 Hydrological models for water resources system design and operation Operational Hydrol Report 34 WMO Geneva Switzerland Brusch W 1996 Das Snowmelt Runoff Model ETH SRM ETH als universelles Simulations und Prognosesyst
70. to that of the Enter lt key Using these keys the model user may move freely about the input data screen entering new and editing existing field values LeftClick on any field to move the cursor to that field Movement between input fields initiated by cursor movement key presses is governed by a forward and a backward progression sequence The forward progression via the Enter lt or Tab keys on a screen beginning at the top leftmost field is from left to right top to bottom Backward progression is the reverse Movement vertically is generally from the current field up down to the next closest field Movement via the Enter key during daily variable parameter data entry is from current day to next calendar day Tab The Tab key functions exactly the same as Enter lt 4 Shift Tab If the cursor is not on a field s first character it shifts to the first character If on the field s first character the cursor moves to a prior field as dictated by the backward progression Home End These keys move the cursor to the first last data field respectively on the screen display PJ Within a field value the screen cursor may be moved left or right from character to character If the cursor is at the leftmost character of the field a lt moves the cursor to the prior field as defined by the screen s backward progression If the cursor is at the rightmost character of the field value a gt moves the cursor to the next field in
71. total basin 468 9 km 92 4 km 1984 179cm 2 10 86 em 1 5960 cm 1155 0 km 1155 0 km 92 4 km T 4 C 2 03 _ _ 0 1624 2 1155 0 km SC Scum ST c 1 434 cm 0 8 1 147 cm The carry over of runoff is computed by the recession flow formula Equation 7 with x 1 052 and y 0 037 as derived in the Illecillewaet basin in the winter In the hydrological year 1984 the runoff of the previous day 30 September 1983 was 22 5 m s and it took 3 days of recession flow totally 0 453 cm runoff depth to reach the computed final runoff on 30 September 1984 which was 18 46 m s Thus ACO 0 453 cm a gain from the carry over balance For T 4 C the computed final runoff is 22 638 mis so that a loss from the carry over balance results AC Oct nm 017 cm ACOcum ACO 0 017 cm 0 453 cm 0 47 cm 80 aa Wm mm mm mm mm wm n e Dm w wm Computed Runoff Measured Runoff 350 300 ON 0 s u asIeyosiq Fig 42 Measured and computed runoff in the lecillewaet basin in the year 1984 I I I I I I I I I I I e I I I I I I I I I I I L I I I I I I I I I I I e I I I I I I I I I I L I I I I pl I I I I I I I e zen O mn mn em mm mm mm mm mm mm mell en ze y ee mm mm mm mm mm bk
72. units have already been entered Notice the units are set to metric and may not be changed Mixed units are not allowed in a WinSRM database As with Units the Number of Zones text box once populated with a value will also become fixed for the life of the database All the other basin definition variables may be changed as necessary These unique basin descriptors are stored in the BasinDescription table in the new database PhysicalData table The PhysicalData table contains all the physical data observations temperature precipitation snow covered areas required by WinSRM to drive its simulations Once the number of elevation zones in the basin is defined Window 31 the user is provided with an opportunity to populate initialize the PhysicalData table using a dialog window shown below Window 32 w Populate physical data table E e ol x Enter the starting and ending dates for the period for which data are available in the basin Starting date mm dd yyyy 01 01 1990 Ending date mm dd yyyy 12 31 1990 Continue without populating Accept dates Window 32 Populate physical data table prompt Using the period of record provided in this dialog WinSRM will populate the PhysicalData table with the necessary records one for each zone day If this step is bypassed the records will be created as required for each new simulation definition Values for the entire period of record for the b
73. unknown Actually this period for Pn usually lasts from 0800 hrs on the day n to fem bay 0800 hrs on the day n 1 and is published as precipitation on the day n In some cases however the same precipitation amount is ascribed to the day n 1 on which the measurement period ended In such case precipitation data must be shifted backwards by one day before input to SRM 5 2 3 Snow covered area S It is a typical feature of mountain basins that the areal extent of the seasonal snow cover gradually decreases during the snowmelt season Depletion curves of the snow coverage can be interpolated from periodical snow cover mapping so that the daily values can be read off as an important input variable to SRM The snow cover can be mapped by terrestrial observations in very small basins by aircraft photography especially in a flood emergency and most efficiently by satellites The minimum area which can be mapped with an adequate accuracy depends on the spatial resolution of the remote sensor Examples are listed in Table 3 Figure 4 shows the snow cover in the alpine basin Felsberg mapped from Landsat 5 MSS data Baumgartner 1987 When the time interval between the subsequent satellite images becomes too long e g due to visibility obscured by clouds the depletion curves derived from the measured points may be distorted by occasional summer snowfalls 26 Table 3 Some of the possibilities of remote sensing for snow cover mapping Pl
74. value applied to the specified data using the specified edit action e Zone the zonal scope of the change all zones or one specific zone A comments text box near the bottom of the window provides a location where the user can self document the scenario 10 3 5 4 Buttons PDelete the current scenario Close the window ele _Delete SC return to the main window Create a new scenario Print the current scenario Show Quickhelp for this window Display Help for this window Access the button functions via a left mouse click or using the keyboard by pressing the Alt key then the corresponding hot underlined key 101 10 3 6 Simulation statistics window 10 3 6 1 Method of access After a current simulation has been identified by clicking on a row in the Basin Simulations data grid a simulation is initiated by clicking the Main Window menu bar Run Simulation Melt Season Simulation Run Simulation Year Round Simulation Run Forecast Melt Season Forecast or Run Forecast Year Round Forecast as in Window 17 WINSRM Snowmelt Runoff Model for Windows Fie Options Data Run Tools Graphics Reports Help Simulation Melt Season Simulation Basin Definition Forecast b Year Round Simulation Hame rio Gre Climate Change Window 17 Run a simulation 10 3 6 2 Purpose The Simulation Statistics window Window 18 appears upon successful completion of the current model simulat
75. viewed given the current point of model operation Selecting a report from this menu replaces the window s text contents with the selected report 114 Help Displays a separate Help window that documents the File Display window and its use 10 3 13 4 Window control The model attempts to display the lines of each report as separate lines in the window If the physical width of the window is too narrow to adequately display a line it will be continued wrapped to the next line in the window making the report difficult to read To alleviate such situations the window may be resized by moving the mouse cursor over a window border until the cursor changes to a double headed arrow then dragging the window into a larger state Alternatively the window can be maximized by clicking the maximize icon on the window s title bar Portions of report text may be copied to the Windows clipboard by clicking some starting point in the window then while pressing the left mouse button dragging the mouse to an ending point The highlighted text 1s copied by right clicking then clicking Copy to Clipboard 10 4 WinSRM operation Prior to the running the snowmelt runoff model processing algorithms there are number of supporting tasks that must first be addressed The task list will vary depending upon the state of the currently loaded database and the processing algorithm simulation or climate change desired The following sections b
76. 0 0 000 011278 4 790 0 000 0412278 4 790 0 000 0713778 8 410 0 000 0714778 14 720 0 000 10715778 16 930 0 000 10716778 19 510 0 000 10717770 ON RAN N NNN 4 one of 3 WW A 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 n ana C Program Filess WwinSAM AioGrandtaQ hl mdb 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 dh hh dh hd S Print Edit multiple days Switch Scroll Style 000 0000 0 000 0 000 0 000 0 000 annn n nnn Runoff Tmax Tmin RER COC Precip NetRadiation 0 000 0 000 0 000 0 000 dh hh Ch Ctrl c Ctrl Gd Print Help Cancel Done 10 26 2005 14 28 PM E Window A2 Replacing basin temperature data with normalized data Click to save the changes made in this window to the database and return to main window 6 Repeat Step 6 for each zone of precipitation data by copying column PrecipCC from the Results table to column Precip in the Edit Basin Variables table for the current simulation 7 Repeat Step 6 for each zone of snow depletion data by copying column CDCclimWAMA from the Results table to column CDC in the Edit Basin Variables table for the current simulation 8 Repeat Step 6 for normalized daily runoff by copying column DailyQ from the Results table column 46 to column Runoff in the Edit Basin Variables table for the current
77. 1 Method of access After a current simulation has been identified by clicking on a row in the Basin Simulations data grid the climate change scenarios for the simulation may be displayed and edited using the Climate Change Scenario Definition window Display this window by clicking on the Main Window menu bar Data Climate Change Scenario Definition Window 15 WINSRM Snowmelt Runoff Model for Windows Fie Options Data Run Tools Graphics Reports Help Climate Change Scenario Definition Basin Deh Raman h Variables d Parameters Units of A Results Window 15 Display the Climate Change Scenario Definition Window 10 3 5 2 Purpose The Climate Change Scenario Definition Window Window 16 is a specialized editor used to create and manage the various climate change scenarios CCS for a given simulation A CCS is a set of user provided rules that WinSRM uses to create a modified version of the physical and or parameter data that will be used to approximate a changed climate for the given simulation period w Climate Change Scenario Definition for Simulation RGWY FF Simulation Period Start Date 9710 07 1976 End Date 0373071 g77 Exiting Scenarios for ROW TT 191 plus Mew Delete Climate Scenario 1977 Tplu4 variable parameter Starting Date Ending bate Edt Action edit Factor zoe Casier s0 01 1976 0973071977 shift Backward m am Es ono io 0 1976 maner apen si 0000 at Y
78. 14 0 20 1 50 1 14 7 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 amp 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 9 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 10 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 14 0 65 0 75 1 14 0 65 0 75 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 65 1 00 1 14 0 80 1 50 1 14 12 0 65 0 75 14 0 65 0 75 14 0 65 1 00 14 0 65 1 00 14 0 65 1 00 14 0 80 1 50 14 hd C Program Files Microsott Visual Studio B9855 AM rpt out 1 26 2002 1 34 PM a Window 30 File Display Window 10 3 13 3 Menu The File Display window menu provides tools to transfer the window s contents to a printer or the Windows clipboard and allows access to other reports and Help Print The print menu pull down provides the choice of printing a multi page report with or without page breaks Suppressing page breaks will minimize paper usage with a cost of reduced readability Also included on the print pull down is the PageSetup choice where the user can customize printer parameters such as margins and page orientation Edit Click to copy the entire contents of the window to the Windows clipboard Refer to the following section for instructions on copying portions of the window to the clipboard SRM Reports Displays a drop down menu of the reports that may be
79. 1984 and climate affected runoff NErEasea DY dildir MEIG mida AR 81 Figure 45 Objects of the Graphical User Interface GUI WINdOW occccccccococcnnnnococoncnrononnnnononnnnnnoss 85 Figure 46 Additional GUI objects used by the WinSRM Interface ococcccccccnccccnnnnoconnnnnnorononnnnoncnnonoss 86 11 List of Ta bles Fable SRM applications ald Results ra EN 14 Table 2 Calculation of the melt of new snow deposited on a snow free area Pa 2 20 cm Toy EE 25 Table 3 Some of the possibilities of remote sensing for SNOW cover MAPPING ceseeceeeeeeeeeeeeeeeeees 26 Table 4 Results of model performances in the WMO project 10 years snowmelt season 46 Table 5 Errors experienced by SRM users and their correction o oococococccnnnnnoccnnnnnnononnnnorornnnororonannoss 49 Table 6 Seasonal redistribution of runoff for 1979 in the Rio Grande near Del Norte Colorado due to imate changes ei ia 69 Table 7 Seasonal redistribution of runoff for a normalized year in the Rio Grande near Del Norte Colorado que to clmate ee La Let 72 Table 8 Elevation zones and glacier areas in the Illecillewaet haen 78 Table 9 WinSRM capabilities and limitations EE 85 13 SNOWMELT RUNOFF MODEL SRM USER S MANUAL UPDATED EDITION 2008 WINDOWS VERSION 1 11 1 PREFACE This 2008 edition of the User s Manual presents a new computer program the Windows Version 1 11 of the Snowmelt Runoff Model WinSRM The
80. 3 26694 Winter clim Total Zonal Input 1 Zone 1 80 41334 2M S 39 92039 2M 1 S 0 ZP rain 40 49295 Winter Change AHW Zone 1 1 1 P P 31 198 Zone Winter Change deficit of 31 19781 was equaled ex ceeded by accumulated zonal melt on 5 5 a shift of 35 days AZM yxc on 5 5 31 42749 Prior day s A ZM 30 5473 Total P Winter Total P Winter clim Winter Total Zonal Input 1 Zone 2 17 38953 P 101 8937 P 122 2724 2M S 0 XM 1 S 4 349365 XP rain 13 04016 Winter clim Total Zonal Input 1 Zone 2 52 66322 2M S 7 564875 XM 1 S 6 676103 ZP rain 38 42224 Winter Change AHW Zone 2 1 1 P P 14 895 Zone 2 Winter Change deficit of 14 89496 was equaled ex ceeded by accumulated zonal melt on 5 17 a shift of 47 days AZM xa on 5 17 15 63624 Prior day s AZMexc 14 55844 160 Total P Winter P 122 8147 Total P Winter clim P 147 3776 Winter Total Zonal Input i Zone 3 0 2M S 0 XM 1 S 0 xP ran 0 Winter clim Total Zonal Input 1 Zone 3 16 24898 M S 8295755 XM 1 S 2 494999 ZP ran 12 92441 Gain g P 1 P 1 g 147 3776 16 24898 122 8147 0 1 067699 Winter Change AHW Zone 3 1 1 P P 8 314 Zone 3 Winter Change surplus 8 313959 was accounted for by stretching MDCs by a factor of 1 067695 Total P Winter Total P Winter clim P 122 8147 P 147 3776
81. 6 As a formal change from the SRM User s Manual of 1983 Martinec et al 1983 a negative sign appears in the exponent of Equation 7 so that the numerical values of x and y are positive 35 Medium Line Ote mes Qn m s Fig 10 Recession flow plot Qn vs Qm for the Dischma basin in Switzerland Either the solid envelope line or the dashed medium line is used to determine k values for computing the constants x and y in Equation 7 Martinec amp Rango 1986 The variability of k according to Equation 7 was also confirmed in other basins This means that the recession does not exactly follow the usual equation O Qs k 10 where Q the initial discharge Q the discharge after n days but the following equation Jaccard 1982 Q x E x where x and y are the constants of Equation 7 1 y 11 36 The envelope line in Figure 10 and the resulting values of x and y must be determined for each basin For ungauged basins and when historical discharge data are insufficient x and y can be derived indirectly from the size of the basin as follows A A Qu ii Ku o Nn 1 12 of one where Xm Ym are the known constants for the basin M Qu 0 are average discharge values from the basin M and the new basin N and Ay Ay are the areas of the respective basins Equation 12 indicates that recession coefficients are generally higher in large basins than in small basins
82. 80 64 64 Publ 1996 1987 1986 1996 2002 1989 1996 1994 2000 2000 Year appl 1988 1991 1983 1972 1982 1988 1990 1975 1988 1991 1993 1995 1995 If more than one year was evaluated averages of R and averages of Dy single values taken in absolute terms are listed climate change evaluated Ref Number of reference Section 11 2 Publ Year of publication Year appl Hydrological year of model application land use zones The accuracy criteria R and Dy listed in Table 1 are defined as where Ro Te DQ Gen D v a measure of model efficiency measured daily discharge simulated daily discharge Ve 8 Ve Va 100 average daily discharge for the simulation year or simulation season number of daily discharge values percentage difference between the total measured and simulated runoff measured runoff volume simulated runoff volume separate mapping of snow cover and glaciers 19 In addition to the input variables the area elevation curve of the basin is required If other basin characteristics are available forested area soil conditions antecedent precipitation and runoff data they are of course useful for facilitating the determination of the model parameters SRM can be used for the following purposes 1 Simulation of daily flows in a snowmelt season in a year or in a sequence of years
83. 90 260 240 VOLUMETRIC DIFF Dv 4 36 220 NASH SUTCLIFFE Ri 0 82 measured 200 180 al forecasted 160 E fis 140 120 100 e 80 60 e a 40 ore WI 20 0 ASAS 0 Discharge ms F gt April May June July August September Fig 22 Simulated real time runoff forecast for the Rio Grande basin using long term average temperatures instead of temperatures for the year 1983 Rango amp van Katwijk 1990 55 Forecast time 0 6 a 18 24 6 Te 18 24 day n day n 1 Fig 23 Real time availability of temperature and precipitation data for short term runoff forecasts in contrast to runoff simulation For other lag times SRM automatically combines Q in the appropriate proportions of two subsequent inputs as explained in Section 5 3 7 For example if L 24 h the input from T and P which might be already known at the time of the forecast is represented by 25 and the input from T and P forecasted values by 75 In the absence of temperature and precipitation forecasts runoff forecasts can be issued on condition for example that long term average values or extreme values maxima minima will occur It is also possible to use fictitious values as will be shown in the section dealing with climate change The feasibility of real time forecasts was demonstrated for two hydroelectric stations in the Swiss Alps Br sch 1996 With the use of snow cover monitoring by Landsat as well as of tempera
84. APPLICATION SRM can be applied in mountain basins of almost any size so far from 0 76 to 917 444 km and any elevation range Table 1 A model run starts with a known or estimated discharge value and can proceed for an unlimited number of days as long as the input variables temperature precipitation and snow covered area are provided As a test a 10 year period was computed without reference to measured discharges Martinec amp Rango 1986 The references pertinent to the following table can be seen in Section 11 2 Specific references for Table 1 at the end of this manual These references appear under the heading Ref and with a number to easily find them in Section 11 2 Table 1 SRM applications and results Size Elevation D y Year Country Basin 2 Range R x ar Zones Ref Publ appl km m a s l seasons 1 USA EGL Rocky Mountains 0 29 3300 3450 N A N A N A N A 72 1991 1989 2 USA WGL Rocky Mountains 0 6 3300 3450 N A N A N A N A 72 1991 1989 3 Germany Lange Bramke Harz 0 76 540 700 N A NA 1 1 25 1984 1981 4 Germany Wintertal Harz 0 76 560 754 N A N A l l 25 1984 1981 5 CzechR Modry Dul Krkonose 2 65 1000 1554 0 96 1 7 2 1 40 12 On eee 1970 1966 6 USA GLEES Rocky M 2 87 3300 3450 N A N A N A N A 72 1991 1989 7 Ecuador Antisana Andes 3 72 4500 5760 N A N A 1 3 19 1997 1996 8 Argentina Echaurren 4 5 3000 4200 0 84 7 5 l 1 14 1997 1985 H Spain Lago Mar Pyrenees 4 5 2234 3004 N A NA
85. C CDCcim wa 5k Replace CDC with CDCcrm wa ma Step 6 Summer simulation changed climate using CDCcim wa ma e Make one final summer climate change simulation using the existing climate change scenario with the derived CDCeLim wa curves to produce Qcrm 174 After a successful climate change computation Micro SRM changes state with references to actual and simulated runoff replaced with simulated runoff before after climate change The post change data is treated as normal Micro SRM data with the exception that 1t may not be used as the base for another climate change If climate modified data is SAVEd and reLOADed it induces the above referenced state change LOADing a normal SRM file resets the model to its normal state G 8 Using Micro SRM trace file options Several Micro SRM command line trace options allow the advanced model user to create and use files containing the commands provided to the model via the keyboard or mouse during all or part of a program execution This feature can prove very useful if a user is faced with the task of performing an elaborate sequence of otherwise identical processing steps on multiple years of data By building a trace file and using a text editor to modify references to any SRM filenames the user insures identical processing for each year of data In order for trace file processing to proceed in an orderly fashion Micro SRM makes the following assumptions when running i
86. Change Statistics Report is produced This printout disk file if no printer is available details the calculated values used by SRM to compute zonal Winter Change and the cutoff points or gain factors used in Step 5c to derive MDCexcr wa Sc Create data for a new curve MDCexcy wa To understand the methodology for modeling CDCerim it is important to understand the MDCexcy curve Each point along the curve is a daily intersection of snow water equivalent independent of melt season snowfalls 2 a T newmelt and daily snow covered area S on the x axis and y axis respectively The data are saved internally by SRM in ascending order by date SRM uses two different methodologies for creating MDCexcr wa 105 e Methodology for winter deficits Beginning on the date following that identified by Step Sb above i e the cutoff date all remaining daily values for MDCexcy S and a T newmelt are shifted backwards in time to the beginning of the melt season Additionally as the curve s x values are shifted each daily value is reduced by a constant equal to the Ist x value following the cutoff date reestablishing an x origin of 0 0 For example if a cutoff date of 15 April is identified 15 days into the melt season then all succeeding days for the data constituting the curve are shifted to the corresponding day 15 days earlier with each daily 2 a T newmelt being reduced by X a T newmelt day 16 e Methodology
87. EnterX lt The Enter lt key is used by Micro SRM to terminate keyboard responses Enter lt 4 implies that information has been provided to the model and the model should now react to that information Most commonly it signals that data entry for the current input data field 1s complete When pressed the model optionally validates the current field s contents for consistency range validity etc see field content validation above Micro SRM remembers each input field s most recent value and so only initiates validation when field contents change If the field value is unchanged or the changed value validates successfully the model accepts it into its internal schema and advances the screen cursor to the next logical field on the screen display If validation fails an appropriate error message is displayed and the cursor relocates to the first character of the offending field s value When used with a menu screen display or pop up window the Enter lt key directs the model to use the menu option marked by the light bar as your choice from that menu Enter lt is also used during the FilelO process to terminate a request for a filename by the model LeftClick another input field or Enter G 4 2 Cursor movement keys The keys described here are all capable of moving the screen cursor from the current data entry field to some other field in the same screen display When that occurs their use initiates a data validation response identical
88. G THE MODE E Ate 21 Sek BaSil Character Ee EE 21 DL Basi ANG ZONAS A ei 21 Dz Areas eleva cOn UVE tds 22 See 23 5 2 1 Temperature and degree days Te 23 DZ al IPAS CII LON LEE 24 D215 SNOW ENEE pl rodas darla 25 SEENEN 29 S3 LARUNGO COC cent oC vats EE ee Eege 29 Die EE ENEE dica 31 Did PEP SOUS UD SS Tate Min ido 32 5 5 4 Critical temperature Tetas das 33 5 3 9 Ralftall CONUMOULING area ACA as 33 3 90 RECESSION COCTIICIONL usa ds 34 SE ON AM A EE 40 Gb ASSESSMENT OF THE MODEL ACCURACY E 42 SEENEN 42 6 1 1 Accuracy criteria in model Tests 44 6 1 2 Model accuracy outside the snowmelt season 46 0 2 Elimination Of POSSIDIG CI Ol EE 46 7 OPERATION OF THE MODEL FOR REAL TIME FORECARTIE 50 7 Extranolation of Ine SHOW ee Ee LEE 50 E Bet E lee BET 55 8 YEAR ROUND RUNOFF SIMULATION FOR A CHANGED CONATE 57 8 1 Snowmelt runoff computation in the winter half vear 5 7 8 2 Change of snow accumulation in the new climate oococcccccccncococnncncononnnnoronnnnoroncnnnnonons 59 8 3 Runoff simulation for scenarios of the future climate ocococccccccccncnnnonononororornrorororonannnos 61 8 4 Model parameters in a changed climate ocoocococccoccnnncocccnnnoroncnnonoconnnnoronennonoroncnnoness 69 8 5 Normalization of data to represent the present cmate 70 8 6 WinSRM to improve real time runoff forecasts oocococccccnncnconcnnnnononononannonnnnorornnnororoannnonoss 17 9 RUNOFF MODELING IN GLACIERI ZED BASI
89. Geophysical Union Fall 1982 San Francisco 1982 11 Dey B Sharma V K amp Rango A 1989 A Test of Snowmelt Runoff Model for a Major River Basin in Western Himalayas Nordic Hydro ogy Vol 20 pp 167 178 12 Dincer T Payne B Martinec J Tongiorgi E Florkowski T 1970 Snowmelt runoff from measurements of Tritium and Oxygen 18 Water Resources Research AGU Vol 6 pp 110 124 122 13 Escobar C F Bales R Pozo V 1998 Aplicaci n de un modelo de derretimiento de nieves SRM 3 2 usando Sistemas de Informaci n Geogr fica SIG e imagines Landsat y NOAA Personal Communication 1998 14 Escobar C F amp Carin C 1997 Balance de Masa en el Glaciar Echaurren Norte 1975 a 1992 Resultados preliminares D G A H A y G 97 1 Complemento No 1 anos 1993 1996 1997 15 Escobar C F 1992 Aplicacion del modelo SRM 3 11 en cuencas de los Andes Centrales In Segundas Jornadas de Hidr ulica Francisco Javier Dominguez pages 283 295 Departamento de Hidrolog a Ministerio de Obras P blicas Santiago Chile Noviembre 1992 16 Ehrler C 1998 Klimaanderung und alpine Schneedecke Auswirkungen auf das Abflussregime am Beispiel des Einzugsgebiets Rhein Felsberg vdf Hochschulverlag an der ETH Zurich NFP 31 Schlussbericht edition 1998 117 Seiten 17 Ferguson R I 1984 Magnitude and modelling of snowmelt runoff in the Cairngorm mountains Scotland Hydrologi
90. HE LONG TERM MONTHLY AVERAGES WINDO Wo cccooccconccnnnoconnncononocnnnaccnnarononccnonacnnnaconanicnos 106 WINDOW 23 LONG TERM MONTHLY AVERAGES WINDOW ccoocccnoccnnnoccnnorononccnnncononarononconanocnnnarnnnarononocnonocnnnaronanicnns 106 WINDOW 24 ACCESS THE LONG TERM MONTHLY AVERAGES WINDONW 107 WINDOW 25 CUSTOM TIME LAG DISTRIBUTION WINDOW ccccccecccsccccsscceescccensccescseeeccseesssescseuecssensesensseeseceues 107 WINDOW 26 VIEW GRAPHICS WINDOW ras A AA A A e 107 WINDOW 2 7 GRAPHICS WINDOW piatsteascoctiacae EE EE EE 108 WINDOW 28 LINE ATTRIBUTES WINDOW A Ad 110 WINDOW 29 QUTPUT DEFINITION WINDOW coa 111 WINDOW 30 FILE DISPLAY WINDOW scsicsccecdcccataxidadcsheslocaatitescsttotecovertdeccdlevlecseieteesedctancseebheasslevionneeitevielsatenseecties 113 WINDOW 31 CREATE THE BASIN DESCRIPTION TABLE 114 WINDOW 32 POPULATE PHYSICAL DATA TABLE PROMDT 115 WINDOW 33 Z ONE DEFINITION GRID EE 115 WINDOW 34 BASIN SIMULATIONS GRID sr A A ia 116 WINDOW A1 COPYING NORMALIZED TEMPERATURE DATA TO THE WINDOWS CLIDPBOARD 135 WINDOW A2 REPLACING BASIN TEMPERATURE DATA WITH NORMALIZED DATA 136 WINDOW A3 NORMALIZED TEMPERATURE USING EXTERNALLY GENERATED ADJUSTMENT FACTORS 00 137 WINDOW El CLIMATE CHANGE DEFICIT GAIN SUMMARY REPORT nor nnnnncn nan nacnocnnnnnss 148 140 Appendix D WinSRM Database Schema WinSRM utilizes a Microsoft Access database as it s primary storage medium The tab
91. Lag Distribution Window 10 3 10 Graphics window 10 3 10 1 Method of access The Graphics window is viewed by clicking Graphics on the Window menu bar Window 26 WINSRM Snowmelt Runoff Model for Windows File Options Data Run Tools Graphic Reports Help Window 26 View Graphics Window 108 The Graphics Window Window 27 is the tool provided by WinSRM to manage the presentation and manipulation of simulation parameter data basin variable data and simulation results in a graphical 10 3 10 2 Purpose format Se x Bone qari rz00z T I I I I I I T I I I I I I i I I I I Avg Computed CO m s Volume Difference Er 0011978 to 09301979 Runoff Measured vs Computed CoPragram Files vvinsRMRioGrand mdb REY Rio Grande Basin Measured Runoff Computed Runoff Measured Runoff Volume 10 nii 1208 94 Average Measured Runoff m74s 38 336 Computed Runoff Volume 10 min 1212 018 36 435 Coefficient of Determination RE 100 50 C Program Files inS RAMA RloGrand rmdb A Sep6 1979 Y 262 5 a ul E i ro E Lo 5 7 SS Si Ee 2 os da an ay af d i MI SI Window 27 Graphics Window 109 10 3 10 3 Button bar menu The button bar menu at the top of the Graphics window see above and Appendix B provides easy access to all the underlying functionality supported by the window The following buttons are found on the button bar Save the image as a bitmap bmp
92. MDCexcr proportionally to this gain Ps 26 where P P precipitation in the present and changed climate I I winter input see Equation 24 in the present and changed climate Re writing Equation 24 AA I I 27 69 If there is no winter input in either case due to low temperatures a precipitation increase by 20 results in g 1 2 The x coordinates of MDCexcy cumulative snowmelt depth are multiplied by g to derive MDCexcL wa which conforms to the increased water equivalent of the snow cover on 1 April In the summer half year the position of CDCeim wa aS opposed to CDC results from the balance of the contradictory effects of an increased initial snow cover and of increased melting Table 6 Seasonal redistribution of runoff for 1979 in the Rio Grande near Del Norte Colorado basin due to climate change 1979 enn Leen Lee ses 72 maa ozs 12000 2 F 8 4 Model parameters in a changed climate SRM parameters are predetermined which requires more hydrological judgment than mechanical calibration or optimizing However as has been pointed out elsewhere Klemes 1985 Becker amp Serban 1990 Nash amp Gleick 1991 calibration models are not suitable for climate effect studies because the parameters cannot be meaningfully adapted to the conditions of a changed climate In the given example from the Rio Grande basin near Del Norte the seasonal change of the degree day factor a and of the runoff coef
93. Modified depletion curves for zone A MDC nc derived from CDC therefore including new snow MDCexc with new snowmelt excluded MDCexc wa with winter deficit shaded area SI 65 10 Figure 30 Accumulated zonal melt curves for zone A AZMiyc computed daily melt depth multiplied by S fom CDG ZO al Mell EE 66 Figure 31 Modified depletion curve adjusted for the winter deficit and including new snow of the changed climate MDCcum wa derived from MDCeycy wa fOr zone A ccececececeeeeeeeeeveeeeeeeereneeeeaees 66 Figure 32 Effect of a changed climate T 42C on snow covered areas of 1979 in elevation zones A B and C of the Rio Grande basin near Del Norte Colorado sssssssssssrsrsrsrsrsrsrrrrrrnrererersrsrrererene 67 Figure 33 Climate affected runoff T 42C in the Rio Grande basin near Del Norte Colorado compared with the runoff simulate by data of 1979 as shown in Figure 27 for April September 67 Figure 34 Simulated runoff in the Rio Grande basin near Del Norte Colorado in the hydrological year1979 and climate affected runoff computed by increased temperatures T 42C and correspondingly changed SNOW Conditions ococcccccocococcnnonococnnnononocnnnocononnnnoneronnnnorennnnenenonns 68 Figure 35 Normalized depletion curves of the snow coverage 9979 in the Rio Grande basin near Del Norte Colorado dashed lines derived from the measured curves of 1979 solid lines zones A B le KEE 73 Figure
94. NS ococcococcncoccococnoconcoconnorononcnnnoronnoronnnrornoronnrrnrnorenanss 78 9 1 Runoff increase in a warmer climate from glacier melt 78 9 2 Long term behavior of glaciers IN a warming mate 82 10 SRM FOR WINDOWS WINSRM COMPUTER PROGRAM 83 OED PO ele Lungen 83 EO Ve Me TE eidel ACK ONO e DEEN SE KEE EIERE Ee 84 10 13 Capabilities and neg el EE 85 10 2 WiInSR MUSE INEM ACC EE 85 102 1 ENV WNGOW featur E Sreet an a ida 85 103 WINSAM WINdOW UESCAPUONS isa dis 87 10 3 1 Welcome to WINSRM WINKOW c ccccceceeeeeeeeeeeeeeeeeeaeaeeeeaeaeesteeneaseetetasaneenetags 87 RAR PUD OS EE 87 EA A pe PRECIO PE RI e A OA 87 10032 WINSRM un Ee ene e UE EE 88 E ET 88 A Pe Un O A rN 88 A gu anaes siete Ee 93 10 3 2 4 Basi acera ton Tra E ooias EE 94 1013 20 Basil SIMULATION he EE 94 TO EI 95 10 3 3 Edit simulation control information window cccececeeceeeeeeeeeeeeaeeueeeeeareetetanens 95 LO 3 32 Method OT eg 95 L03 3 2 RUIDO Si EE 96 AB Et TEE 96 10 3 4 Edit type of data EEN ee EE e 96 AE Method Re Tee 96 EE Ee 97 10 3 4 3 Data selection data nl Vii ia 97 TS As BUON aeri E E AT 98 10 3 5 Climate change scenario definition window 99 t0331 MEA OA dada da 99 UR El ie EE 99 10 3 5 3 Data entry and Selecta A 100 LO 3704 De EE 100 10 3 6 Simulation statistics WINDOW EE 101 LO 5 0 Ee Ee 101 ELE 101 UA BURON EE 102 10 3 7 Run a climate Change WINdOW ssssesseresrsnrerrerrurnnrerrnrrrrnrrnrrnrnernrrnrrerrrr
95. PERATURE PRECIPITATION SNOW COVERED AREA 5 2 1 Temperature and degree days T In order to compute the daily snowmelt depths the number of degree days must be determined from temperature measurements or in a forecasting mode from temperature forecasts Temperature average Program options 0 daily mean 1 Min Max The program accepts either the daily mean temperature option 0 or two temperature values on each day Tmax Tmin option 1 The temperatures are extrapolated by the program from the base station elevation to the hypsometric mean elevations of the respective elevation zones For option 1 the average temperature is computed in each zone as Tax T Thin 2 T 3 When using daily means option 0 or when using Tmax Tmin option 1 it is recommended that negative temperature values when they occur be used in the calculation In line with this recommendation the original effective minimum temperature alternative automatic change of negative temperatures to 0 C was removed from the computer program beginning with Version 3 0 If the user still prefers this alternative the occasional negative temperatures can be changed manually to 0 C when inputting the data to SRM Because the average temperatures refer to a 24 hour period starting always at 0600 hrs they become degree days T C d The altitude adjustment AT in Equation 1 is computed as follows l AT y h h 4 temperature lapse rate C
96. Runoff Model Help Screen Version 4 0 HELP Program Options Start End 4 digits to define the processing period The period runs Date Prom Wat Uo Sita AMM elmo Wich ere Model Mode O Simulation n Updating where calculated Q is adjusted every nth day using actual flow values Temperature O Daily mean temperature provided Average F C 1 Use daily max min temperature to calculate daily means Temperature O a single set of daily temperatures will be applied to each Eeer hypsometric mean elevation using temperature lapse rates l Daily mean temperature to entered by user for each zone SRM will NOT lapse temperature Not used for Max Min temp Lapse Rates O Enter variable parameter values for zone A only all Precipitation remaining zones will be assumed identical to A Runoff Coeffs l Enter zone specific variable parameter values Critical Temp Runoff Avail required to run SRM in updating mode Esc exit help PgUp go to prior help screen PgDn go to next help screen Typical Help Screen 154 lt j Snowmelt Runoff Model SRM Version 4 0 Basin Variable Parameter Definition Menu IT to select Enter lt to edit values Precipitation P A Actual Stream Runoff ACTUAL B Maximum Daily Temperature TMAX Basin C Minimum Daily Temperature TMIN Variables D Average Daily Temperature T E E Snow Covered Area in Zone S Critical Temperature Tc Basin La age L Parameters ACTION la
97. Scenario Definition window is unable to create This example substitutes the climate of one year 1977 as the changed climate for a climate change simulation for year 1979 l Start the WinSRM program Load the Rio Grande database that is included with the software by clicking File Open In the Open an SRM database dialog double click the entry RioGrand mdb Control will return to the main window populated with the new database information 2 Click on RGWY79 in the Basin Simulations grid making it the current simulation 3 Click on Data Climate Change Scenario Definition When this window appears click on the Existing Scenarios drop down list and select Manuall This scenario will be used to recreate the SimResults table It changes temperature in the new climate by adding 0 to each temperature value shifts a parameter cs 31 days but does nothing else to the new climate Click when you have examined the scenario to return to the main window 133 Click on Run Climate Change The Run a Climate Change window appears Select Manual from the Scenario Definition s drop down list Click on Step1 Winter Present Climate As explained above Example 4 Step 5 the model recreates the SimResults table using the selected scenario to generate the changed climate variables and or parameters In the example the only effective change is the 31 day shift of snow runoff coefficients cs Simulat
98. T Plot File All the input and model generated values used to create Micro SRM s plots are saved to a user named external disk ette file PLT qualified The PLT file format is described in more detail in the sample PLT file included on the Micro SRM distribution diskette G 5 4 1 Plot displays climate change SRM Version 4 0 and later will automatically produce several additional plot displays during climate change simulation runs MDCrxc vs MDCexct wa The first curve depicts the snow water equivalent of the existing snow pack new snow excluded under normal climate conditions the amount of melt needed to decrease the snow covered area to a certain proportion of the total The second curve is derived from MDCrxcL by introducing the effect of a winter climate change upon areal snow water equivalent AZMincL A Zen AZMexct wa Accumulated zonal melt curves for the melt season with and without new snow falls and winter adjusted AZM The first two curves depict areal water equivalent of the initial snow cover under normal climate conditions the last areal water equivalent in a changed climate MDCrxc wa VS MDCcum wa These curves depict areal snow water equivalent of the initial snow cover in a changed climate with and without new snow added CDC vs CDCcy 1m wa These curves depict snow covered area for the normal and changed climate The first curve is taken from input data provided by the user The second curve is derived
99. Winter Total Zonal Input 1 Zone4 0 2M S 0 XM 1 S 0 xP ran 0 Winter clim Total Zonal Input 1 Zone 4 0 2M S 0 XM 1 S 0 xP ran 0 Gain g P 1 P 1 g 147 3776 0 122 8147 0 1 2 Winter Change AHW Zone 1 1 P P 24 563 Zone 4 Winter Change surplus 24 56294 was accounted for by stretching MDCs by a factor of 1 2 161 The climate change scenario described produced this Climate Change Statistics Report T 4 C P 120 31 day shift of a and Cs Note Zones 3 4 exhibit a positive Winter Change due to the increase in P 120 overwhelming the temperature induced 4 C deficit Climate Change Statistics Screen G 4 Keyboard definition User interaction with the model is initiated via the standard PC keyboard or the optional mouse During data entry the program recognizes and accepts all the standard ASCII characters present on the keyboard In addition Micro SRM provides both intuitive and non intuitive functionality to many of the special PC keyboard keys At any given point in the Micro SRM modeling process one or more of these keys may be hot that is set to initiate a specific program defined activity when pressed These hot key definitions are task specific and will be redefined or undefined as the model user changes tasks These special keys can be grouped into several broad categories G 4 1 Global definitions The following four keyboard keys
100. a da DI Blat Pra n 1 Fla Tr S s tao T a 1 S n Bis a n 1 where A Hw difference between the present and future areal water equivalent of the snow cover on April cm a degree day factor cm C d T temperature in the present climate at mean hypsometric elevation as degree days C d T temperature in a warmer climate as degree days C d S ratio of snow covered area to total area present climate S ratio of snow covered area to total area warmer climate Pz rain according to Tcrrr present climate P rain according to Tcrrr warmer climate 182 number of days October through March Equation 24 thus summarizes the SRM input to runoff which consists of snowmelt from the stable snow cover S melting of snow which temporarily covers the snow free area 1 S and rain The distinction between a stable and temporary snow cover during the snow accumulation in winter is rather arbitrary due to insufficient satellite monitoring but the total of both snowmelt inputs always equals 100 of the occurring snowmelt M M S M 1 S M S 1 S M 25 60 In March however if S is put to 0 while there is a stable snow cover and there happens to be little snowfall the snowmelt input may be underestimated Figure 25 illustrates the areal water equivalent of the snow cover on April as a difference between the winter precipitation and the winter input to runoff left side of Figure
101. abilities and limitations Maximum basin area No limit 917 444 km max to date Maximum simulation time period 366 days 10 2 WinSRM user interface 10 2 1 Entry window features Figures 45 and 46 show some of the commonly used Graphical User Interface GUI objects WinSRM uses with the features labeled Ti H e ba r WIR Smo Runoff Hodel for Windows 5 x gt Menu bar with Fle Options Data Bum Tock Grohe epis Help Dasin De fiatien Pull downs Text box fraen macia Member of Zoss Reference Seaton pl be Uais af Mesurement rte Letipeds dee fes Lengltude echz 33 Pecripbiea Display data Sp ETT Data entry Tables e ese P __ ate simstation otero simulation mp simlation Command button Status bar gt Cra Moda 1072575005 A Fig 45 Objects of the Graphical User Interface GUI window Feature Function Window Title bar Displays title of the window Select menu options with further sub options In this Menu Bar with Pull downs text references to menu choices will be appear in bold italics File Save Enter a single value into one field Data entry tables Enter related data into cells of a row Display Data Display only cannot enter data into a display area 86 Click to access the functionality In this text references Command button to command buttons will appear in a shaded bounded box Help Status Bar Displays program status information
102. able but deceptive value for R 0 97 is realized This value is high because the long term Q is much higher than any Q in 1977 Consequently the daily deviations of the simulated runoff become relatively insignificant although in absolute terms they are not negligible In addition to these criteria which are automatically computed and displayed after each model run the coefficient of gain from daily means DG could be computed by the user as follows 43 VQ Or DG p 23 gt Q OI 1 1 where 1s the measured daily discharge Q Q is the computed daily discharge Q isthe average measured discharge from the past years for each day of the period n is the number of days Thus R compares the performance of a model with no model average discharge and DG with a seasonal model long term average runoff pattern Negative values signal that the model performed worse than no model or worse than the seasonal model DRY YEAR 1977 T MEASURED E COMPUTED DI 1 D el G UI T Fig 13 Runoff simulations in the basin of the Rio Grande near Del Norte Colorado 3419 km 2432 4215 m a s l Martinec amp Rango 1989 44 6 1 1 Accuracy criteria in model tests The World Meteorological Organization WMO organized an international comparison of snowmelt runoff models in which hundreds of model runs were performed in six selected test basins Figure 14 shows a summary of all nu
103. and menu screens Each screen display can be broken into several component parts e The screen header lines 1 3 of the display identifies the screen and in some cases includes run specific information e The prompt line line 24 columns 1 71 provides a location where the program can display prompts and warning error messages as required e The keyboard state indicators line 24 columns 74 80 display the current state on off of the CapsLock and NumLock keyboard keys e The function key line line 25 describes important functions available while the screen is displayed e Screen hot areas highlighted bold text or characters may appear anywhere on the screen They draw attention to the screen area as with function keys and they provide a region that may be mouse clicked to provide a mouse equivalent to a keypress response e The remaining screen area lines 4 23 will have a type specific definition Micro SRM uses three types of screen display to communicate with the model user 153 G 3 2 Text screens The model introduction screen and fifteen Help screens provide the user with on line documentation to assist in resolving questions concerning the use of the computer program and to provide limited information about the model and the variables and parameters required to use it The introduction screen appears immediately after the program welcome screen The help screen most appropriate to the user s current location
104. annoversch Munden Deutscher Verband fur Wasserwirtschaft und Kulturbau Bonn Germany 123 26 Ishihara K Inoue M Takeda K 1985 Application of Martinec Rango model to river basin in Japan In Proceedings of the Second US apan Snow and Evaporation Workshop Honolulu Hawaii pages 53 59 NASA Conference Publ 2363 27 Jones E B Frick D M Barker P R 1984 Application of a snowmelt runoff model for flood prediction Presented at the Symposium on A Critical Assessment of Forecasting in Western Water Resource Management American Water Resources Association AWRA Seattle Washington USA 28 Kawata Y Kusaka T amp Ueno S 1988 Snowmelt Runoff Estimation using Snow Cover Extent Data and its Application to Optimum Control of Dam Water Level In GARSS 88 Symposium in Edinburgh Scotland pp 439 440 Paris 1988 ESA SP 284 Vol 1 29 Kawata Y amp Ueno S 1987 Basin Streamflow Estimation from Snow Cover Extent Data JAACE 19 pp 129 132 30 Kawata Y 1982 Estimation of Daily Basin Streamflow using Snow Cover Extent Data and its Application to Optimum Control of Dam Water Level Technical Report Kanazawa Institute of Technology Ishikawa 921 pp 255 260 Japan 31 Kim S Busch J R Molnau M 1986 Snowmelt Runoff Simulation using a Degree Day and Non Linear Multiple Recession Model Proceedings of the 54 Western Snow Conference Phoenix Arizona USA pag
105. ansfers control to the Output Definition window Window 29 where all the available WinSRM reports are selected and produced Reports has no drop down menu WINSRM Snowmelt Runoff Model for Windows Fil Options Data Run Tools Graphics Reports Help Window 9 WinSRM Main Window Reports 93 Menu e Explanation Displays the WinSRM Reports selection window the window EE that controls production of the available tabular reports Help The Help menu pull down organizes the on line help choices available for the main WinSRM window WINSRM Snowmelt Runoff Model for Windows S File Options Data Run Tools Graphics Reports Help Bazin Definition Help For Wim main window Show WinSRM Welcome Name Rio Grande Basin l El About SRM For Windows Fi Units of Measurement A etric System Information Window 10 WinSRM Main Window Help menu Pull down Choice Explanation Item Help for WinSRM SR Man Window Fl Displays help for the main window Show WinSRM l l E Help Displays WinSRM Welcome Window About SRM for Displays description of the program disclaimer program Windows version number and date System Information Displays information about the your computer 10 3 2 3 Buttons dd Simulati l e SE Click this button to create a new simulation for the current database The Edit Simulation Control Information window is displayed allowing the user to create the shell parameter data table usin
106. aph of modified depletion curves for the elevation zone B 1284 km 2926 3353 m a s l of the Rio Grande basin near Del Norte Colorado sssssssssssssssrsrsrsrsrerererrrererererrrereseses 54 Figure 22 Simulated real time runoff forecast for the Rio Grande basin using long term average temperatures instead of temperatures for the year 1983 Rango van Katwijk 1990 54 Figure 23 Real time availability of temperature and precipitation data for short term runoff forecasts in contrastito uno En UE Le EE 55 Figure 24 Discharge simulation in the Dinwoody Creek basin 228 km 1981 4202 m a s in Wyoming a without updating b with updating by actual discharge on 1 AUGUST ccccceceeeee sees eeeeeees 56 Figure 25 Illustration of the snow accumulation in the winter and snowmelt in the summer in the present climate hypothetical example sancire tiatn a a a 60 Figure 26 Illustration of the snow accumulation in the winter and snowmelt in the summer in a warmer Climate hypothetical example ccceccececceceeseceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeaeeeeeeeeeeeseraeeeeaeseteneeags 61 Figure 27 Measured and simulated runoff in the Rio Grande basin near Del Norte Colorado in the Marodi cal yea TOTI EE 63 Figure 28 Conventional depletion curves of the snow coverage from Landsat data in the elevation zones A B and C of the Rio Grande basin near Del Norte Colorado in 1979 ssssssssssssssesrsrrrrrrrererns 65 Figure 29
107. are avoided In any event the SRM program finishes this task including printout of figures tables numerical results and hydrographs within minutes Another advantage is the independence from calibration enabling the model parameters to be meaningfully shifted in time or adjusted if so indicated by climate scenarios If SRM is used only to predict snow covered areas and regional snow accumulation in a changed climate for example in mountain areas which are not hydrologically defined basins the computer program requires the following data e Number of elevation zones their areas and hypsometric mean elevations e Current snow covered areas daily values S e Daily average or Max Min temperatures T e Precipitation daily P e Degree day factor a e Temperature lapse rate y e Critical temperature Tcrrr For runoff computation the remaining SRM parameters are required Cs Cr RCA k L and the initial Q Originally SRM was applied to simulate and forecast runoff only during the snowmelt season It was later run year round for international tests of model performance WMO 1986 The first evaluation of the effect of climate change of SRM dates back to 1980 Martinec 1980 At that time there was no consensus yet whether temperature will increase or decrease Therefore the runoff in the snowmelt season was computed for 1 C and UC Since then climate scenarios anticipating increasing temperatures have become available f
108. asin are stored in this table and are shared by each of the basin s simulations Providing values for these new records is described below in another section ZoneDescription table After the Populate dialog is released by clicking or Accept dates control returns to the main window At this point the user provides the critical values required to describe the zones that will subdivide this basin Each line in the grid shown below Window 33 represents a single elevation zone in the new basin Currently the model requires values for zone elevation and zone area only The remaining columns in the grid may be provided for descriptive purposes but are not used by the model s processing algorithms The information contained in this grid is stored in the ZoneDescription table Fone Ib Zone Area Hypsometric SNE Aug Elevation ME E SE Aug Elevation SE L NW E Elevation MW km mean Seaton Aspect Aspect m Aspect Aspect mi Aspect Aspect mi 100 T REH S 3 Window 33 Zone Definition grid 116 SimulationIndex table As its name indicates this table is the index of all the simulations that exist in the basin database Each record in the table represents a unique basin simulation and stores the control information name run starting and ending date etc required by the model when the simulation is run This information is displayed in the Basin Simulations grid on the WinSRM main window Window 34 Each simu
109. atform Spectral Bands Spatial Minimum Repeat Sensor resolution area size period Aircraft Orthophoto Visible NIR 2 m 1 km flexible Green to NIR 2 km 1 3 Green to NIR 2 5 5 km 1 Red 2 NIR 10 20 km 1 3 Green to NIR 1 4 Green to NIR 10 20 km 16 18 days 1 4 Green to NIR 2 5 5 km 16 18 days Visible to NIR 2 3 km 16 18 days Terra Aqua ASTER 1 3 Visible to NIR 2 3 km 16 days 1 Red 2 NIR 20 50 km 1 day MODIS 3 8 Blue to MIR 50 100 km 1 day NOAA AVHRR 1 Red 2 NIR 10 500 km 12 hr Meteosat 1 3 Red to NIR 3 km 500 1000 km 30 min SEVIRI 12 Visible 1 km 10 500 km 30 min Acronyms ASTER Advanced Spaceborne Thermal Emission and Reflection Radiometer e AVHRR Advanced Very High Resolution Radiometer HRVIR High Resolution Visible and Near Infrared IRS Indian Remote Sensing e LISS Linear Imaging Self scanning Sensor e MIR Middle Infrared MODIS Moderate Resolution I maging Spectroradiometer e MSS Multi Spectral Scanner NIR Near Infrared Pan Panchromatic SEVIRI Spinning Enhanced Visible and Infrared Imager SPOT Satellite Pour l Observation de La Terre TM Thematic Mapper WiFS Wide Field Sensor ETM Pan Enhanced Thematic Mapper Panchromatic Landsat 6 and 7 only Depends on availability 2 16 HMAY 1965 24 MAY 1965 Fig 4 Sequence of snow cover maps from Landsat 5 MSS Upper Rhine River a
110. attered a correctly evaluated degree day factor will produce less meltwater than if a 100 percent snow cover were assumed A meltwater difference that arises from erroneous snow cover information should not be compensated for by optimizing the degree day factor Instead the correct areal extent of the snow cover should be determined and used 4 In large area extrapolations point measurements should be weighted depending on how well a specific station represents the hydrological characteristics of a given zone Shafer er al 1981 In the absence of detailed data the degree day factor can be obtained from an empirical relation Martinec 1960 a 11 Ls 6 oe where a the degree day factor cm C d P density of snow Pw density of water When the snow density increases the albedo decreases and the liquid water content in snow increases Thus the snow density 1s an index of the changing properties which favor the snowmelt 31 A 1979 Fig 7 Average runoff coefficient for snow cs for the alpine basins Dischma 43 3 km 1668 3146 m a s l and Durance 2170 km 786 4105 m a s 1 Martinec amp Rango 1986 APRIL JUNE JULY AUGUST Fig 8 Average runoff coefficient for rainfall e for the alpine basins Dischma 43 3 km 1668 3146 m a s 1 and Durance 2170 km 786 4105 m a s 1 Martinec Rango 1986 a2 Figure 9 illustrates the seasonal trend of the degree day factor in the Alps and i
111. audio tone level and video graphics mode Included on the distribution disk are several sample CFG files and a utility program CONFIG EXE that gives a user the ability to customize the model s interface for his her unique hardware configuration G 2 4 Operating instructions The following command syntax runs Micro SRM from the default directory Filenames must be complete pathnames if the target file does not reside in the default directory SRMA configfile option s Enter lt 4 The following list describes valid Micro SRM command line options Values enclosed in brackets denote SRM default values Argument Description Config File See description above SRM CFG TIN tracefile Control SRM via a trace file not the keyboard See previous section for a complete description of the trace feature TOUT tracefile Save keypresses from current run in a tracefile TDELAY seconds Use with TIN to vary speed of trace operation 5 See previous section for more on the trace delay P printfile Redirect LPT 1 printed output to a disk file The following examples demonstrate several common command line option combinations for initiating Micro SRM C gt SRM4 Run the model from the default in this case the root directory using the default configuration file C SRM CFG stored in that directory C gt A 152 A gt SRM4 EGA CFG P SRMPRINT PRT Switch to the A drive then run the model from the A default direc
112. basin according to Equation 1 A 10000 Quis esn 8 T AT Spt Cra Pal seng En Ob Ge where Q average daily discharge m s c runoff coefficient expressing the losses as a ratio runoff precipitation with cs referring to snowmelt and c to rain a degree day factor cm C d indicating the snowmelt depth resulting from 1 degree day T number of degree days C d AT the adjustment by temperature lapse rate when extrapolating the temperature from the station to the average hypsometric elevation of the basin or zone C d S ratio of the snow covered area to the total area P precipitation contributing to runoff cm A preselected threshold temperature Terrr determines whether this contribution is rainfall and immediate If precipitation is determined by Terry to be new snow it is kept on storage over the hitherto snow free area until melting conditions occur A area of the basin or zone km 20 k recession coefficient indicating the decline of discharge in a period without snowmelt or rainfall k Ce m m 1 are the sequence of days during a true recession flow period m n sequence of days during the discharge computation period Equation 1 is written for a time lag between the daily temperature cycle and the resulting discharge cycle of 18 hours In this case the number of degree days measured on the nth day corresponds to the discharge on the n 1 day Various lag times can be
113. berg from the Landsat imagery Baumgartner 1987 Baumgartner amp Rango 1995 Together with temperature and precipitation data such depletion curves enable SRM to simulate runoff in a past year In September the depletion curve in zone E refers to the glacier which prevents it from decreasing any further For real time runoff forecasts however it is necessary to know the snow covered area within days after a satellite overflight and also to extrapolate the depletion curves of the snow coverage to the future weeks This procedure 1s explained in the Section 7 5 3 Parameters RUNOFF COEFF SNOW CRITICAL TEMPERATURE RUNOFF COEFF RAIN RAINFALL CONTRIBUTING AREA DEGREE DAY FACTOR RECESSION COEFF TEMPERATURE LAPSE RATE TIME LAG The SRM parameters are not calibrated or optimized by historical data They can be either derived from measurement or estimated by hydrological judgment taking into account the basin characteristics physical laws and theoretical relations or empirical regression relations Occasional subsequent adjustments should never exceed the range of physically and hydrologically acceptable values 5 3 1 Runoff coefficient c The runoff coefficient accounts for the losses which are the difference between the available water volume snowmelt rainfall and the outflow from the basin On a long term basis 1t should correspond to the ratio of the measured precipitation to the measured runoff In fact compari
114. by the climate change algorithms of SRM G 5 5 Printed reports Micro SRM includes 7 different printed report formats each display one or more model variable s The menu controlling selection of printed reports allows the user to select an individual report or the entire report set see previous sections for more information on the F6 print function Printing can be canceled by pressing ESCape at any time during printing if print buffering is in effect there will be some delay before ESCape takes effect Each selected report will print even if there are no data present 1 e neither input nor computed The following reports are available 169 Temperature Values form Prints temperature values entered by the user in the temperature form selected max min mean for each day in the snowmelt period as specified on the Program Options Screen Degree Day Factors Runoff Coefficients Daily degree day factors AN and snow rain runoff coefficients are printed for each day of the melt period Lapse Rate Critical Temperature Rainfall Contributing Area Lag Time Parameters displayed control how discrete temperature values are applied basin wide lapse rate when to treat precip as snow critical temp how to calculate the effect of precipitation RCA and how to distribute resulting runoff over time lag time Zone Degree Days Observed Precip Snow Covered Area A zone specific report that displays DD temperatur
115. cal Sciences Journal Vol 29 1 pp 49 62 18 Fitzharris B 1991 Runoff Personal Communication 19 Galarraga R 1997 Runoff Personal Communication 20 Garr T 1996 Runoff Geehi River Australia Personal Communication 1996 21 G mez Landesa E 2002 Snowmelt Runoff Model Worldwide Applications Personal Communication nternational Snowmelt Forecast Seminar U S AID Central Asia Natural Resources Management Project NRMP Tashkent Uzbekistan J une 2002 22 G mez Landesa E Rango A 2002 Operational Snowmelt Runoff Forecasting in the Spanish Pyrenees using the Snowmelt Runoff Model Hydro ogical Processes Vol No 16 pp 1583 1591 23 G mez Landesa E Rango A Hall D K 2001 Improved Snow Cover Remote Sensing for Snowmelt Runoff Forecasting In Remote Sensing and Hydrology 2000 Proceedings of the Santa Fe Symposium April 2000 AHS Publication No 267 pp 61 65 24 G mez Landesa E 1997 Evaluaci n de Recursos de Aqua en Forma de Nieve medianteTeledetecci n usando Satelites de la Serie NOAA Estimation of snow water resources using remote sensing with NOAA satellites Ph D Thesis ETSICCP Universidad Polit cnica de Madrid Spain February 1997 25 Herrmann A amp Rau R G 1984 Snow cover stores and winter runoff behaviour of a small basin in the German highlands n Proceedings IUFRO Workshop on Snow Hydrologic Research in Central Europe number 7 H
116. cal data are unique Many simulations in the database may utilize the same set of daily values If modeling requires differing sets of these physical data the solution is to create another copy of the database see Window 3 with the alternative set of physical data Data ge Field Name Description Field Name Type Zone Integer Zone identifier 1 10 Date Month Day Year these data were measured mm dd yyyy Runoff Sinale These values represent the total runoff flowing from the basin They are 9 NOT zone specific Runoff is always stored in zone 1 f s m s Daily maximum temperature at the hypsometric mean elevation HME of the zone If basinwide is indicated see Simulation ndex table the temperatures are assumed to have been measured at the reference elevation see BasinDescription table and will be lapsed to the HME Basinwide variables are always stored in the first zone F C Daily average temperature Same conditions as Tmax The model uses Tavg 2 Single max min or average temperatures as specified in a given simulation s temperature type flag see Simulation ndex table F C Precipitation measured either at the HME of the zone or at the reference Precip 1 Single elevation As with the other variables if precipitation is flagged as basinwide see Simulation ndex it is stored in the first zone snow covered These values MUST be provided for each zone NetRadiation unused 1 required 2 eit
117. ccomplished by clicking the Use the Climate Changes Defined in Results check box Note that you must still select a scenario from the combo box as that is where WinSRM obtains the Winter End Date value needed during processing Control After the scenario has been determined the model is ready to begin processing A climate change run is in reality four complete iterations through the model algorithms Steps 1 2 3 6 one partial model execution Step 4 and a data generation step Step 5 wherein climate changed snow cover data is modeled Each step is executed in sequence by clicking the active button on the Climate Change Processing Steps frame Reporting as each processing step is completed the Statistics data grid is updated to display the statistics resulting from the completion of the step Steps 1 2 3 6 only In addition to this reporting of simulation statistics the user can access the main window s plotting and reporting capabilities by clicking on the main window menu bar Graphics or Reports 10 3 7 3 Buttons E Return to main window Print statistics Display help window Enabled after step 6 after Step 6 V completes Ba Print Help A ara Toggle Quickhelp Terminate the run return to main window 10 3 7 4 Processing steps controlled by the step buttons Step 1 Winter Present Climate Prior to a WinSRM simulation with on exception a copy of the pertinent
118. d to climatic conditions In SRM simulations a lapse rate of 0 65 C per 100 m was usually employed Slightly higher seasonally changing values appeared to be adequate in the Rocky Mountains e DISCHMA e ONWOOLDY A DURANCE Fj ZL Fig 9 Average degree day ratio a used in runoff simulations by the SRM model in the basins Dischma 10 years Durance 5 years and Dinwoody 228 km 1981 4202 m a s 1 Wyoming 2 years Martinec amp Rango 1986 33 The computer program accepts either a single or a basin wide lapse rate option 0 or different rates for each zone option 1 Seasonal variations can also be accommodated by inputting predetermined lapse rates every 15 days and the lapse rate can be changed manually on selected days if a special meteorological situation for example a temperature inversion requires a different value Temperature lapse rate Program options 0 basin wide 1 by zone If the temperature station is situated near the mean elevation of the basin possible errors in the lapse rate are to some extent canceled out because the extrapolation of temperature takes place upwards as well as downwards If on the other hand the temperature station is at a low altitude SRM becomes sensitive to the lapse rate For some basins in Wyoming for example the closest temperature station was more than 100 km away and 2000 m lower than the highest snow covered parts of the basins In such a case an error in the laps
119. de Window 22 Access the Long term Monthly Averages window 10 3 8 2 Purpose The Long term Monthly Averages Window Window 23 calculates and displays monthly averages for daily maximum temperature daily minimum temperature daily average temperature and daily precipitation for the period of record reflected in the current basin database The averages are stored in the Basin Averages database table w Long term Monthly Averages E ol x Long Term Monthly Averages Close window return to Maximum Minimum Average main window no ae p e mape rete SE updating 0 000 0 000 5 826 0 091 Update BasinAverages table return to main window SI Print Help Cancel Accept yGrand mdb 127 10 20 Zone 1 of 3 C Program Files ains A Display complete Compute long term monthly Show quickhelp Print values to the Help window averages for the current zone panel system printer Window 23 Long term Monthly Averages Window These monthly averages are used during climate change processing when the climate change scenario includes the normalize edit action for temperature and or precipitation To normalize daily temperature values the program calculates the difference between the monthly average temperature simulation and monthly average temperature period of record and applies that difference or normalization factor to each daily temperature value in the respective month The resulting
120. del applied to the Tillouguit Basin of Morocco Master s Thesis South Dakota School of Mines and Technology Rapid City South Dakota USA 3 Bagchi A K 1983 Generation of the snowline Photogrammetric Engineering and Remote Sensing Vol 49 12 pp 1679 1689 4 Baumgartner M F 2002 Runoff in Pskem Chatkal Angren Naryn Karadarya Vaksh Zerafshan Personal communication 2002 5 Baumgartner M F Spreafico M amp Weiss H W 2001 Operational snowmelt runoff forecasting in the Central Asian mountains In Remote Sensing and Hydrology 2000 Proceedings of the Santa Fe Symposium April 2000 AHS Publ No 267 pp 66 71 6 Baumgartner M F 1987 Schneeschmelz Abflusssimulationen basierend auf Schneeflachen bestimmungen mit digitalen Landsat MSS und NOAA AVHRR Daten PAD thesis University of Zurich 1987 Referent H Haefner Gutachter K Seidel 7 Baumgartner M F amp Rango A 1995 A Microcomputer Based Alpine Snow Cover Analysis System ASCAS Photogrammetric Engineering and Remote Sensing Vol No 61 12 pp 1475 1486 December 1995 8 Bedford D 1996 Data on Kafirnigan Vakhsh Pyandzh Persona communication 1996 9 Boyer E W 1990 Martinec and Rango s Snowmelt Runoff Model Applied to the Upper Yakima River Personal Communication 1990 10 Burn D E amp McBea E A 1982 River flow forecasting model for the Sturgeon River In Meeting of the American
121. ditions modeled by a climate scenario The present conditions can be represented by precipitation temperatures snow covered areas and simulated runoff of a single year especially if this year appears to be a nearly average year Apparently averaging data from a number of years would be more representative However using long term average temperatures which smooths out daily fluctuations for snowmelt runoff computations for example may result in a considerable underestimation of snowmelt and runoff as mentioned in Table 5 case 13 The following procedure is recommended see also Appendix A Example 6 1 Select a hydrological year 2 Determine deviations of monthly average temperature of the selected year from long term means of monthly average temperatures see the example ahead AT Ls Mies ee AT different in each month 3 Adjust daily temperatures of the selected year to obtain daily normalized values with daily fluctuations preserved Lom lya AI AT different in each month 4 Determine ratios of the monthly mean precipitation of the selected year to long term monthly mean precipitation P long term average Lern aeee r different in each month Pie 5 Adjust daily precipitation of the selected year to obtain daily normalized values Pomi Poar T r different in each month The third variable the snow cover S is evaluated differently in the winter half and in the summer half of the
122. e either computed from max min average daily temperature or entered directly as DD s observed precip these values may be zone specific or apply across the entire basin and snow covered area of zone covered on a given day Melt Depth M S Melt New Snow M 1 S Precip Contributing to Runoff Intermediate results of Micro SRM computations For each elevation zone this report details daily total melt depth zone contribution to runoff M S and precip contributing to runoff either as rain or as melted new snow Measured vs Computed Snowmelt Runoff Simulated forecast and measured streamflow if provided are plotted against time Run statistics R Volumetric Difference are meaningful only 1f actual runoff values have been provided Input Summary Report Run Statistics A printout of the model summary display Bi monthly values 1st 16th for 7 model parameters are printed for each zone in the basin Run statistics are for the most current model run All the Above Generates a complete set of printed reports G 5 6 Printed reports climate change Version 4 0 of SRM includes one additional printed report the Climate Change Statistics Report that presents the computed values used to derive climate affected snow cover 170 G 6 Using Micro SRM Assuming the availability of the data necessary to run the model there are only two requirements for using Micro SRM The target basin must contain 8 or fewer elevation zone
123. e 12 32 Kumar V S Haefner H Seidel K 1991 Satellite Snow Cover Mapping and Snowmelt Runoff Modelling in Beas Basin 20 General Assembly IUGG in Vienna 1991 IAHS IUFRO Symposium Snow Hydrology and Forests in High Alpine Areas pp 101 109 AHS Publication No 205 33 Kumar V S 1991 Runoff Parbati Saing Buntar Personal Communication 1991 34 Litwin C J 1991 Runoff Tierra del Fuego Argentina Personal Communication Consejo Federal de Inversiones San Martin 871 Buenos Aires Republica Argentina 35 Ma Hong 2000 Runoff simulations Gongnisi Urunqi Master s Thesis Chinese Academy of Sciences Institute of Geography Xinjiang Peoples Republic of China 2000 Personal communication 36 Martinec J 1982 Transfer of results on snowmelt runoff from small to big basins In Proceedings of the Symposium on Hydrological Research Basins pages 801 809 Landeshydrologie Berne Switzerland 1982 37 Martinec J amp Rango A 1979 Discharge forecasts in mountain basins based on satellite snowcover observations Operational Applications of Satellite Snow Cover Observations Proceedings of a Workshop at Sparks U S A pp 223 238 NASA Conference Publ 2116 38 124 Martinec J 1975 New methods in snowmelt runoff studies in representative basins AHS Symposium on Hydrological Characteristics of River Basins Tokio Japan pp 99 107 AHS Publ No 117 39 Martinec
124. e Change Progress Screen Step 1 Winter ai j normal climate determine total zonal melt Step 2 Winter simulati changed climate determine total zonal melt SES tep 3 Summer simulation normal climate save MDC simulated runoff S Step 4 Summer simulation changed climate save MDCcrrm Step 5 Create 1M wa using MDCcrrm adjusted for Winter Change S 4 tep 6 Summer simulation changed climate using CDCorim wa Step 1 2M S y A Ale cada Zonal le Winter Change 404 O28 INES CANES 0998 2000 000 349 13 040 EE 000 000 CHOC SSES 0000 000 000 000 0 000 0000 000 Simularced O on 3 31 EscAbort FlHelp F2Summary F5Plot F6Print lt J Next Step Climate Change Progress Screen Step 6 models CDCcum wa ma During this step several new output products are automatically generated by the model A printed report the Climate Change Statistics Report or optionally a disk file SRM LOG on systems without a printer displays the computed values used to derive these climate modified CDC s The following curves are displayed in four new plots allowing the user to visualize the modeling process MDCexct MDCexcr WA AZMicL AZMgxcL AZMexcr wa MDCexcL wa MDCcLim wa CDCcum LC DC cm wa Date 10 10 1997 File IL840606 SRM Total P Winter P 82 57877 Total P Winter clim P 99 09452 Winter Total Zonal Input i Zone 1 32 69977 2M S 2 404175 XM 1 S 7 028653 XP rain 2
125. e losses and of the forthcoming precipitation Seasonal runoff forecasts can be issued for different precipitation eventualities 5 The climate program is run for the summer half of the year considering daily normalized T P see Section 8 5 as a new climate and MDCcym wa in other words MDCinct E is derived The conventional curve CDC is derived from MDCnct r SRM is run with Thom Prom and CDC This is a day to day forecast for the next 6 months assuming normal precipitation and temperature This forecast can be updated by actual and forecasted temperatures and precipitation It is also possible to use as a new climate instead of Thorm and Phorm the T and P values of an extremely dry or extremely wet summer which could occur with a certain probability 78 9 RUNOFF MODELING IN GLACIERIZED BASINS 9 1 Runoff increase in a warmer climate from glacier melt The presence of glaciers or persistent late lying snow fields in an elevation zone is indicated when the decrease of the snow coverage stops at a certain level which corresponds to the glacier area Approximately from this date on the daily melt depths printed out by SRM Version 4 as well as by WinSRM can be attributed to glaciers instead of the seasonal snow cover and the total glacier melt depth can be computed In a warmer climate the depletion curves of the snow coverage decreases to the glacier area on earlier date and more glacier ice is m
126. e of computer analysis and a Digital Elevation Model DEM 8222 ft 2506 m ad STREAMGAGE METEOROLOGICAL STATION km mi 2 ZONEA 125 48 22 SS ZONE B 269 104 48 ZONE C 165 64 20 TOTAL 559 216 100 Fig 2 Elevation zones and areas of the South Fork of the Rio Grande basin Colorado USA 5 1 2 Area elevation curve By using the zone boundaries plus other selected contour lines in the basin the areas enclosed by various elevation contours can be determined by planimetering These data can be plotted area vs elevation and an area elevation hypsometric curve derived as shown in Figure 3 for the South Fork basin This area elevation curve can also be derived automatically if the user has access to digital elevation data and computer algorithms used in an image processing system The zonal mean hypsometric elevation h can then be determined from this curve by balancing the areas above and below the mean elevation as shown in Figure 3 The h value is used as the elevation to which base station temperatures are extrapolated for the calculation of zonal degree days 22 ELEVATION op a s 1 Fig 3 Determination of zonal mean hypsometric elevations h using an area elevation curve for the 2314 ZONE C h 3512 m rai A GC wv EN ee ee ro E AAA RA ESE ZONE B h 23443 m ZONE A h 2742 m 100 200 ERR q AREA kr South Fork of the Rio Grande basin 23 5 2 Variables TEM
127. e rate of only 0 05 C per 100 m causes a deviation of 1 degree day from the correct degree day value which corresponds to an error of about 0 5 cm of the computed daily meltwater depth late in the snowmelt season Such situations sometimes necessitate an adjustment of the originally selected lapse rate taking into account the course of the depletion curves of snow coverage If high temperatures result from extrapolation by a certain lapse rate but no change in the snow areal extent is observed then probably no appreciable snowmelt is taking place at this altitude The high temperatures result from a false lapse rate which must be increased or decreased depending on whether the temperature station is lower or higher than the mean zone elevation 5 3 4 Critical temperature T crit The critical temperature determines whether the measured or forecasted precipitation is rain or snow Models which simulate the build up of the snow cover first m order to simulate the runoff depend heavily on this parameter not only in the ablation period but particularly in the accumulation period SRM needs the critical temperature only in order to decide whether precipitation immediately contributes to runoff rain or if T lt Terr whether snowfall took place In this case SRM automatically keeps the newly fallen snow in storage until it is melted on subsequent warm days as explained in the Section 52 2 When SRM was applied in the alpine basin Dischma Tcrrr
128. e user may interrupt the data entry activity by pressing an active function key to request help save data display the data summary screen etc After data entry is complete the simulation or forecast can be computed press F7 and the results displayed F5 for plots F6 for print Prior to each F7Compute SRM performs a simple non elegant test to determine if all appropriate variables and parameters have been supplied by the user For each required variable or parameter type determined by Program Option field values SRM simply checks for the presence of a non zero value anywhere in the period If any required variable or parameter fails this test a data validation error screen is displayed that identifies the variable s and zone s that have failed the test To view this screen simply run Micro SRM enter some values on the Basin Definition Screen and press F7Compute Because certain model variables and parameters may legitimately contain all zero values De precipitation the model user is given the opportunity to continue the simulation or forecast or abort processing so the missing data condition may be corrected 171 G 7 Using Micro SRM to simulate a year round climate change Beginning with Version 3 11 Micro SRM allowed a user to define a climate change scenario a seasonal change of temperature and precipitation and simulate its effect during the melt season This feature was an important step in implementing the ongo
129. ed on the three successive days This procedure is slightly changed in the winter as it will be explained later 25 In this example S is decreasing on consecutive days because it is interpolated previously from the snow cover depletion curve In reality it should remain constant as long as the seasonal snowpack is covered with new snow however the model currently uses the incremental decrease of S shown in Table 2 1 0 C Table 2 Calculation of the melt of new snow on a snow free area Pa 2 20 cm Torr Melted P P contributing Depth Stored to Runoff a T cm em a T 1 S em pon os o ov 22 o 20 us om or o oos 215 n 1 me oss 270 oss o 12 os ma oas 370 066 o 093 o Sharp peaks of discharge are typical for rainfall runoff as opposed to the relatively regular daily fluctuations of the snowmelt runoff SRM has been adapted to better simulate these rainfall peaks whenever the average daily rainfall calculated over the whole basin equals or exceeds 6 cm This threshold can be changed by the user according to the characteristics of the selected basin The procedure is described in the Section 5 3 6 1 in connection with the recession coefficient In spite of these precautions rainfall runoff peaks may cause problems because local rainstorms may be missed if the network of precipitation stations is not dense enough Also the timing of rainfalls within the 24 hour period is frequently
130. elevation zone within the basin Micro SRM paints a screen display format that represents two calendar months of a given parameter or variable for a given elevation zone The initial two months of data displayed in the format are for the first two months of the specified snowmelt period and for the lowest or only elevation zone zone A The model then uses the PgUp PgDn keys to superimpose the next prior two months values onto the screen If the basin contains multiple zones Function keys F5 F6 are used to superimpose the prior next zone s values for the 2 month period shown Any given display is self explanatory showing the variable parameter name elevation zone month names and hot key areas and definitions See previuos section for more information on these hot keys lt i LS Snowmelt Runoff Model SRM Version 4 0 Basin Definition Basin Name Durance Basin Number of Zones 1 8 5 Model Year 1975 SE chibi rae bl eI Nh ek 2 Parameters for computing recession coefficients X 1 0849 2 220 e Y M ee AO A Effective Date 0401 Bitective Date Reference Elevation 0 Retin ala ares nica oo Zone Hypsometric Zone Hypsometric Zone Area Mean Elevation Zone Area Mean Elevation A 219 1141 B 495 1680 C 783 2154 D 536 250d ESO E TD OSes 3074 F SS G H EscQuit FlHelp F2Summary F3FilelO F5Plot acicate F7Compute Basin Definition Data Entry Screen 157 1975 Snowmelt Runoff Model SRM Version 4 0
131. elted Again the total amount can be computed from the daily values For comparison the glacier melt can also be evaluated from the resulting increase of climate affected runoff However the carry over from runoff and unmelted snow from the previous hydrological year and to the following hydrological year must be taken into account as follows Rom R Pcrm P Ma cum Me c ST cum ST c ACOciim ACO 28 Where R runoff depth in a hydrological year cm P precipitation cm M glacier melt depth cm ST storage of unmelted snow on 30 September cm water ACO difference of carry overs by recession flow from the previous year and to the next year cm c runoff coefficient expressing losses SRM automatically computes the storage of unmelted snow The carry over must be separately computed by the recession formula Equation 7 The procedure is demonstrated on the basin Illecillewaet in British Columbia Canada Glacier areas estimated from the periodic satellite monitoring are listed in Table 8 Table 8 Elevation zones and glacier areas in the Illecillewaet basin Elevation Elevation Range Glacier Area Zone m a s l Figure 42 shows a runoff simulation in the Illecillewaet basin with the following preselected parameters degree day factor a 0 2 to 0 6 cm C d snowmelt runoff coefficient cs 0 6 to 0 9 rain runoff coefficient cp 0 6 to 0 9 temperature lapse rate y 0 65
132. em von Schneeschmelz Abflussmengen The Snowmelt Runoff Model ETH as a universal simulation and forecast system for snowmelt runoff in German Remote Sensing Series 27 Remote Sensing Laboratories RSL University of Zurich Zurich Switzerland Carlson T N amp Bunce J A 1991 The effect of atmospheric carbon dioxide doubling on transpiration In Proc American Meteorological Society Special Session on Hydrometeorology Salt Lake City Utah USA 196 199 Cronshey R G R T Roberts amp Miller N 1985 Urban Hydrology for Small Watersheds TR 55 Revised Proc ASCE Hydraulic Division Specialty Conference Orlando Florida USA Gifford R M 1988 Direct effects of CO concentrations on vegetation In Greenhouse Planning for Climate Change ed by G L Pearlman CSIRO Melbourne Australia 506 519 Gyalistras D H von Storch A Fischlin M Beniston 1994 Linking GCM Simulated Climatic Changes to Ecosystem Models Case Studies of Statistical Downscaling in the Alps Climate Research Vol 4 pp 167 189 119 Higuchi K Ageta Y Yasunari T Inoue J 1982 Characteristics of precipitation during the monsoon season in high mountain areas of the Nepal Himalaya In Hydrological Aspects of Alpine and High Mountain Areas Proc Exeter Symposium IAHS Publ no 138 21 30 Hall D K amp Martinec J 1985 Remote Sensing of Ice and Snow Chapman amp Hall Ltd London New York IBM VS FORTRAN Applicat
133. er must decide on which date and the computer program should be switched to option 1 Now if rain falls on this snow cover it is assumed that the same amount of water is released from the snowpack so that rain from the entire zone area is added to snowmelt The melting effect of rain is neglected because the additional heat supplied by the liquid precipitation is considered to be small Wilson 1941 5 3 6 Recession coefficient k As is evident from Equation 1 the recession coefficient is an important feature of SRM since 1 k is the proportion of the daily meltwater production which immediately appears in the runoff Analysis of historical discharge data is usually a good way to determine k Figure 10 shows such an evaluation for the alpine basin Dischma 43 3 km 1668 3146 m a s 1 Values of Q and Qm are plotted against each other and the lower envelope line of all points 1s considered to indicate the k values Based on the relation k Qp 1 Qn it can be derived that for example k 0 677 for Qa 14 mie and ky 0 85 for Q 1 m s This means that k 1s not constant but increases with the decreasing Q according to the equation Ko A Q 7 where the constants x and y must be determined for a given basin by solving the equations k SS xQ k x Q log k log x y log Q 8 log k log x y log Qo 9 In the given example log 0 677 log x y log 14 log 0 85 log x y log 1 x 0 85 y 0 08
134. ervening weather conditions In spite of inherent approximations this method may prove to be more realistic than mere speculations about the fate of glaciers As a practical exploit for the management of water resources in particular hydropower generation the increase of runoff from glacierized basins in the coming decades can be evaluated 83 10 SRM FOR WINDOWS WINSRM COMPUTER PROGRAM 10 1 Program overview WinSRM provides the user with a complete modeling environment in which snowmelt runoff is simulated for mountain basins where snowmelt provides a major contribution to that runoff The model manages a database of physical data for a given basin and any number of individual simulation data sets stored as unique tables in the database Each simulation is an independent entity operating on a 2 366 day subset of the database s physical database Simulation results are available to the user via a series of tabular reports and an extensive library of graphical plots The model environment includes a unique ability to design complex scenarios that describe a future changed climate in a basin and then simulate the effect of such a scenario on runoff for that basin 10 1 1 Historical background The Martinec Rango Snowmelt Runoff Model SRM was originally a FORTRAN model designed to operate on an IBM 370 series mainframe computer The first computerized version of the model was developed by Martinec et al 1983 In 1986 the m
135. esent Climate Step 2 Winter Changed Climate Step 3 Summer Present Climate Step 4 Summer Changed Climate et present climate Step 5 Derive CDCclimiata Step 1 has completed The statistical results are displayed above Click Step 2 Step 6 Summer Changed Climate to continue the climate change run C Program Files inSAM SAG rand mdb 12 3 2002 2 45 PM a Volume Difference 4 6 1669 oefficient of Determination A 1 3886 Window 20 Run a Climate Change 103 Selection The first step in modeling climate change is to identify the climate change scenario that will control the process A climate change scenario See Climate Change Scenario Definition Window Appendix D ClimateScenario table is a database table containing a set of rules for altering the original variables and or parameters used with a simulation to approximate a changed climate Included in the scenario is a WinterEndDate which separates the simulation period into winter and a summer half years The scenario is identified by selecting an entry from the Existing Scenario Definition s dropdown combo box located in the upper left corner of the window This combo box contains a list of all the climate change scenarios that have been defined for use with the current simulation An alternative to be discussed later in this document is to re use the climate changes from an earlier climate change run This is a
136. etric altitudes of the respective zones by an altitude gradient for example 3 or 4 per 100 m If two stations at different altitudes are available it is possible to assign the averaged data to the average elevation of both stations and to extrapolate by an altitude gradient from this reference level to the elevation zones It should be noted that the increase of precipitation amounts with altitude does not continue indefinitely but stops at a certain altitude especially in very high elevation mountain ranges A critical temperature see Section 5 3 4 is used to decide whether a precipitation event will be treated as rain T gt Terr or as new snow T lt Tcrir When the precipitation event is determined to be snow its delayed effect on runoff is treated differently depending on whether it falls over the snow covered or snow free portion of the basin The new snow that falls over the previously snow covered area is assumed to become part of the seasonal snowpack and its effect is included in the normal depletion curve of the snow coverage The new snow falling over the snow free area is considered as precipitation to be added to snowmelt with this effect delayed until the next day warm enough to produce melting This precipitation is stored by SRM and then melted as soon as a sufficient number of degree days has occurred The following example in Table 2 illustrates a case where 2 20 cm water equivalent of snow fell on day n and then was melt
137. f record Parameters Current simulation Units of A Latitude dec 23 Results Window 13 Display the Edit Basin Variables Dialog 97 10 3 4 2 Purpose The Edit Basin Variables Window Window 14 is used for detailed editing of basin variables The same window with appropriate changes to its title and data grid contents 1s used for displaying editing simulation parameters and simulation results im Edit basin variables E i 0 x Scroll bar Top of grid Basin Rio Grande Basin Simulation RGW r73 Prior page Prior line Focus rectangle 0 000 0 000 5 320 0 000 0 000 0 000 Target cells for the edit action Es E E Right click for the A Ch A o A K Copy Chic Click the arrows 1 4 ___ to change 1 4 t Ctrl V SE 4 790 0 000 0 000 10590 print KE L edit context menu Edit multiple days Switch Scroll Style Scroll bar Ta Next line 0 000 0 000 0 000 0 000 0 000 Next page Bottom of grid i P Frint Help Cancel Done C Program Ples AnS RAMA RloGrand mdb 12 12 2002 10 19 AM E Window 14 Edit Basin Variables Dialog The basic structure of the window is the data grid which displays information related to a given elevation zone Each line of the grid represents a single day for that elevation zone with each cell in the line containing the variable or parameter value for that particular day and zone 10 3 4 3
138. ff Volume 10 6 m 938 753 Average Measured Runoff m s 21 014 Computed Runoff Volume 10 6 nF 908 996 Average Computed Runoff m s 68 763 Volume Difference 3 1699 Coefficient of Determination R 0 8957 These comments have been placed in the sample samt data file in order to test the model enhancement feature allowing the user to imbed comments m a data file to document the file s evolution P Print Help Done 9 After the successful simulation you can familiarize yourself with the graphical and tabular output products available to the modeler by clicking Graphics or Reports on the main menu Example 2 Adding physical data to a database from another digital source 1 Load the database created in Example 1 or any of the other sample WinSRM databases 2 Click Add Simulation enter a new simulation in the Name the New Simulation dialog then click Ok The Edit Simulation Control Information window appears 3 For purposes of this example the default information in this dialog will be used as presented Click to add the new simulation to the database Two dialogs appear as shown below The parameter table For this simulation will be created using the initial parameter table values shown op this window Once the parameter table For a simulation has been created it may be edited by clicking Data Parameters Click Ok 130 Click to add new records for the specified date period
139. ficient for snow cs was taken into account The degree day factor gradually increases in line with snow density while cs reflects the decline of the snow coverage and the stage of vegetation growth Since the original CDC s are moved by about one month earlier see Figure 32 the values of both parameters were shifted accordingly by 31 days in the climate run For example a 0 45 cm C d selected for May in the present climate was used in April in the warmer climate The climate program provides for automatic shifting by any number of days When September values are shifted to August the value of 30 September is repeated in September The shifting is stopped in January in order to prevent winter conditions being transferred to the autumn There is no consensus yet whether a warmer climate will increase losses in which case the values of cs and cr would be generally decreased because a decreased evapotranspiration due to the CO increase might offset the temperature effect Carlson amp Bunce 1991 Gifford 1988 Selected parameters can be shifted or changed in accordance with the expected conditions of a future climate Future versions will also enable the constants x y for the recession formula to be adjusted if for example a steeper recession would be indicated by drier soil conditions 70 8 5 Normalization of data to represent the present climate The climate effect is evaluated by comparing present conditions with con
140. for winter surpluses increased P overwhelms temperature increase for the zone stretch the MDC Compute the proportional increase in winter ending snow water equivalent using values obtained in Steps 1 2 above 2 Pam amp Zonal Mell orm 29 2P 2 Zonal Melt Multiply each x value of MDCgxcr by this factor to stretch MDCrxcr creating MDCexcr wa 5d Derive a new curve MG cm wa by adding to MDCexcL wa the corresponding daily melt depths of new snow surviving in the warmer climate 2 newmelt c im from Step 4 Se For each daily value n in MDCexcy wa find the first day on 2 a DeLim from Step 4 when the value is equaled or exceeded Move the corresponding day s snow cover to day n of the CDCcim wa array On days when multiple S values for the same day in CDCerm result the first highest value is used the remaining values are ignored Sf After all derived values have been calculated for CDCerm wa SRM completes the new curve CDCcum wa ma by supplying all missing daily snow cover values using the following logic e Missing days that lie between two days with derived values will inherit the earlier day s derived S value e All missing days following the last derived daily value will inherit a residual value that 1s determined in two different ways depending upon the difference between the last derived daily value and the corresponding value in CDCyormat The normal difference will be negative CDCcum lt CDCy
141. forecast for the Rio Grande basin in which the forecasted temperatures were replaced by seasonal average temperatures the precipitation was 110 of the average precipitation randomly distributed over each month and the snow covered area was forecasted by using temperatures and the appropriate modified depletion curve Evidently the seasonal average temperatures do not show the cold spell in the second half of May 1987 and therefore the runoff decline is not simulated The difference between the runoff simulation and short term forecast is illustrated by Figure 23 The temperature T precipitation P and snow covered area S are used to compute Q with L 18h At the time of simulation these values are known When Qn 1s forecasted in the morning of the day n T and P are not yet known and forecasted values must be used In order to forecast further ahead Qy 2 Qn 3 the forecasted values Tan and Ban Trio and P are used The snow covered areas Dua Sn 1 Sn are extrapolated by using temperature forecasts and the modified depletion curve MDC 54 SHOW COVERED AREA D 200 CUMULATIVE SNOWMELT DEPTH cm CORRECTED FOR NEW SNOW Fig 21 Nomograph of modified depletion curves for the elevation zone B 1284 km 2926 3353 m a s l of the Rio Grande basin near Del Norte Colorado The curves are labeled with the areal average water equivalent 1 April of the snow cover which they represent Rango amp van Katwijk 19
142. g default parameter values specified on the window _Delete Simulation Click this button to delete the current simulation as denoted on the main window The simulation is highlighted and an are you sure prompt is issued before the desired action is completed Edit Simulation SEN Click this button to display and edit the current simulation in the Edit Simulation Control Information window Be forewarned that editing parameter value s for an existing simulation here will over write ALL values for ALL zones with the default value specified 94 Ka When the QuickHelp for Main Window button is clicked context sensitive help is displayed for the screen elements and data entry structures within the main window that are subsequently clicked The help data is displayed near the bottom of the window While in help mode a question mark 7 is displayed next to the cursor No data can be entered and menu item and button functionality is suspended Clicking the quickhelp button a second time will disable quickhelp return to the standard cursor and enable data entry and other functionality 10 3 2 4 Basin definition frame The basin definition frame Window 2 is a container for the physical characteristics that name describe and geo locate the basin This information is simulation neutral that is it applies for any simulation defined for the database Two values NEVER change after their initial definition Those values are U
143. ge data however are normally published for midnight to midnight intervals and need adjustments in order to be compared with the simulated values Conversely the simulated values can be adjusted Shafer et al 1981 to refer to the midnight to midnight periods Figure 12 illustrates the procedure for different time lags For L 6 hours 50 of input computed for temperature and precipitation on the nth day Ai plus 50 of Lk results in the n 1 day s runoff after being processed by the SRM computer program L 6h 0 50 Ia 0 50 In gt Dias 17 Similarly L 12h 0 75 1 0 23 Lu gt Qu 18 L 18h hkh gt OU 19 L 24h 0 25 1 0 75 Lu gt EK 20 This procedure is preferable at least in mountain basins smaller than 5000 km to evaluations of L by calculating the velocity of overland flow and channel flow It has been shown by environmental isotope tracer studies Martinec 1985 that overland flow 1s not a major part of the snowmelt runoff as previously believed There is increasing evidence that a major part of meltwater infiltrates and quickly stimulates a corresponding outflow from the groundwater reservoir With this runoff concept in mind the seemingly oversimplified treatment of the time lag in the SRM model is better understood Fig 12 Snowmelt hydrographs illustrating the conversion of computed runoff amounts for 24 hour periods to calendar day periods The various time lags bold lines are taken into accoun
144. he Variable Parameter Input Data Screen may be generated externally and directly loaded into SRM To be accessible to SRM all import data files must be qualified DAT Once a FileIO option is selected Micro SRM displays a file selection definition screen made up of two display panels The upper panel displays the individual system s available devices and lists all subdirectories and filenames for the current default device directory The lower panel displays the current pathname 1 e the device directory or filename that will be used to attempt to fulfill the current FilelO operation The selection screen uses two simple approaches for identifying target file names The first approach is point and shoot With the upper panel active an object from that panel may be selected by moving the light bar to the desired object using the cursor keys then pressing Enter lt 4 or by clicking on the desired object with the left mouse button Alternatively the user can press the TAB gt key to switch to the lower panel and use the keyboard to enter the desired filename then press Enter lt 4 to select the file If the object selected entered is a valid pathname 1 e a device A B directory name ending with TV or lt Parent gt the default device directory will be changed and all appropriate files on the new path will be displayed F5 Plot The PLOT Function is invoked by pressing clicking function key FS from the Program Option
145. he model has been applied by various agencies institutes and universities in over 100 basins situated in 29 different countries as listed in Table 1 More than 80 of these applications have been performed by independent users as is evident from 80 references to pertinent publications Some of the localities are shown in Figure 1 SRM also successfully underwent tests by the World Meteorological Organization with regard to runoff simulations WMO 1986 and to partially simulated conditions of real time runoff forecasts WMO 1992 13 3 06 N 2z sobuer MFLS ogo PUES UY 3 04 N ote EMEYD WayYsY J Z9 NJE Yzpuedg 3 2SFL S Co My 14995 AA ELL NZ wa MA bZL N E AE Ewe ddn 3 86 N Ep 5uBuos AMF GEN SE MOA M 589 5 8 E9 spurg M 8S N Ze WABroqy a Soe S rr Bee J BEL NG Jeary IES SANAAT Ta oe J F H LSE seag 3 81 NE PE AH Mea 3 561 N 26r 2feung 3 55 NOS INOW d LINE ISMA StL OL N LbS PEGA 3 5011 NEIE aye quie 7 3 96 N or Binqsi a 3 29 HN CR gt ouring ME MEG PUEGODS aysay 3 2 0 5 Zr seeusiAg jem obeT SoU PUCoS awe wj addy d ZS ZZ OD OC wed gt M 28 NE EM M L Sep Un y oben M 04 5 22 eBundny Mc LOL NEIE SPUEI on AW SLL N WL Or Wad ysjueds AL SELL N S 191405 M S6LL NIE Jann sBury M 60L Ner Apoomug AF LLL N S rr Wo 6 AUSH AW Lal NFC 95 08 AA LALL N LS PEAD j et DO Du OD JL a Fig 1 Selected locations where SRM has been tested 14 3 RANGE OF CONDITIONS FOR MODEL
146. her Tmax Tmin or Tavg are required 143 Simulation Parameter Table Unlike the tables previously defined the following table is replicated for each simulation in the database see the SimulationIndex table and the table s name is identical to the simulation s name Data SC Field Name Description Field Name Type Zone Integer Zone identifier 1 10 Month Day Year these parameters represent mm dd yyyy Critical temperature See Section 5 3 4 determines whether precipitation Single is treated as rain or snow by the model Tcrit can be provided for each zone or aS a basinwide parameter See SimulationIndex table C F Lag time see Section 5 3 2 is the time difference between the start of Single increasing temperatures and the corresponding increase in runoff from ERR the basin This is always a basinwide parameter and is entered in zone l If a single set of temperature data are provided at the reference elevation the model uses lapse rates see Section 5 3 3 to adjust the LapseRate Single values for the effects of elevation to each zone s hypsometric mean elevation C 100m or F 1000 ft Also see Simulation ndex table description The degree day factor see Section 5 3 2 converts the number of Single degree days in daily snowmelt depth An must be provided for each day zone See Section 5 3 2 The snow runoff coefficient see Section 5 3 1 takes care of losses due Single to s
147. here Rango amp Martinec 1994 In view of the stepwise character of the cumulative snowmelt depths a slightly higher snow water equivalent than the winter deficit may be cut off to derive MDCexcr wa see Step 9 above which would ultimately accelerate the decline of CDC cum wa On the other hand searching for cumulative snowmelt depths equaled or exceeded in deriving CDCeLim wa may result in a very slight delay of the decline 12 CDCcrm wa 18 used to compute the climate affected runoff in the summer half year It should be noted that in contrast to the model runs in the simulation mode Ri in the climate change runs does not indicate the model accuracy but results from the difference between the hydrographs computed for the present and changed climate 100 75 5 LA 50 APR MAY JUN JUL Fig 28 Conventional depletion curves of the snow coverage from Landsat data in the elevation zones A B and C of the Rio Grande basin near Del Norte Colorado in 1979 5 50 A 2 D e Aa M cm Fig 29 Modified depletion curves for zone A MDCner derived from CDC therefore including new snow MDCexc with new snowmelt excluded MDCexcr wa with winter deficit shaded area cut off 66 100 90 80 ZU 60 LMS 50 cm Ap 30 20 10 fy Apr May Jun Jul Aug Sep Fig 30 Accumulated zonal melt curves for zone A AZMixcL computed daily melt depth multiplied by S from CDC zonal melt
148. hour perioids to calendar day periods Martinec amp Rango 1986 oocccocccccnococnnnncnncnnonorornnnororonnnnos 40 Figure 13 Runoff simulations in the basin of the Rio Grande near Del Norte Colorado 3419 km 2432 4215 mast Matinee Rango 1989 andara 43 Figure 14 Combined representation of model performances for average inaccuracies using three criteria Ri DG and D Rango and Martinec 19981 45 Figure 15 Combined representation of model performances for maximum inaccuracies using three criteria Ri DG and D Rango and Martinec 10991 45 Figure 16 Runoff simulations in the basin Dischma 43 3 km 1668 3146 m a s l Martinec Rango Eege 46 Figure 17 Runoff simulation in the catchment area of the hydroelectric station Sedrun Swiss Alps 108 km 1840 3210 m a s l Baumann et al 1990 ii ENEE 50 Figure 18 Runoff simulation in the catchment area of the hydroelectric station Tavanasa Swiss Alps 215 km 1277 3210 m a s l Baumann et al 1990 iio 51 Figure 19 Extrapolation of snow cover depletion curves in real time from modified depletion curves with the use of temperature forecasts Hall amp Martinec 1985 oooocccccnconocococococornnnanonononororornannss 52 Figure 20 Elimination of the effect of antecedent snowfall on the extrapolation of depletion curves of snow coverage Hall Martinec 1985 coococccoccccnnococnnnononocnnnonoroncnnonoronnnnoronnnnnneronnnnnnernnns 52 Figure 21 Nomogr
149. hs in winter in each zone Figs 25 26 The snow in the B zone was completely melted for T C on 1 April so that no winter deficit resulted for T 4 However in reality some snow survived for T C due to variable snow depths as seen by satellites S 27 on 1 April With no computed snow for T C as well as for T 4 no winter deficit resulted and therefore the existing snow cover was falsely preserved for T 4 A manual correction elimination of CDCcum was necessary In the C zone there was enough snow to avoid the error from uniform instead of variable snow depths in winter so that the snow cover duly disappeared as a result of the evaluated winter deficit This error can occur only with very little snow in a zone on 1 April and can be readily corrected Whenever satellites see some snow on 1 April and the winter accumulation of snow from precipitation and snowmelt is zero eliminate CDCc iy The recession of the natural runoff slows down with a decreasing discharge To this effect y must be gt 0 Ina log log representation Fig 10 the points outside the range between the 1 1 line and a parallel line for points in the range of high flows below the dashed line in Fig 10 must be disregarded With this measure the problem of the runoff simulation was eliminated 50 7 OPERATION OF THE MODEL FOR REAL TIME FORECASTS In order to be applied for real time discharge forecasts a model should be able to simulate the runoff not
150. ia the Internet by accessing the SRM home page at http www ars usda gov Services docs htm docid 8872 This site also includes an electronic version of the SRM User s Manual A distribution diskette available on either 5 or 3 media containing the executable code supporting files and example data files can be provided upon request To obtain the latest version of Micro SRM contact USDA ARS Hydrology and Remote Sensing Laboratory Bldg 007 Room 104 BARC W 10300 Baltimore Avenue Beltsville MD 20705 2350 USA 301 504 7490 or via electronic mail to ralph roberts ars usda gov alrango nmsu edu Micro SRM has been distributed to several hundred individuals and institutions worldwide since 1985 We appreciate the feedback we receive from users of Micro SRM both positive and negative and view it as a valuable resource in the process of debugging and improving the software 176 177 New Mexico State University is an equal opportunity affirmative action employer and educator NMSU and the U S Department of Agriculture cooperating
151. id on the right side of the window At this point the user can click on the main window then view the simulation s results using Graphics and or Reports The user can also examine the date in SimResults by clicking Data Results data editing is disabled during this process Notice that several of the menu choices on the main window are disabled not available for use while the climate change process 1s active 7 Click the remaining Step buttons as they become active After steps 2 3 and 6 you may repeat the post completion directions detailed above in paragraph 6 8 When is clicked the Winter Change for Zone dialog appears one for each zone The value shown in the dialog is the calculated melt gain loss realized during the winter half year resulting from the climate change Click OK to accept these values Winter Change for Zone 1 KE E Click UK to accept the computed Change or type a Ok new value Cancel 9 The climate change simulation is completed once the six step buttons have been clicked Click the Print button to print the statistics displayed in the dialog to the system printer 10 Click Graphics or Reports to examine the plots and reports respectively available after a successful climate change simulation Example 5 Developing a custom climate change scenario This example demonstrates a method for modeling an extremely complex or unusual climate change one that the Climate Change
152. identifying the specific iterative version of the data displayed G 5 3 SRM data file As described before the FilelO menu option Save creates a disk file containing all the information necessary to recreate the model run at some future time via the Load function G 5 4 Plot displays Micro SRM produces several different plots in one of several different screen modes Once plotted and assuming a compatible printer a hardcopy of the plot image can be produced Measured vs Computed Streamflow Runoff is plotted along the Y axis instantaneous flow time along the X axis Run statistics are displayed in the upper right corner 168 Snow Depletion vs Time Snow depletion curves for each zone are produced The initial image displays all the basin s snow depletion curves on a single image Press a key to produce a separate plot of each specific zone in ascending order Press ESCape to terminate the plot Snow Depletion vs Total Snowmelt Depth Daily snow covered area is plotted on the Y axis corresponding accumulated snowmelt depth and accumulated snowmelt reduced by the effect of new snow are plotted on the X axis Each zone is plotted separately Press Enter lt to see the next zone ESCape to terminate and return to the Plot Menu Accumulated Zone Melt Depth vs Time Accumulated zone melt depth areal coverage y is plotted against time x Press Enter lt 4 to change zones ESCape to terminate the plot PL
153. ield allows the user to enter the number of days of shift desired Shifts are currently restricted to model parameters only When shifting parameters the model also modifies the shift in two critical ways in order to impose reality on a mechanistic approach see Section 8 4 1 To insure that parameters for a cooler autumn month s are not shifted into the final month s of the melt season for example October snow runoff coefficients shifting to September on a defined shift of 31 days SRM shifts parameters only within a hydrologic year repeating the final daily value for the year the required number of days to complete the shift With this approach a shift of 31 days for a summer half year 4 1 to 9 30 would have September values shifted to August and all September days would take on September 30 s daily value 158 Snowmelt Runoff Model SRM Version 4 0 Climate Change Control Screen Winter Summer 1001 0331 0401 0930 cr KA SALELE Amount Sit Amount Temperature T N A Precipitation P Lapse Rates y Snow Runoff Coefficients Es Rain Runoff Coefficients Cp Degree Day Factors a Critical Temperature era EscAbort FlHelp Climate Change Control Data Entry Screen 2 Winter parameter shifts operate only on the last coldest half of the winter season This precludes the possibility of having the desired warmer winter climate negated by moving parameters associated with colder winter
154. in MDCexcy wa find the first day on 2 a Deim from Step 4 when the value is equaled or exceeded Move the corresponding day s snow cover to day n of the C DC om wa array On days when multiple S values for the same day in CDC 1im result the first highest value is used the remaining values are ignored Sf After all derived values have been calculated for CDCcLim wa SRM completes the new curve CDCcim wa ma by supplying all missing daily snow cover values using the following logic e Missing days that lie between two days with derived values will inherit the earlier day s derived S value e All missing days following the last derived daily value will inherit a residual value that is determined in two different ways depending upon the difference between the last derived daily value and the corresponding value in CDCyormat The normal difference will be negative CDCcum lt CDCnormat In this situation the residual value is set equal to the last S value in the CDCuormaL This value will typically be zero except in cases where the zone includes a glacier or permanent snow cover In the infrequent cases where the difference is positive CDCcum gt CDCyormat each missing daily value after the last derived day will take on the corresponding day s S value from CDCyormat plus the difference 5g Plot MDCrxcL MDCexcr wa Sh Plot AZMynct AZMexcr AZMexct wa 5i Optionally plot MDCexcr wa MDCcum wa 5j Plot CD
155. in October December If S 1 on 1 April from satellite data assume S 1 in March as well If it is less than 1 on 1 April the snow coverage in March is put to O or which may be more accurate it is interpolated from 1 on 1 March to the S value on 1 April Naturally S 1 can be assumed for a longer part of the winter in higher elevation zones while in the lowest zone option 1 may be preferable The present program keeps the snow coverage estimated for the present climate unchanged for climate runs It is therefore recommended to assume a complete snow coverage only in months in which it 1s expected to last in a warmer climate as well Whenever S 1 is introduced the SRM program cancels any existing storage of preceding temporary snowfalls because such snow becomes a part of the seasonal snow cover Snow storage is also automatically canceled on the 1 April date because all existing snow is then accounted for by the depletion curve of the seasonal snow coverage With the variables thus accounted for SRM is run as usual Because the estimated snow coverage is less accurate than measured CDC in the summer the water equivalent of new snow is reduced at once by the simultaneous 1 S to prevent an inadequate S from influencing the computation on a later melt day This deviates from the summer procedure explained in Section 5 2 2 Table 2 The model parameters should be adapted to winter conditions In particular the constants x y for the rece
156. in the data entry process is displayed upon user request by pressing Function Key 1 Once in help mode the user may PgUp to a prior help screen PgDn to the next help screen or ESCape to return to the screen from which help was initiated G 3 3 Menu screens Micro SRM uses the menu screen to provide a simple method for controlling processes that require a single selection from among an array of possible choices The menu screen presents the user with a list of possible selections along with their associated letter code The user indicates a menu selection by entering its appropriate letter code clicking on the selection with left mouse button or by moving the light bar the menu selection displayed in contrasting color or reverse video up down the menu list using the arrow up down TV keys The user activates the selection by pressing Enter lt There are three full screen menus used within SRM The Plot and Print Menus control selection of available plot and print products respectively The Basin Variable Parameter Menu controls which of the 13 daily basin variable or parameter data input screens is to be displayed for data entry In addition to these three full screen menus SRM uses several pop up menus A pop up is a display that is temporarily superimposed over some existing screen An example is the FilelO function s pop up menu that controls selection of the model s input output options lt i L gt Snowmelt
157. ine Attributes Window 10 3 11 3 Buttons Done Set 111 Close the Line Attributes window Redraw plot in Graphics window using the new attribute Cancel Close the window return to the Graphics window Ignore any changes If Close the window return to the Graphics window Ignore any changes Toggle quickhelp comments related to the window elements fields 10 3 12 Output definition window 10 3 12 1 Method of access The Output Definition window is viewed by clicking Reports on the Window menu bar Window 9 10 3 12 2 Purpose The Output Definition Window Window 29 is the tool provided by WinSRM to manage the presentation of simulation parameter data basin variable data and simulation results in tabular formats The report choices on the window are dependent upon the current state of model operation For example reports that include calculated results are not available until a successful simulation has completed Reports are unavailable when they appear in the window as disabled choices light gray in color and unresponsive to mouse clicks SS Output Definition Output Definition Available Reports This report will be sent to a Temperature Values printer as determined by IT Degree Day Factors Runoff Coefficient the Print To radio button Lapse Rate Critical Temperature Rain Con Area Lag Time IT Zone Degree Days Observed Precipitatio dai Covered Area 5 Melt Depth M 5 Melt Ne
158. ing climate change research that SRM model developers had conducted over the prior several years However we have recognized the limitation of the original climate change algorithm the fact that the change was isolated only on the melting season In other words the snowpack used during any climate change scenario was the observed real seasonal snow cover Version 4 0 incorporates much of the model developers latest research on the effect of climate change on snowmelt runoff over an entire hydrological year see Section 8 The original SRM 3 11 program s climate change algorithms processing steps have been replaced with new algorithms that model future snow conditions as well as climate affected runoff The computer program has been enhanced to manage the steps involved in simulating climate change by providing a new data entry screen the Climate Change Control Screen This screen allows the user to easily describe climate change conditions that span an entire hydrological year specifying both a winter and a summer season climate change scenario By simulating the effect of winter climate change on the existing snowpack the program can model the summer season s climate affected snow conditions and use those curves CDC w during the melt season climate change simulation In Version 4 0 a climate change run is in reality five complete iteractions through the model along with a series of additional steps required to model clima
159. ining a variety of control information the model needs to initialize the physical and parameter data sets prior to simulation or climate change processing See Appendix D SimulationIndex Table for a description of these control columns The current simulation name is displayed below the basin simulations grid table in bold blue 95 10 3 2 6 Status bar The path and filename of the current basin database is displayed on the left side of the status bar at the bottom of the window Window 2 If no database is open the text File appears in this location The remainder of the status bar contains fields for the current date and time The same status bar appears on several other larger size data entry windows 10 3 3 Edit simulation control information window ia Edit Simulation Control Information E lol x m Simulation Name Fun Number Snowmelt Period Start Date 18 9380 ddddd la End Date 09 30 1981 Units p Temperature gt p Initial Runoff m s Rainfall Threshold cm SS Ce Daily Average Wes C Max Min Values en AL D EE Input by Zone x yes Comments Temperature Comments concerning the simulation go here Precipitation Je Temperature Lapse Rates IT Runoff Cocfficients Critical Temperature These data boxes are IV Recession coefficients EE changed to display Degree Day Factor Jo 650 Runoff Cocfficients Snow Jo 700 fields when an existi
160. introduced by a subroutine 10000 E conversion from cm km7d to m el 36400 T S and P are variables to be measured or determined each day cr Cs lapse rate to determine AT Terr k and the lag time are parameters which are characteristic for a given basin or more generally for a given climate A guidance for determining these parameters will be given in Section 5 3 If the elevation range of the basin exceeds 500 m it is recommended that the basin be subdivided into elevation zones of about 500 m each For an elevation range of 1500 m and three elevation zones A B and C the model equation becomes A 1000 86400 1000 A CS T SE AT pn S pn SS C RBn Pon levers 2 A 1000 it T J AT o So ole C RCn Pon P 1 E Kes Ok Oz La T AT an JS an TC hant An The indices A B and C refer to the respective elevation zones and a time lag of 18 hours is assumed Other time lags can be selected and automatically taken into account as explained in the Section 5 3 7 In the simulation mode SRM can function without updating The discharge data serve only to evaluate the accuracy of simulation In ungauged basins the simulation is started with a discharge estimated by analogy to a nearby gauged basin In the forecasting mode the model provides an option for updating by the actual discharge every 1 9 days Equations 1 and 2 are written for the metric system but an option for model operation in Engli
161. ion After Step completes click on to terminate the simulation Control returns to the main window Now we will begin to customize the SimResults table with a changed climate that would be impossible to accomplish with a climate change scenario by cutting and pasting the climate variables temperature and precipitation from 1977 to 1979 s new climate 10 11 12 Click on RGWY77 in the Basin Simulations grid Click on Data Variables Current Simulation The Edit basin variables dialog appears Click on the column header Tavg The entire column should be highlighted Right click on this column then select Copy or press Ctrl C to copy the year of temperature data for zone A to the Windows clipboard Click to close the dialog and return to the main window Click on RGWY79 in the Basin Simulations grid Click on Data Results The Simulation results editing dialog appears Notice that a number of the column headers end with CC These columns contain the post climate change values for that variable parameter If a column header is only partially visible you can resize the column by placing the mouse on a column border in the header When the mouse changes state press the left mouse button and drag to complete the resize Click on the column header TavgCC to select the entire column The column will be highlighted Right click the column and select Paste from the popup menu or press Ctrl V The data c
162. ion The window displays several important statistics that quantify the accuracy of the calculations Also displayed for reference purposes are the current date the source of the simulation data the database name and the simulation table name and the user provided comments that describe and document the simulation The window and any other can be captured to the Windows Clipboard by clicking on the window title bar to give it focus then pressing Alt PrintScrn keyboard keys The window s title bar caption will identify the processing choice selected from the Run menu see above iw Simulation Statistics i ioj x Basin Name Rio Grande Basin Simulation ARGWYy 79 Run Date 12 9 2007 Measured Runoff Volume 10 6 m 1208 960 Average Measured Hunoff ms 30 336 Computed Runoff Volume II Gei 1212018 Average Computed Runoff Im Zelt 38 433 Volume Difference 0 2530 Coefficient of Determination A 0 9556 Fie AGW ZO Winter climate change T 4 a amp Cs shifted 31 davsearlier De April Cs becomes March Cs etc Summer climate change T 4 a Cs shifted 31 days vd Print Help Done i i Cl indow return t Window 18 Simulation Statistics Window 102 Once in the clipboard the window s image can be pasted into another application that can access the Windows clipboard via a paste command The information shown on the window is available as part of the WinSRM Measured vs Computed Snowmelt Runoff
163. ion Programming Language Reference GC26 3986 3 March 1983 111 112 Jaccard C 1982 Recession coefficient formula Personal communication Klemes V 1985 Sensitivity of water resources systems to climate variations WCP Report 98 WMO Geneva Switzerland Kotlyakov V M amp Krenke A N 1982 Investigations of the hydrological conditions of alpine regions by glaciological methods In Hydrological Aspects of Aloine and High Mountain Areas Proc Exeter Symposium AHS Publ no 138 31 42 Martinec J 1960 The degree day factor for snowmelt runoff forecasting UGG General Assembly of Helsinki AHS Commission of Surface Waters AHS Publ no 51 468 477 Martinec J 1970 Study of snowmelt runoff process in two representative watersheds with different elevation range AHS Unesco Symp Wellington New Zealand IAHS Publ no 96 29 39 Martinec J 1975 Snowmelt Runoff Model for stream flow forecasts Nordic Hydrol 6 3 145 154 Martinec J 1980 Snowmelt Runoff Forecasts based on Automatic Temperature Measurements AHS Symposium in Oxford 1980 AHS Publ No 129 pp 239 246 Martinec J 1985 Time in hydrology In Facets of Hydrology ed by J C Rodda vol HIE John Wiley amp Sons London 249 290 Martinec J 1989 Hour to hour snowmelt rates and lysimeter outflow during an entire ablation period In Snow Cover and Glacier Variations Proc Baltimore Symp AHS Publ no 183 19 28
164. ions that utilize the edited data Changes to parameter values are local Each simulation has its own set of parameter values so changes to these data only affect that simulation 10 4 3 Running WinSRM After a current simulation has been identified by clicking on a row in the Basin Simulations data grid a simulation is initiated by clicking the main Window menu bar Run Simulation Melt Season or Run Simulation Year round Window 6 As mentioned earlier in the section describing the WinSRM main menu the distinction between these two alternatives lies in the package of output graphics available upon completion of the simulation See Appendix B for more information about the graphics alternatives 10 4 3 1 WinSRM processing steps WinSRM assembles the basin variables for the required simulation period with the simulation parameter data into a temporary database table named SimResults The model uses the data in SimResults to drive the simulation and stores many of its temporary and final calculations in the same table All model tabular and graphic output is generated using the information found in SimResults see Appendix D The model Develops temperature profile at zone hypsometric mean elevation s Determines the effect of daily precipitation upon basin runoff Determines basin runoff derived from melt of the existing snow pack Derives total basin runoff by combining snowmelt runoff with net daily precipitation contributing
165. ith drop down plot menus as portrayed in the following picture See previous sections for a description of the buttons on the Graphics window button bar shown below ER A alallala 24 em melo Runa Measured vs Eomputed Plots of calculated values are available after completion of a simulation Snow covered rea 5 Average Temperature Tawg zonal Degree Days Precipitation D Critical Temperature Tecrit Lag Time Li Lapse Rate LR Degree Day Factors An Snow Runoff Coefficients 125 Rain Runoff Coefficients CR Rainfall Contributing Area RCA After completion of Potential Melk Depth G a Melt Season Snow Depletion ve Total snowmelt Depth simulation these Accumulated Zonal Melk Depth vs Time plots are available Cumulative Runoth Components Simulated Runoff Present vs Ghanged Climate These plots are Snow Covered Area Before Uber climate change available upon WC eegi vs Mitex coto completion ofa climate change Plots of variables and parameters are available for the current simulation Accumulated Zonal Mel Changed climate i l FIDE scl vs Mitel aL simulation C Program Eleswir AM FioGrand mdb E 139 Appendix C Program Windows Page WINDOW 1 WELCOME TO WINSRM WINDOW cccccecccssecccsccccscccescscusccessccecscseesescuescusscseescseescseuseseusecsauscseuscscs 87 WINDOW 2 WINS RM MAIN WINDO Wins EES 88 WINDOW 3 WINSRM MAIN WINDOW
166. itude Basins HYDALP Technical Report Mitteilung 4 Institute for Meteorology and Geophysics University of Innsbruck Austria 60 Rott H Nagler Th 1995 Monitoring temporal dynamics of snowmelt with ERS 1 SAR In T Stein editor GARSS 95 Quantitative Remote Sensing for Science and Applications Florence Italy pp 1747 1749 J uly 1995 61 Schaper J Seidel K amp Martinec J 2001 Precision snow cover and glacier mapping for runoff modelling in a high alpine basin Remote Sensing and Hydrology 2000 Proceedings of the Santa Fe Symposium April 2000 AHS Publ No 267 pp 105 111 2001 62 63 64 65 66 126 Schaper J 2000 Fernerkundungsbasierte Kartierung von Schnee und Eisflachen hochalpiner Gebiete Ph D Thesis University of Zurich Remote Sensing Laboratories RSS 2000 Referent H Haefner Gutachter K Seidel Schaper J Martinec J amp Seidel K 1999 Distributed Mapping of Snow and Glaciers for Improved Runoff Modelling Hydrological Processes Vol 13 pp 2023 2031 Seidel K Martinec J amp Baumgartner M F 2000 Modelling Runoff and Impact of Climate Change in Large Himalayan Basins ntegrated Water Resources Management for Sustainable Development Mol 2 pp 1020 1028 December 2000 National Institute of Hydrology Roorkee New Delhi India Seidel K Ehrler C Martinec J 1998 Effects of climate change on water resources and
167. l l 39 1966 1965 10 Spain Llauset dam Pyrenees 7 8 2100 3000 0 69 55 1 2 23 2001 1999 1969 11 USA W 3 Appalachians 8 42 346 695 0 81 8 8 10 l 79 1986 1978 Lainbachtal 1978 12 Germany Allpater AID 18 7 670 1800 N A NA 5 1 74 1978 1979 13 Spain pasa hbase 22 2 1460 3200 072 43 3 3 22 2002 1999 Pyrenees 14 Spain EE apor anien 36 8 1480 3000 0 71 3 7 3 2 22 2002 1995 Baserca Pyrenees 15 Switzerland Rhone Gletsch Alps 38 9 1755 3630 N A N A 1 4 49 1980 1979 e 1975 1973 16 Switzerland Dischma Alps 43 3 1668 3146 0 86 2 5 10 3 38 79 1986 1970 79 28 29 1982 1979 17 Japan Sai Japan Alps 57 300 1600 0 86 N A 3 3 30 1987 1981 18 Spain Tor en Alins Pyrenees 60 1880 3040 0 71 7 3 1 4 21 2002 1999 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 Country Spain Spain Austria United Kingdom Switzerland Austria Austria Australia USA Austria Switzerland Switzerland India Switzerland USA Italy USA India Argentina Norway Norway USA Spain Japan USA Spain Spain USA Basin Flamisell en Capdella Pyrenees Vell s en A isclo Pyrenees Rofenache Alps Feshie Cairngorms Sedrun Alps Tuxbach Schlegeis Geehi River Snowy Mtns American Fork Utah Venter Ache Landwasser Frauenkirch Alps
168. late the expected numbers of degree days for each elevation zone 5 2 2 Precipitation P The evaluation of representative areal precipitation is particularly difficult in mountain basins Also quantitative precipitation forecasts are seldom available for the forecast mode although current efforts in this field are improving this situation Fortunately snowmelt generally prevails over the rainfall component in the mountain basins However sharp runoff peaks from occasional heavy rainfalls must be given particular attention and the program includes a special treatment of such events see Section 5 3 6 Rainfall input Program options 0 basin wide 1 by zone The program accepts either a single basin wide precipitation input from one station or from a synthetic station combined from several stations that is option 0 or different precipitation inputs zone by zone option 1 If the program is switched to option 1 and only one station happens to be available for example in the zone A precipitation data entered for zone A must be copied to all other zones Otherwise no precipitation from these zones is taken into account by the program Further program options refer to the rainfall contributing area as explained in Section 5 3 5 In basins with a great elevation range the precipitation input may be underestimated if only low altitude precipitation stations are available It is recommended to extrapolate precipitation data to the mean hypsom
169. lation name must be unique in the database as that value is used to create and access the simulation s simulation parameter table described below Basin Simulations Runoff Temperature Temperature Precipitation Lapse Rate SE Critical Temp Values Distribution Distribution Distribution Distribution Distribution Simulation Starting Title Date a oaie 00 100141976 09730 1977 Average Bosinwide By Zone Bosinwide By Zone By Zone Rou y yo 10 10011972 0043041979 Average Basinwide By Zone By Zone By Zone Basinwide Rou y 6 15 102011975 09730 1976 Average Eosinwide By Zone Basinwide By Zone Basinwide normalize l 1020142078 09730 2079 Average Basinwide Basinwide By Zone Basinwide Basinwide 4 Current Simulation Name RGWY 79 q Add Simulation Delete Simulation Edit Simulation Window 34 Basin Simulations grid Simulation Parameter Table Each line in the Basin Simulations grid represents a unique simulation parameter table stored in the current database named using the value in the line s Simulation Title grid cell RGWY79 in the example depicted in Window 34 The simulation table is the repository for the model parameters used by WinSRM for that simulation Each record in the table represents a parameter set for a given zone day Each new simulation 1s created by clicking the button on the main window Once a name for the new simulation is provided Window 12 control passes to the Edit Simula
170. les described below hold the variable and parameter information for a single mountain basin BasinDescription Table The basin description table occurs once in each database A single record in this table holds the physical description of the basin that the database represents Two fields in this record zones and units are permanently fixed at file creation Field Name Data Type Description Name of the basin 1 50 characters BasinDescription Brief basin description 1 50 characters Latitude Decimal latitude Longitude Decimal longitude RefElev Elevation of met station feet meters Number of elevation zones 1 16 Units Integer English 0 Metric 1 1 required ZoneDescription Table The zone description table occurs once per database The table contains one record for each elevation zone into which the basin has been subdivided the number of zones is defined at database creation Each zone record contains three required values the zone id and the area and hypsometric mean elevation of the zone Aspect related fields are currently unused hypsometric mean elevation of the zone s NE aspect ee 1 required 141 SimulationIndex Table The simulationindex table occurs once per database The table contains one record for each simulation that has been developed and saved in the database Each record contains a unique value in the field SimName that ties the simulation to a separate simulation parameter table
171. lues are then transferred to the equidistant dashed curve in Figure 20 and the real snow covered areas are obtained 74 on the date of forecast and 50 after one week These values together with the interpolated data for the intermediate days are used for runoff computations Figure 20 shows just one snowfall but the computer program takes into account each new snowfall during the snowmelt period 52 DATE OF FORECAST M 15 cm M 15 cm CONVENTIONAL gt al e DEPL CURVE 100 10 m3MELTWATER 80 D D SNOW COVER km FORECAST PERIOD 0 20 40 60 80 100 120 APRIL MAY CUMULATIVE SNOWMELT DEPTH cm 1 APRIL Fig 19 Extrapolation of snow cover depletion curves in real time from modified depletion curves with the use of temperature forecasts Hw max 1s the water equivalent of the snow cover at the beginning of the snowmelt season Hall amp Martinec 1985 DATE OF FORECAST 100 M 10 cm NEW SNOW SR 8 S E aaa 1 MAY E 60 Ze WARM M 15 cm ds O S fen O a eege S 5 8 MAY O 40 O Z D 20 0 20 40 60 80 APRIL MAY CUMULATIVE SNOWMELT DEPTH cm Fig 20 Elimination of the effect of antecedent snowfall on the extrapolation of depletion curves of snow coverage Hall amp Martinec 1985 53 Figure 21 shows that it is possible to derive a nomograph of modified depletion curves Rango amp van Katwijk 1990 for a given basin from the past years As noticed in Figure 19 the area below a modified
172. mal arrowhead shape to a move cross with four arrowheads shape when moved over a legend box To move a legend move the cursor over a blank area of the legend press and hold the left mouse button and drag the legend to the desired location then release the mouse button The legend s upper left corner will assume the position at the cursor when the mouse button is released For this reason we recommend that when moving a legend you position the mouse cursor over the existing upper left corner of the legend box before initiating the drag drop Also the program will not allow one legend to be dropped over itself or another legend 10 3 11 Line attributes window 10 3 11 1 Method of access The Line Attributes window is viewed by clicking Ka on the Graphics window button bar 10 3 11 2 Purpose The Line Attributes window Window 28 is the tool provided by WinSRM to customize and enhance the plots presented in the Graphics window through use of color line thickness and line style UA tributes lola Line sample displayed eg 1O x at specified attributes a Line Attributes SE dd dd d d Line Number Line Style Line Vr idth d Solid 1 4 Click to increase decrease thickness Lines of width gt 1 Click to change line numbers s zones for multi zone plots otherwise to the order the lines are drawn Select line style here only may only be solid when line width 1 Window 28 L
173. mate Save the data needed to recreate modified depletion curves MDCexcz accumulated zonal melt curves AZMexcr and the simulated runoff hydrograph Qsm Step 4 Summer Changed Climate Run a summer simulation for the changed climate Save gt al kom gt newmelt cum Step 5 Derive CDCCLIM WA using MDCCLIM adjusted for Winter Change Sa Compute AHw Winter Change 2winter zonal meltuorma winter zonal melt from Steps 1 2 above Reduce enhance effect of temperature induced deficit by adding any net change to seasonal precipitation resulting from the climate change scenario In the case of Illecillewaet 1984 zones C and D the scenario T 4 C P 1 2 results in a positive Winter Change increases in P overwhelm the winter deficit in melt In that case MDC xc is stretched rather than cut off as explained in Step 5c The Winter Change for Zone dialog Window 21 allows a user to over ride a calculated zonal Winter Change value at this point in the processing sequence See previous sections for a description of when such intervention might be advisable Winter Change for Zone 1 x Click OK to accept the computed Change or type a new value Cancel Window 21 Winter Change for Zone Dialog 5b For each zone develop zonal melt curve AZMexcr Q a T S vs Time for the normal climate Find the date along AZMgxcL Where zonal Winter Change is equaled or exceeded At this point the Climate
174. ment Cronshey er al 1985 Micro SRM consists of an integration of the mainframe SRM FORTRAN algorithms converted to Basic and a variation of TR 55 s data entry data management algorithms 10 1 2 Program description The Snowmelt Runoff Model for Windows WinSRM is the first version of SRM that has been adapted for use with the Microsoft Windows operating system Win 95 98 2000 XP Limitations imposed on existing and future research by the physical constraints of DOS operating system as well as its relative obsolescence were the primary reasons for the transition to Windows This new version of the program is compatible with pre existing DOS SRM data files The basic SRM model algorithm remains unchanged producing identical simulation results to those obtained by the DOS version of the model The graphical user interface GUI and data storage philosophy have undergone a complete redesign to take advantage of the enhanced capabilities available on the modern windows platform All of the original DOS output products graphs and reports remain available with enhancements provided that improve on line analysis and usability and off line quality Basins may now consist of up to 16 elevation zones Modern database technology is utilized to manage the data storage requirements of the model Storage is organized using the concept of a basin database Within a basin database resides a single copy of the physical variables for the basi
175. menu is displayed by pressing clicking function key F6 from the Program Options Basin Definition or Variable Parameter display screens The print products available in Micro SRM are e Temperature Values e Degree Day Factors Runoff Coefficients e Lapse Rate Critical Temperature Rain Contributing Area Lag Time e Zone Degree Days Observed Precipitation Snow Covered Area S e Melt Depth M S Melt New Snow M 1 S Contributing Rain Cpr 166 e Measured vs Computed Snowmelt Runoff e Input Summary Report Run Statistics e All the above The validity of some of these reports is contingent upon successful completion of a simulation forecast F7 Compute This function is invoked from the Program Options and Basin Definition data entry screens or the Variable Parameter display screen by pressing clicking function key F7 It initiates actual model processing using the current version of input data Processing may be suspended due to missing or invalid input variables If processing is successful a results panel will be displayed showing results of the computations SIMULATION SS Measured Volume 10 m 938 75 Average Measured Q m s 71 01 Computed Volume 10 m 910 10 Average Computed Q m s 68 85 R Goodness of Fit 0 8959 Volume Difference 3 0526 Screen display after compute G 4 5 Alternate function keys In addition to the functions described above there are several alterna
176. merical values of R DG and D published by WMO 1986 Each prism refers to a tested model The length along the x axis corresponds to the arithmetic mean of all LR values length along the y axis to the arithmetic mean of all 1 DG values and length along the z axis to the arithmetic mean of all D values as achieved in the snowmelt seasons of 10 test years Inaccurate results mean low values of R and DG thus longer dimensions along the x and y axis Volume deviations prolong the dimension along the z axis Consequently the volume of a prism 1s proportional to an overall average inaccuracy of a model In Figure 15 the dimensions of the prisms are determined by the worst results of each model for R DG and D 1 e 1 R max 1 DG max and Dimax All available data for the individual years and snowmelt seasons as listed in the WMO tables WMO 1986 are thus contained within each prism The differences between models are larger than in Figure 14 which means that some calibrating models had difficulties in the years with unusual runoff conditions but improved their average results Figure 14 by the more normal years A more detailed assessment of accuracy criteria with regard to the needs of the model user is published elsewhere Martinec amp Rango 1989 Prisms in Figures 14 and 15 labeled G refer to SRM Table 4 compares the numerical results of each model in the WMO project In Table 4 columns 1 3 refer to Figure 14 and columns
177. months into autumn months Amount Daily temperature and precipitation values are normally changed for climate by some constant amount or percentage Values entered in the amount field for a variable are applied to each daily observed value in the period during climate change runs Values in the amount fields are assumed to be numeric constants unless otherwise noted by including a character Parameters may also be modified by amount though model developers prefer to utilize shifting to adjust parameters for climate change G 3 8 2 Climate change progress screen The Climate Change Progress Screen tracks the multi step process that SRM follows when performing a climate change simulation see previous sections for a detailed explanation of the steps involved At the completion of Steps 1 4 the progress screen displays basic zonal melt totals the model will use in Step 5 to derive climate modified snow depletion curves After each of these initial steps the user can initiate step specific interim plots and reports using the function keys displayed near the bottom of the screen At the beginning of Step 5 SRM computes net zonal Winter Change and prompts the user to verify the results At this point computed zonal Winter Change may be overridden with some other value see Section 8 3 for discussion of a situation where such replacement would be appropriate 159 ICS Sh Snowmelt Runoff Model SRM Version 4 0 TL840606 SRM Climat
178. mperature increase of 4 C on runoff in the Rio Grande basin near Del Norte Colarado 3419 km 2432 4215 m a s l elevation zones A B and C in the hydrological year 1979 Figure 27 The following symbols are used CDC conventional depletion curve of snow covered area interpolated from periodical snow cover mapping MDCineL modified depletion curve of snow covered area with new snow included This curve is derived from CDC by relating the snow coverage to the accumulated computed snowmelt depth It indicates how much snow including new snow falling during the snowmelt period must be melted in terms of computed snowmelt depth in order to decrease the snow covered area to a certain proportion of the total area and ultimately to zero The shape of this curve depends on the initial water equivalent of the snow and on the amount of new snow 62 MDCexci modified depletion curve of snow covered area with new snow excluded This curve is derived from MDCieL by deducting the melt depths of new snow from the accumulated snowmelt depth The shape of this curve depends on the initial water equivalent of the snow cover and is independent of subsequent snowfalls The area below this curve indicates the areal water equivalent of the initial snow cover AZMincL accumulated zonal melt with new snow included This curve accumulates daily computed snowmelt depths multiplied by the respective snow coverage as decimal number and shows the total
179. n s period of record All the simulations developed for the basin are likewise located in the same database along with supporting information used by the modeler such as climate change scenario definitions The most significant improvement in the model s capability is in the area of climate change simulation The limitations imposed on climate change processing by DOS have been removed allowing a much more detailed and robust definition of change to be studied Finally the path of future research using the model has been laid by moving to an operating system and a model design that minimizes or removes the restraints to development inherent in the earlier design and operating system environment The current version of the model WinSRM Version 1 11 was developed using Microsoft Visual Basic 6 0 and contains one custom control Spread 2 5 from Farpoint Technologies Trademarks used in this document e g IBM PC IBM CORP Visual BASIC Microsoft Corp Spread grid OCX Farpoint Technologies are used solely for the purpose of providing specific information Mention of a trade name does not constitute a guarantee or warranty for the product by the U S Department of Agriculture or an endorsement by the Department over other products not mentioned 85 10 1 3 Capabilities and limitations The Snowmelt Runoff Model for Windows WinSRM has the capacity to analyze watersheds that meet these criteria Table 9 WinSRM cap
180. n trace mode 1 Micro SRM processing TIN begins from the same DOS path as did execution when TOUT created the trace file 2 All pathnames and filenames referenced in a trace file exist on the PC During trace file creation TOUT file each keystroke and mouse click is saved in file Trace file creation can be terminated at any point in the model session by pressing ALT_ When the model runs under control of a trace file TIN file the program reads and processes one command from the trace file every Y second unless the TDELAY option is included to specify the time increment This control sequence will continue until one of the following three trace ending events occurs End of Program the trace file includes the keystrokes necessary to end execution of Micro SRM End of Trace File When the last command is removed from the trace file a message is displayed and program control is returned to the keyboard and mouse Trace Suspended the user may suspend trace file processing by pressing Esc After suspension the user may optionally continue trace file processing with the same or a modified trace delay or terminate trace file processing at the suspension point with control returning to the keyboard and mouse 175 G 9 Micro SRM availability The Hydrology Laboratory supports and distributes the Snowmelt Runoff Model free of charge to any interested party The program and related files are available v
181. n 1 15 May 1993 with usual temperature lapse rate of 0 65 C 100 m that the snow covered area was exaggerated Temperature was extrapolated by too low a lapse rate Snowmelt was not computed separately for elevation zones but for the entire basin This usually leads to overestimation of meltwater input Simulation was improved by re evaluating the snow coverage and selecting a higher temperature lapse rate corresponding to the climate With regard to the large elevation range the basin should have been divided in several elevation zones Values of recession coefficient false Negative y values used with the equation k x Q so that the exponent became positive Consequently k increased with the increasing Q while it should have decreased Use equation k xQ and positive values of y so that the exponent is always negative Precipitation data from one of the stations were ascribed to the date on which it was measured at 0800 Data had to be shifted one day backwards See also Section 5 2 2 Snowmelt was computed by smooth long term average temperature thus decreasing the number of degree days Reintroduce the daily fluctuations of temperatures Erratic precipitation data Automatic extrapolation by altitude gradients from two stations resulted in negative precipitation amounts in the highest zone when gradient was negative Extrapolate averaged precipitation data by a uniform gradient as recommended in Section 5 2 Always vi
182. n out of MG ower to derive MDCexcy wa melt depths of new snow surviving in the warmer climate are put back to derive MDCcim wa as illustrated for zone A in Figure 31 11 The climate affected conventional depletion curves adjusted for the winter deficit CDC cum wa are derived as follows MDCcrm wa indicates for example that a snowmelt depth of 22 cm is needed to decrease the snow coverage to 50 Figure 31 This occurs in the present climate according to CDC in Figure 28 on 5 May In a warmer climate T 4 C in this example a cumulative snowmelt depth of 22 cm and a corresponding decrease of the snow coverage to 50 are reached already on 9 April so that the 50 point is shifted to that date Figure 32 The program takes the cumulative snowmelt depth computed by present temperatures on each day and searches for the date on which this snowmelt depth was equaled or exceeded when the higher temperatures are used for computation If the new climate implies changes of the degree day factor see Section 8 4 the cumulative snowmelt depth must be computed not only by higher temperatures but also by changed usually higher a values The computer program takes care of this matter Comparable snowmelt depths are reached about one month earlier so that CDCcrm wa 1s shifted in time against the original CDC as illustrated in Figure 32 for all elevation zones The method of CDC shifting in the summer is also explained by a numerical example elsew
183. n the Rocky Mountains Because the geographic latitude of a basin influences the solar radiation it may be advisable to adjust the degree day factors accordingly In glacierized basins the degree day factor usually exceeds 0 6 cm Cd towards the end of the summer when ice becomes exposed Kotlyakov amp Krenke 1982 The computer program accepts different degree day factors for up to 16 elevation zones which are usually changed twice a month although daily changes are possible Sometimes the occurrence of a large late season snowfall will produce depressed a values for several days due to the new low density snow The a values in the model can be manually modified and inserted to reflect these unusual snowmelt conditions As is evident from Equation 1 the degree day method could be readily replaced by a more refined computation of snowmelt without changing the structure of SRM Such refinement appeared to be imperative in a study of outflow from a snow lysimeter Martinec 1989 but is not considered to be expedient for hydrological basins until the necessary additional variables and their forecasts become available The degree day method is explained in more detail in a separate publication Rango amp Martinec 1995 5 3 3 Temperature lapse rate y If temperature stations at different altitudes are available the lapse rate can be predetermined from historical data Otherwise it must be evaluated by analogy from other basins or with regar
184. name in the Basin Simulations data grid Then click Data Climate Change Scenario Definition Click on the Existing Scenarios for RGWY79 combo box and click Normalized5794 The dialog window shown below will display the manual normalization that model developer s performed on a simulation year for the Del Norte Rio Grande basin using a 38 year period of record These monthly edit factors were developed externally and provided using the Climate Change Scenario Definition window i Climate Change Scenario Definition for Simulation RGWY 79 O O x Winter End Date a 03 31 1979 P ae Help Simulation Period Start Date E 107017197 End Date Exiting Scenarios for ROWY79 Normalized prg Mew Delete Print Climate Scenario Normalized 5794 anae parameter Starting bate Ending bate Edit Action Edi Factor Zone apse Tenens EE 6 Average Temperature tee ETA EZEN a Normalize average temperature by month using differences calculated externally between average monthly temperature for water year 1978 and average monthly temperature for the entire pernod of record 1957 1994 Normalized precip by multiplying each day by a factor denved by dividing long term monthly average F by 1979 monthly F Window A3 Normalizing temperature using externally generated adjustment factors 138 Appendix B Graphical Plots The interface for graphical plots consist of a Graphics window w
185. ng Critical Temperature lz 000 Runoff Cocfficients Rain D 600 simulation iS being Temperature Lapse Rate D 650 Rainfall Contributing rea lo di t d Lag Time to Gage hr las X recession cocfficient la 0200 edite Y recession coefficient 0 880 T mx anci accent C Program Files Microsoft Visual Studio VB9S WinSRM Setup Files RioGrand mc 1 17 2002 11 37 4M Window 11 Edit Simulation Control Information Dialog 10 3 3 1 Method of access The Edit Simulation Control Information window for a new simulation is accessed from the main window by clicking the button or alternatively when a DOS SRM data file is imported using the menu bar File Import function When dealing with a new or imported simulation a naming dialog window is displayed Name the New Simulation i x Enter a unique name to associate with this data Cancel on Window 12 Naming dialog for a new simulation After a name has been assigned the window is then displayed To display the window loaded with an existing database simulation first identify the simulation by clicking on the appropriate grid line then click the Edit Simulation button A modified version of the window will display allowing changes to the simulation control information values only 96 The Initial Parameter Values frame displays the parameter values for day 1 zone 1 as found in the simulation s parameter table 10 3 3 2 Purp
186. ng will be performed The window also displays all simulations that have been defined for this basin Control objects a window menu and command buttons are provided on this window to provide links to the functionality supported by WinSRM A description of these control objects follows WINSRM Snowmelt Runoff Model for Windows Oj x File Options Data Run Tools Graphics Reports Help Basin Definition Name lt basin name gt Reference Elevation m Units of Measurement D Latitude dec Longitude dec Description Zone Area HYPsometric Avg Elevation NE SE Avg Elevation SE Avg Elevation NW Km Mean Elevation Aspect m Aspect Aspect m Aspect m m Basin rea km Basin Simulations Runoff Simulation Starting Endina Date Temperature Temperature Precipitation Lapse Rate Ei int Critical Temp Title Date Values Distribution Distribution Distribution Distribution Distribution Current Simulation Name P Add Simulation Delete Simulation Edit Simulation File 11225 2002 11 00 M Window 2 WinSRM Main Window 10 3 2 2 Menu The WinSRM program menu is positioned at the top of the main window Window 2 The menu consists of eight menu headings under which all related program functions are grouped When a main menu heading is clicked or accessed via the keyboard using the menu shortcut key pressing Alt and the approp
187. nits of Measurement and Number of Zones All other descriptive variables in the frame may be changed at will Generally however these variables will also be set only once The zone definition grid table Window 31 is the tool for entering and displaying the zonal characteristics of the current basin Zone area and hypsometric mean elevation are the only values currently in use by the model The remaining aspect related table cells are intended for possible future use as variations of the basic model algorithm are developed The basin drainage area is automatically calculated using the areas entered in the zone definition grid table and displayed in bold blue immediately underneath the table 10 3 2 5 Basin simulation frame The basin simulation frame Window 2 Window 34 is a container used to display the individual simulation definition s that have been created for the current basin Each line in the frame s data grid Window 34 represents a unique simulation present in the database Clicking on a grid line identifies that simulation as the current simulation The current simulation is an important property the software uses whenever any simulation related program functionality is accessed For instance if you click the window s dit Simulation button the parameters for the current simulation are loaded into the Edit Simulation Control Information window Each grid line constitutes a simulation conta
188. nn Cut a Copy Fi Paste Print 0 000 0 000 0 000 0 000 n nnn Edit multiple days Switch Scroll Style 0 000 0 000 D nnn 0 000 0 000 D nnn 0 000 0 000 0 000 0 000 n nnn Chrl x Chic Ctrl 0 000 0 000 naonn 0 01 RU 0 01 0 01 0 01 0 01 O_ 01 RU 0 01 0 01 DRUI DRUI Of 0 01 0 01 opi nor P Print Help Cancel Done 10 26 2005 14 30 PM E C Program Files WwinSAM Viola endtol lh mdb Window A1 Copying normalized temperature data to the Windows clipboard b Copy the selected data to the Windows clipboard Right click the column then click Copy c Click to return to main window 136 Assuming the current basin simulation is unchanged from Steps 3 5 open the Edit Basin Variables window by clicking Data Variables Current Simulation see below Select the target column for the copy the column corresponding to that chosen in 6a by clicking its column heading In this example the column name is Tavg After the column is selected highlighted right click on the column then click Paste to copy the data from the Windows clipboard to the data grid w Edit basin variables Ioj x Edit Basin Rio Grande Basin Simulation AGw 79 0701778 5 320 0 000 070278 5 240 0 000 003278 5 040 0 000 00478 4 870 0 000 00578 4 790 0 000 070678 4 370 0 000 02 07 78 4 700 0 000 0708278 4 700 0 000 009278 4 700 0 000 01078 4 87
189. odel s FORTRAN code was downloaded to an IBM PC and modified to operate in the PC environment That same year a decision was made to develop a unique PC oriented version of the model taking full advantage of the PC s inherent capabilities The result of that decision was Micro SRM Version 1 0 Additional refinements were incorporated in several subsequent Micro SRM Versions up to and including 4 0 However SRM itself remained unchanged and relatively simple so that the computations by Equation 1 could still be performed by any pocket calculator which has a function x The DOS PC program automatically handles the multiple input of temperature and precipitation for up to 8 elevations zones of a basin any desired lag time and complicated snow rain situations A model simulation for up to 365 days is finished within several seconds the computed hydrograph is immediately displayed in comparison with the measured discharge and if desired quickly printed Also the achieved accuracy is automatically computed and displayed A summary of parameter values can be displayed after each run so that adjustments can be made and their effect assessed SRM does not require numerous runs because calibration is not necessary The ease with which the results are obtained should not lead to a replacement of the deterministic approach of SRM by a try and see philosophy SRM is designed to operate with physically based estimates of parameters that should not requi
190. ons by moving the menu bar using the up arrow or down arrow then pressing Enter lt pressing the desired sub option s highlighted letter or left clicking on the menu selection FilelO provides the modeler with tools to load save and delete model data files and to import digital input data from non model sources Load The Load option allows the user to identify and input a SRM data file see SAVE below into the model After a successful load the model assumes a state identical to that existing when the referenced data file was created Save The Save option allows a user to create and name a file containing the ASCII representation of the data currently resident in the model s data structures The file qualifier SRM completes each save file s name making it easy to recognize and manage model data A data compression algorithm shrinks the 100K internal data structures into a more manageable size a complete five zone basin might require 10K using shorthand notation identical to the FORTRAN repeat count concept IBM VS FORTRAN 1983 Delete This option provides the model user with the ability to delete any SRM data file SRM qualified currently residing on any of the PC s storage devices Import The Import option provides a method for introducing digital input data from a non SRM source to the model Using a simple format press F1 for a popup describing the import format any of the 13 model variables parameters detailed on t
191. opied to the clipboard in Step 6 will replace the data in the column Since the temperature distribution for RGWY79 is Basinwide see the main window no further temperature modification is necessary Basinwide variables are always stored in Zone in WinSRM tables Click to close the dialog and return to the main window Click on RGWY77 in the Basin Simulations grid Click on Data Variables Current Simulation The Edit basin variables dialog appears Click on the column header Precip The entire column should be highlighted Right click on this column then select Copy or press Ctrl C to copy the precipitation data for zone A to the clipboard Click Cancel to close the dialog and return to the main window Click on RGWY79 in the Basin Simulations grid Click on Data Results The Simulation results editing dialog appears Click on the column header PrecipCC The entire column should be highlighted Right click the column and select Paste from the popup menu or press Ctrl V The 1977 precipitation data copied to the clipboard in Step 9 will replace the data in the column Click to close the dialog and return to the main window Copy the 1977 precipitation for zones B and C to SimResults zones B and C by repeating steps 98 10 for each zone At this point your customized climate change definition is complete Before running the climate change simulation you may want to save the database to preserve the SimResults
192. or the whole year Consequently the SRM climate program has been extended to handle the winter half year as well 8 1 Snowmelt runoff computation in the winter half year In the winter half year usually October March in the Northern hemisphere the evaluation of the snow coverage is more difficult than during the snowmelt season because of persistent cloudiness and low sun angles Satellite data if available are not frequent enough to distinguish the stable snow cover from frequent transitory snowfalls which are subsequently melted An assumption of a stable snow cover between two available satellite measurements may lead to an overestimation of the snow coverage Two options are available to handle this problem during the winter months 58 Option 1 Put S 0 so that each precipitation event recognized by Terry as snow is automatically stored and subsequently melted over the whole area 1 S 1 This method is applicable only if precipitation data are good enough to represent the input or if they can be adequately adjusted This is in many cases not possible due to the well known catch deficit of precipitation gauges and due to the lack of measurements in the high elevations of mountain regions It is for this reason that SRM uses the snow covered area whenever possible for computing the runoff input Option 2 Assume a stable snow cover S 1 or a little less in a rugged terrain in January and February for example and S 0
193. ording to the balance method 114 6 10 m from computed daily melt depths and 142 10 m according to the increase of the yearly runoff This last amount is of course obtained only from an incomplete water balance This hypothetical example assumes that the warm year occurs in present times In reality it can be expected that when such temperature increase is reached the glacier areas will be in the meantime reduced Modeling of a long term sequence of years would thus be needed as outlined in the next section 9 2 Long term behavior of glaciers in a warming climate The WinSRM program enables climate runs to use as a new climate a set of normalized daily temperatures and precipitation to derive a normalized year Section 8 5 as well as a data set T P of a historical year if available and a data set of a future year if the data can be generated So far hydrological models took climate scenarios into account in one step For example the present climatic conditions were replaced by conditions predicted for the year 2050 This procedure is inadequate for glaciers because temperature will not continuously increase by a fraction of a degree each year On the contrary as in the past temperature will fluctuate from year to year with an underlying increasing trend A stochastic series of temperature and precipitation data can be generated Gyalistras et al 1994 which can reach the values given by a climate scenario in the
194. ormalized year The procedure just described in theory 1s used to derive a normalized year from the 1979 data set The long term data are monthly average temperatures and monthly average precipitation totals from the period 1957 1994 The program takes these data into account as a climate scenario and derives the normalized snow cover Figure 35 which serves to compute the normalized runoff Figure 36 This normalized year is labeled 9979 since it is derived from the year 1979 The climate program is then run with the desired climate scenario in our example T 4 C Figure 37 shows the climate affected snow cover and Figure 38 the climate affected runoff Numerical data indicating the redistribution of the winter and summer runoff are listed in Table 7 The climate effect for the normalized year 9979 is greater than the climate effect for the selected single year 1979 Table 7 Seasonal redistribution of runoff for a normalized year in the Rio Grande basin near Del Norte hydrological year MTI due to climate change The extremely cold winter in 1979 reduced the effect of a temperature increase T 4 C so that the winter runoff increased only about 60 With a normalized year for the present climate 1957 1994 the winter runoff is doubled as a result of the same temperature increase The WinSRM program enables sets of precipitation and temperature data from various years to be transferred to an actual year as a changed clima
195. ormat In this situation the residual value is set equal to the last S value in the CDCnuormaL This value will typically be zero except in cases where the zone includes a glacier or permanent snow cover In the infrequent cases where the difference is positive CDCcim gt CDCnormat each missing daily value after the last derived day will take on the corresponding day s S value from CDCyormat plus the difference 5g Plot MDCexc and MDCexcL wa Sh Plot AZMyct AZMgxcL AZMexcr wa 5i Optionally plot MDCexcr wa MDCcum wa 5j Plot CDC CDCcim wa Sk Replace CDC with CDCcim wa ma Step 6 Summer Changed Climate Run a final summer simulation for the changed climate This run uses the existing climate change scenario with the derived CDCcrm wa ma Curves to produce runoff after the climate change Qsm After completion of step 6 the model changes state Several new plots Appendix B and a new tabular report are enabled to augment the model s output options 106 10 3 8 Long term monthly averages window 10 3 8 1 Method of access The Long Term Monthly Averages Window is displayed by clicking the Main Window menu bar Tools Calculate Long term Monthly Averages Window 22 WINSRM Snowmelt Runoff Model for Windows File Options Data Run Tools Graphics Reports Help Calculate Long term Monthly Averages Define a custom time lag distribution For the basin Easin Definition Hame Rio Gran
196. ose The Edit Simulation Control Information window is used to define the initial parameter values and the control information for a new simulation When a new simulation is created a parameter table named after the simulation is added to the database The default parameters provided from this window are used to populate the new table by assigning the parameter default value to all corresponding daily values for each zone for the entire simulation period Once created the simulation parameter table can be edited in more detail using the Edit Simulation Parameters window by clicking the Data Parameters on the main WinSRM window menu 10 3 3 3 Buttons Accept Accept and apply the changes to the simulation and return to the Main Window Cancel Ignore any changes Return control to the Main Window Help Display help window describing the window and its functions Toggle quickhelp comments related to the window elements fields 10 3 4 Edit type of data window 10 3 4 1 Method of access After a current simulation has been identified by clicking on a row in the Basin Simulations data grid the variables and parameters used for the simulation may be displayed and edited using the Edit Window The appropriate window is accessed through the Main Window menu bar WINSRM Snowmelt Runoff Model for Windows Fie Options Data Run Tools Graphics Reports Help Basin befi Climate Change Scenario Definition Mamas h variables Period o
197. porary increase in the snow coverage by a summer snowfall and to missing Landsat data from the preceding overflight Hall Martinec A a ice deine 28 Figure 6 Depletion curves of the snow coverage for 5 elevation zones of the basin Felsberg derived from the Landsat imagery shown in Figure 4 Baumgartner 1987 oococcccccncococcnnncocnnnnnnnnnnnnonononos 28 Figure 7 Average runoff coefficient for snow cs for the alpine basins Dischma 43 3 km 1668 3146 m a s l and Durance 2170 km 786 4105 m a s l Martinec Rango 1986 nsss 31 Figure 8 Average runoff coefficient for rainfall Cr for the basins Dischma and Durance Martinec ROMO LIGO A 31 Figure 9 Average degree day ratio a used in runoff simulations by the SRM model in the basins Dischma 10 years Durance 5 years and Dinwoody 228 km 1981 4202 m a s l Wyoming 2 Veas iMartinec SeRando 1 IO laa as 33 Figure 10 Recession flow plot Q vs Dun for the Dischma basin in Switzerland Either the solid envelope line or the dashed medium line is used to determine k values for computing the constants x and y in Equation 7 Martinec amp Rango 1986 ocococcccccoconconococoncononoconnnnoroncnnonoronnnnonerornnnoross 35 Figure 11 Range of recession coefficients k related to discharge Q resulting from various evaluations Martinec S El ele e EE 36 Figure 12 Snowmelt hydrographs illustrating the conversion of computed runoff amounts for 24
198. poration conditions this error is eliminated In a warmer climate the winter input to runoff increases as shown in Figure 26 so that there is less snow on 1 April if winter precipitation remains unchanged A combined effect of a warmer climate and increased winter precipitation can result in rare cases low temperatures in high mountains with little or no effect of the temperature increase in an increased snow accumulation 61 GLOBAL WARMING AREAL WATER EQ cm Ale ct WA A Al All 1 E MM Y Y VE W MI IX M My S Fig 26 Illustration of the snow accumulation in the winter and snowmelt in the summer in a warmer climate hypothetical example 8 3 Runoff simulation for scenarios of the future climate In order to evaluate the effect of a warmer climate on runoff in mountain basins the SRM program uses the real seasonal snow cover of the present as monitored by satellites and models a climate affected seasonal snow cover The future snow conditions in terms of snow covered areas and areal water equivalents in different elevation zones constitute useful information for planning water management and winter tourism Using snow coverage data as model input the climate affected runoff during the whole hydrological year is computed It allows examination of the changes of daily runoff peaks and the redistribution of runoff volumes in the winter and summer half years The procedure is illustrated by evaluating the effect of a te
199. pper Rhine 25 August 1985 Rainfall peak inadequately simulated because the daily amount was below the special program automatic threshold Threshold was lowered but the peak simulation deteriorated further small tributary basins of upper Rhine 1985 47 Table 5 Errors experienced by SRM users and their correction Cause Snow cover depletion curves distorted by summer snowfall see Figure 5 Eliminate satellite observation after a snowfall event and redraw the depletion curves Input of snow coverage data S shifted by 1 month Correct the S input Input of snow coverage data broken off by the end of August Complete the S input Decimal point in the measured precipitation and discharge data shifted and thereby increasing the measured runoff 10x but the simulated runoff only slightly Correct the decimal point Values x y for the recession coefficient formula Equation 7 were taken over from a much larger basin so that k exceeded 1 0 for low Q Correct x y with regard to the basin size Possibility of this error was eliminated in the computer program Version 2 01 and later versions 3 0 3 1 3 11 and 4 0 SRM simulates natural runoff while the measured runoff was influenced by storage on weekends and release from artificial reservoirs for hydropower These interventions must be corrected in order to compare simulated and measured runoff Rainfall input is concentrated to shorter periods than snowmelt inp
200. pse Rate LR ctors AN oefrLicilencts CS oefficients CR ributing Area RCA SEPEN Ee l Variable Code A FlHelp F2Summary F3Filel0 F5Plot F6Print F7Compute Full Screen Menu and FilelO Pop Up Menu G 3 4 Data entry screens Micro SRM s data input screens are simply electronic versions of standard printed forms income tax forms being a good example which consist of one or more pages of information requests with associated reply spaces or fields provided for entering data values In Micro SRM the basic data entry screen display consists of background text that names and or briefly describes each field field shadows strings of a special character that visually define both the physical dimension and screen location of each data field and a blinking data character or cursor 1 that marks the model user s current position within a data screen The current value of the model variable represented by each field is superimposed over its corresponding field shadow Field values are accentuated on the display by varying light intensity or color on monochrome and color monitors respectively Cursor IT Background Text sd Name John Doe Address Data Field Shadows Data Entry Screen Components 155 The model user types text or other data onto the data fields provided on the screen display at the position marked by blinking cursor Micro SRM interactively controls and validates the characters entered using each
201. re much change after the initial selection Seemingly unsatisfactory results have been frequently improved not by adjusting the parameters but by correcting errors in data sets and in the input of variables A prime consideration in the design of Micro SRM was to develop a snowmelt modeling environment such that the model user was provided not just model algorithms but a complete set of tools for managing the associated model processes data entry storage and retrieval display of data and results Traditionally the most time consuming and error prone activity involved in using any physically based model has been the accumulation of large amounts of input data in the form and format needed to drive the simulation with actual execution of the model a trivial task by comparison Recognizing this we chose to pattern the design of Micro SRM after that originally developed during the automation of the Soil Conservation Service s SCS Technical Release Number 55 TR 55 Urban Hydrology for Small Watersheds Soil Conservation Service 1986 This joint ARS SCS effort provided valuable experience in developing highly interactive front ends for interfacing complex models with model users 84 The approach used by the ARS SCS programming team that automated TR 55 was to develop an efficient easy to use highly interactive data entry manipulation environment and include model algorithms as just one of many functions that support and use that environ
202. red areas according to CDCcrm wa compared with the original runoff simulation in Figure 27 The run is started by discharge computed on 31 March with T 4 C see Step 3 The year round climate affected hydrograph thus consists of a winter simulation with T 4 C and estimated S and of a summer computation by the climate program with T 4 C and S from CDCecrm wa Figure 34 shows year round hydrographs computed with temperatures of 1979 and with temperatures increased by 4 C As listed in Table 6 the future winter runoff would be increased at the expense of the summer runoff The actual effect is greater because the increased climate affected runoff in late March is carried over to April as recession flow as explained elsewhere Rango amp Martinec 1997 Proportionally the redistribution of runoff is relatively small because the cold winter of 1979 did not allow much snowmelt even with the increased temperatures see Section 8 5 about normalization In the given example precipitation was not changed in the new climate but the climate program can also handle the combined effect of changed temperatures and changed precipitation Usually the temperature effect prevails but as mentioned in Section 8 2 increased precipitation and absence of melting conditions in high altitudes in winter may convert the winter deficit to winter gain or positive winter adjust ment In this case the computer program derives MDCexcr wa by stretching
203. relates the snow coverage with cumulative snowmelt depths including new snow in the summer Figure 29 D MDCexcry is derived by eliminating melt depths referring to new snow from cumulative snowmelt depth Figure 29 The area below this curve indicates the initial areal water equivalent of the snow cover as shown in Section 7 8 The climate effect is taken into account by depriving MDCrexc of the winter deficit computed in step 4 and MDCexcr wa winter adjusted is derived The program prints both MDCexc and MDCexcr wa In zone A AHw 36 94 cm was computed so that MDCexcr 1s cut off on the day when this value was equaled or exceeded This happened on 27 April when XM for MDCexcy was 42 04 cm not reduced by S therefore higher MDCexcz wa thus derived is shifted to start on 1 April 9 The cumulative snowmelt is printed in relation to time as zonal snowmelt that is to say reduced each day by the respective percentage of the snow coverage The accumulated zonal melt curve AZMexcy indicates graphically Figure 30 that the winter deficit in this example AHw 36 94 cm was exceeded on 27 April when the accumulated zonal melt amounted to 37 69 cm The previous day s total 35 9 cm is also printed so that the user can cut off the MDCrxcr at the previous day s value if the next day s value is much higher than the computed AHy 10 After melt depths of new snow of the present climate and the winter deficit had been take
204. riate underlined character a drop down sub menu appears see Window 3 for an example of a drop down providing access to logically related program functionality File The file menu pull down Window 3 groups together all of the functionality for moving information to and from WinSRM memory 89 WINSRM Snowmelt Runoff Model for Windows Eile Options Data Run Tools Graphics Reports Help Mea Open F4 Import Ctrl I ef Save Save Os Page Setup Exit M C Program Files Microsoft Visual Studio BSS wins R M Setup Files RioGrand mdb C Program Files Microsoft Visual Studio VB9stest mdb Program Files Microsoft Visual Studio WBS WinsSRM Setup Files Tlecillayaet mdb Window 3 WinSRM Main Window File Menu Drop down Choice Explanation New Create a new basin database Import a DOS srm data file into a new existing database Save current input database using current file name ile Save As Save current input database with different filename Open Open an existing WinSRM database Page Setup Setup printer for page output Exit Exit software List of Recent Click to open one of the four most recently accessed Files WinSRM databases Options The Options menu item provides access to functionality related to the operation and display of the WinSRM software WINSKRM Snowmelt Runoff Model for Windows File Options Data Run Tools Graphics Reports Help Be vw Run year ro
205. riefly describe these tasks 10 4 1 Building a WinSRM basin database There are two procedures for creating a new WinSRM basin database Each procedure begins with the program in a new data state either at program startup or by clicking the File New menu option on the main window menu bar 10 4 1 1 Manually populate your WinSRM database The procedure that will be used by most new users of WinSRM will involve manual population of a new database This sounds worse than it should prove to be in that the model supports cut and paste of data from within WinSRM as well as from other sources typically spreadsheet programs Several types of data are required before a simulation can be attempted These data are stored in database tables within the new database The tables are described in the order in that they are created during development of the new database BasinDescription table When a new basin database is created the first step in the definition process is entering several values that are unique and will typically not change This is accomplished by typing the values in text boxes in the Basin Definition portion of the main window see below Easin Definition Home new basin Humber of Zones Reference Elevation m Units of Measurement Aetric Latitude dec Longitude dec Description Window 31 Create the basin description table 115 In the example above the basin name new basin and the
206. rrrerrnro 102 A EI Le ele 102 A ie 102 EE sng eae nae ene Ee 102 10 3 7 4 Processing steps controlled by the step button 106 10 3 8 Long term monthly averages window 106 105320 MENO Ol le 106 LO 30 2 RUIDOS ua eee eee eee toe eee eles 107 10 3 9 Custom lag time distribution WINDOW ccccceceeseeeeeeeeeseeeeeeeeueeeeeaeaueetetanantetes 107 10 39 LME OA e 107 A E ne 107 1053310 Grapes WINGO Wa arial 107 10 3 10 L Method OF ACCESS aia n 107 10 3102 PUMO SE seen de tall a EE ee 108 RA E LA Button Me EE 109 023 40 LING SatCEIDULES Lg ele EE 110 10 30 Ll ER giele OF AC e EN 110 LO Se LEA PP O A 110 TS a RE BU CEO TEE 111 10 312 O tp ut aent WINGO EE 111 t0312 MENO Be 111 TOA RUDO ada ia 111 TUS IS te EE 112 TO 31278 AV aAiaDIS repor EE 112 RW 13 INESIS Pl WING OW EE 113 TO Sek Sob giele ee 113 TO E PUDO SE EE 113 SL MCN EE 113 103 134 WWIRGOWCONUON EE 114 104 le Eeer Le EE 114 10 4 1 Building a WINSRM basin database occccccccccnccnnonococonnnnononcnnonorornncnrnrnnannoross 114 10 4 1 1 Manually populate your WinSRM databaee 114 10 4 1 2 Import an existing DOS SRM data nie 116 104 2 Data DaSe Gata elias 117 104 Se RUANDO WINSRM eg sae dice Ai 117 10 4 3 1 WINSRM processing StEPS ccccccsesesesecsecuueueveueuseraeguaueueraveneenesuserarass 117 EU ARE r IEN 118 11 1 General rererences ina id 118 EE Tererences TOF Table La io 121 Append A EXa Mp EE 128 Example 1 Create a new basin da
207. s Basin Definition or Variable Parameter Menu display screens Micro SRM will refuse this function request and generate an appropriate error message if the PC has no graphics adapter present or a 165 simulation forecast has not been run since last data entry If the FS key is hot the plot options menu is displayed The five Micro SRM plot products available during normal processing are e Measured vs Computed Streamflow e Snow Depletion vs Time e Snow Depletion vs Total Snowmelt Depth e Accumulated Zone Melt Depth vs Time e Build External Plot File If Micro SRM performs a climate change run the state of the plot menu changes presenting the user with the following choices e Computed Flow Before After Climate Change e CDC Clim Model Adjusted vs Time e Build External Plot File The following keys are hot while this menu is displayed Fl Help for Plot Menu Up Down Arrow Move menu bar to next prior choice Enter lt 4 Invoke selected plot option ESCape Exit function After a plot has been displayed the user may request a hard copy by pressing function key F10 Press Enter lt to cycle through any additional zone s in the basin Press ESCape to return to the Plot Menu The fifth option on the menu Build Plot File generates an ASCII file containing all the information necessary to recreate a user selected combination of the Micro SRM plot products using non model hardware software F6 Print The PRINT
208. s Shift values backward edit factor number of days Normalize values temperature and precipitation only Table D 2 Numeric equivalents for edit actions 148 Appendix E Additional Windows BH File Display Print Edit SRM Reports Help Climate Change Deficit Gain Summary Close 210 x Date 11 02 2004 File C Program Files tinsEM BioGrand _mib Units cm Climate Scenario Tplus4 one 1 Total P Winter PF 4 12 Total P Winter clim F 174 12 fainter Total Zonal Input i fone 1 10 3385 Sum M 5 H Sum M 1 5 2 185 Sum P rain 7 52 t nter climj Total Zonal Input 1 one 1 417 2761 Sum M 5 H Sum Mii 5 24 1767 See Figure 21 and Sum P rain 23 1 Section 8 3 8 Winter Change OHW done 1 36 93 Zone 1 Winter change deficit of 36 9382 was equaledfexceeded by accumulated zonal melt on 4 27 1979 a shift of 27 days AZMexcl on 4 27 1979 37 6915 Prior day s AZMexcl 35 89883 See Section 8 3 9 el C Program Files inssAb SAM pts out 11 2 2004 2 52 Phi a Window El Climate Change Deficit Gain Summary Report 149 Appendix F Program Message Dialogs Import failed Simulation units inconsistent with basin Physical variable data For simulation time period are present in the database Overwrite i Mo Cancel Message 4 Physical data already present in database Enter Forecast Frequency l x Enter the frequency 1 9 days
209. s and the physical dimensions of required input data variables may not exceed the size of the screen fields defined for their storage This should only occur during variable parameter data entry where daily values may contain no more than 6 digits The following sequence of steps describes a typical Micro SRM session 1 User types SRM4 then presses Enter lt to invoke Micro SRM 2 After viewing the Introduction screen the user presses PgDn to view the Program Options data screen 3 The user makes any required changes to the program control information displayed on this screen then presses PgDn to view the Basin Definition data screen 4 After filling the required data fields on this screen the user begins the variable parameter data entry process by pressing PgDn to view the Basin Variable Parameter Definition Menu 5 The user selects a variable parameter from this menu then presses Enter lt to display daily values for the first two months of the snowmelt period for the selected variable 6 The user fills the daily value fields with appropriate data then presses PgUp or PgDn to view a different two month slice of daily values for the same variable 7 After data entry for the selected variable parameter is complete the user presses ESCape to return to the Basin Variable Parameter Definition Menu 8 The user repeats steps 5 7 for all required variables and parameters At any point during these steps th
210. s case the column is highlighted Then press the right mouse button to expose the edit function context pop up menu and click on the desired function to apply to the selected cells e Identify multiple complete columns by clicking the first column header pressing the Shift key and clicking the last column The target column s are highlighted Proceed with the edit as described above e Identify a cell or range of cells by clicking on the first cell then while pressing the Shift key clicking on the last cell in the range Proceed with the edit as described above e Equivalent keyboard edit commands are supported within the data edit grid Identify the cell by setting the focus rectangle using the cursor keys Identify a block of cells by identifying a corner cell of the desired block of cells using the cursor keys then while pressing the Shift key using the cursor keys to move to the opposite corner of the block Once the block is highlighted the edit function can be selected from the right mouse click edit context menu or by pressing the Ctrl key then pressing X cut C copy or V paste 10 3 4 4 Buttons Display the Help A Save changes return Window for additional to Main Window information Help Cancel Done Turn Quick Print the ENTIRE Ilgnore any changes Help on off contents of the grid return to Main Window 99 10 3 5 Climate change scenario definition window 10 3 5
211. s on a time scale AZMexci accumulated zonal melt with new snow excluded This curve is derived from AZMnct by deducting the zonal melt of new snow from the accumulated zonal melt Again it relates the successive totals to time The final total is the areal water equivalent of the initial snow cover as also indicated by MDCexcr MDCcum modified depletion curve of snow covered area for a changed climate This curve takes into account the amount of snowfalls changed by the new climate If there is no change it is identical with MDCwct CDCenm conventional depletion curve of snow covered area in a changed climate CDC cum ma conventional depletion curve of snow covered area in a changed climate derived from MDCincL adjusted for the input to SRM runoff computation model adjusted It appears in publications disregarding the winter effect of a changed climate MDCrxcL wa winter adjusted curve The effect of a warmer winter is taken into account by decreasing the curve according to the winter deficit With a simultaneous increase of winter precipitation a positive balance of the winter snow accumulation may result in which case the curve is increased The area below this curve indicates the areal water equivalent of the initial snow cover in a changed climate AZMexcuwa winter adjusted curve The effect of a warmer winter and if necessary of a changed precipitation is taken into account The final total 1s the water equivalent
212. s the working copy that the simulation uses as its input source and output destination All reports and graphical plots that display calculated values obtain their information from this table CI wame O m TmaxCC Maximum temperature changed climate Ps tn se Mnmumtempestwe Pe to ge meee ooo f PrecipCC Measured precipitation changed climate E afse E SOS CAES E 7 T n 145 Single Rain runoff coefficient see Section 5 3 1 Single Lag time see Section 5 3 2 Integer Rainfall Contributing Area see Section 5 3 5 single x If a single set of temperature data are provided at the reference elevation the model uses lapse rates see Section 39 Cpr Single 5 3 3 to adjust the values for the effects of elevation to each zone s hypsometric mean elevation C 100m or F 1000 ft Also see Simulation ndex table description osm moe ra neve mm ele E SES sme SSS lege 1 e o rs ome E S rs Toomer mm ole fen eleng mm wea See wen se ween A SS SS ACTA REES loss s o EIC f s S 146 ClimateScenarioIndex Table The ClimateScenarioIndex table occurs once per database The table contains one record for each climate scenario table present in the database Each record in this index table also contains the name of the simulation for which the scenario table was prepared Simulations may own multiple climate change scenarios Field Name Data Type Description
213. set of normalized temperature values are then used to define the changed climate 107 Normalizing precipitation involves multiplying each daily precipitation value for the zone by the ratio of the long term monthly average precipitation over the current monthly average precipitation 10 3 9 Custom lag time distribution window 10 3 9 1 Method of access The Custom Lag Time Distribution Window 1s displayed by clicking the Main Window menu bar Tools Define a custom time lag distribution for the basin Window 24 WINSRM Snowmelt Runoff Model for Windows File Options Data Run Tools Graphics Reports Help Calculate Long term Monthly Averages Basin Definition e i Define a custom time lag distribution For the basin Name rio Grande J Window 24 Access the Long term Monthly Averages window 10 3 9 2 Purpose The Custom Lag Time Distribution window Window 25 provides an alternative method for establishing a time distribution for allocating snowmelt over multiple days When used this custom allocation overrides the time lag values present in the simulation tables found in the basin database See Section 5 3 7 for more information on the time lag parameter w Custom Time Lag Distribution Ioj x of days contributing to a daily melt total 1 6 Current Doay l Daye Days Day 4 Doy 5 SS SS SS SS SS SS Distribution Total must equal 1 04 U Help Cancel l Window 25 Custom Time
214. sh units is also provided in the computer program 21 5 NECESSARY DATA FOR RUNNING THE MODEL 5 1 Basin characteristics 5 1 1 Basin and zone areas The basin boundary is defined by the location of the streamgauge or some arbitrary point on the streamcourse and the watershed divide is identified on a topographic map The basin boundary can be drawn at a variety of map scales For the larger basins a 1 250 000 scale map is adequate After examining the elevation range between the streamgauge and the highest point in the basin total basin relief elevation zones can be delineated in intervals of about 500 m or 1500 ft In addition to drawing the basin and zone boundaries several intermediate topographic contour lines should be highlighted for later use in constructing the area elevation curve Once the boundaries and the contours have been determined the areas formed by these boundaries should be planimetered manually or automatically Figure 2 shows the elevation zones and areas of the South Fork of the Rio Grande basin in Colorado USA The elevation range of 1408 m dictated the division of the basin into three elevation zones Once the zones are defined the various model variables and parameters are applied to each zone for the calculation of snowmelt runoff To facilitate this application the mean hypsometric elevation of the zone must be determined through use of an area elevation curve Many of these steps can be expedited through the us
215. simulation 9 The physical basin variables temperature precipitation and snow covered area for the time period denoted in the current simulation have now been normalized Save the mirror database at this point to preserve the changes made above File Save The normalized year thus created can now be utilized as a non biased source year for other climate change studies 137 Manual alternative for normalizing temperature and precipitation for a basin The climate change scenario describe in step 4 above takes advantage of a built in WinSRM function that automatically generates monthly temperature and precipitation adjustment factors which serve as the climate change scenario These factors are the differences between the long term monthly averages for T amp P for the basin and monthly averages for the current simulation year Daily temperatures T for each month are adjusted or by the difference between the current year s monthly average T and the long term monthly average T for the corresponding month Daily precipitation P amounts for each month are adjusted by multiplying each daily value by a proportional factor LT monthly average P monthly average P An example of Step 4 performed manually can be found in one of the sample databases RioGrand DN mdb included in the standard installation package Load this database into the program and select the simulation RGWY79 by left clicking the
216. son of historical precipitation and runoff ratios provide a starting point for the runoff coefficient values However these ratios are not always easily obtained in view of the precipitation gauge catch deficit which particularly affects snowfall and of inadequate precipitation data from mountain regions At the start of the snowmelt season losses are usually very small because they are limited to evaporation from the snow surface especially at high elevations In the next stage when some soil becomes exposed and vegetation grows more losses must be expected due to evapotranspiration and interception Towards the end of the snowmelt season direct channel flow from the remaining snowfields and glaciers may prevail in some basins which leads to a decrease of losses and to an increase of the runoff coefficient In addition c 1s usually different for snowmelt and for rainfall The computer program accepts separate values for snow Cs and rain Cr and allows for half monthly and if required daily changes of values in each elevation zone Examples of seasonal trends of runoff coefficients are given in Figures 7 and 8 with the half monthly values connected by straight lines The runoff coefficients can even reach lower values in certain semiarid basins particularly in the lowest elevation zone of such basins Of the SRM parameters the runoff coefficient appears to be the primary candidate for adjustment if a runoff simulation is not at once successful
217. ssion formula Equation 7 derived usually for summer conditions sometimes allow the discharge to decrease too low so that a slower recession formula is indicated By using the Equation 14 l Le gt E 14 x y can be adjusted in order to prevent the discharge from sinking below a selected level after a long recession period In view of frequent snowfalls values of the degree day factor lower than those used in the summer are recommended Values of cs and cr higher than in summer can be expected The main purpose of the winter runoff simulation is the evaluation of the runoff redistribution between the winter and summer half years Consequently it is more important to compute the winter runoff volume as accurately as possible rather than to try to improve the daily accuracy R 59 8 2 Change of snow accumulation in the new climate As the first step the effect of a climate change on snow covered areas and runoff was evaluated in the summer half year only assuming an unchanged initial snow cover on April Martinec et al 1994 For a year round temperature increase the seasonal snow cover on April is deprived of a certain snow water equivalent and perhaps snow cover by additional snowmelt and by a conversion of some precipitation events from snow to rain in October through March This decrease of the snow water equivalent is computed by rewriting the input part of the SRM formula as follows A Hw Y as Ts S
218. started at 3 C in April at the beginning of snowmelt and diminished to 0 75 C in July This seasonal trend with a narrower range appears to be applicable in other basins At certain times SRM may not take notice of a sharp rainfall runoff peak because the corresponding precipitation is determined to be snow the extrapolated temperature being just slightly below the critical temperature In such cases the assignment of critical temperature and the temperature lapse rate values should be reviewed and logical adjustments made in order to change snow to rain It is of course difficult to distinguish accurately between rain and snow because the temperature used is the daily mean while precipitation may occur at any time day or night Le in the warmer or colder portion of the daily temperature cycle As a possible refinement formulas have been proposed Higuchi et al 1982 to determine the proportion of rain and snow in mixed precipitation conditions 5 3 5 Rainfall contributing area RCA When precipitation is determined to be rain it can be treated in two ways In the initial situation option 0 it is assumed that rain falling on the snowpack early in the snowmelt season is retained by the snow 34 which is usually dry and deep Rainfall runoff is added to snowmelt runoff only from the snow free area that is to say the rainfall depth 1s reduced by the ratio snow free area zone area At some later stage the snow cover becomes ripe the us
219. step A is Average Select the A zone if temperature option selected is Max Min Repeat modifying rule steps above for precipitation Change the Winter End Date 1f necessary Enter comments in the Comments box if desired Click to return to the main WinSRM window 135 5 Run a climate change simulation using the scenario created in Step 4 Click Run Climate Change Select the scenario created in Step 4 from the Scenario Definitions drop down menu Proceed with the simulation by clicking the six Climate Change Processing Steps buttons Click after completion of the climate change simulation The Results database table now contains a normalized set of temperature and precipitation data a set of normalized snow depletion curves derived using said data and a year of normalized runoff values To preserve this normalized year for use in subsequent climate modeling activities these data need to be saved to the mirrored database created in step 2 overwriting the corresponding actual observed data The following steps accomplish this task 6 Copy the normalized temperature values generated by the climate change run Step 5 to the corresponding physical variables in the current mirror database For purposes of this example we will assume that basin wide average temperatures were used in the original simulation Step 3 Click Data Results to display the
220. sually inspect input precipitation data Occasionally missing temperature data were interpreted by the computer program as 0 0 C Inspect temperature data and complete missing values by interpolation The winter deficit was exceeded on a warm day with a high snowmelt depth so that too much was cut off from MDCexc Consequently CDCcum was shifted too much and less runoff resulted Cut off MDCgxc by the value of the previous day nearly equaled Frequent fohn wind in this period temperature differences between Weissfluhjoch 2693 m a s l and Davos 1560 m a s l correspond to a lapse rate as high as 0 95 C 100 m The value of 0 8 C 100 m was used 49 Table 5 cont 18 Evaluation of the climate effect in the Kings The decrease of the snow water equivalent on 1 April due River basin 1973 For a temperature increase 4 C snow cover on 1 April disappears in the C zone 1700 2300 m a s but still exists in the lower B zone 1100 1700 m a s l Failure of runoff simulations in the basin Pskem the derived constant y in the recession formula Eq 7 is negative so that the recession is steeper for low flows than for high flows The envelope line in Fig 10 was derived taking into account evidently false points in the low range of flows Such points may occur for example when the river flow is artificially stopped by freezing to warming is evaluated from uniform amounts of precipitation and snowmelt dept
221. t Felsberg 3250 km 560 3614 m a s l Baumgartner 1987 Black is snow free area gray is cloud covered area and white is snow covered area 28 OVERFLIGHT 80 ERAGE a 40 A t a SS MONITORING INTERVALS SNOW COV gt SEES SS 20 YU NONEXISTENT LY SNOW COVER Fig 5 Example of a possible distortion of a depletion curve due to a temporary increase in the snow coverage by a summer snowfall and to missing Landsat data from the preceding overflight Hall amp Martinec 1985 100 OU DU S AD 20 Fig 6 Depletion curves of the snow coverage for 5 elevation zones of the basin Felsberg derived from the Landsat imagery shown in Figure 4 A 560 1100 m a s l B 1100 1600 m a s l C 1600 2100 m a s l D 2100 2600 m a s l E 2600 3600 m a s l Baumgartner 1987 29 As a result excessive meltwater production would have been calculated To avoid such errors satellite images showing the short lived snow cover from the summer snowfalls must be disregarded when deriving the depletion curves In order to identify the new snow events coincident precipitation and temperature data should be consulted The transitory new snow is accounted for as stored precipitation eventually contributing to runoff as explained in the Section 5 2 2 As an example Figure 6 shows depletion curves of the snow coverage derived for five elevation zones of the alpine basin Fels
222. t by proportions of the daily inputs Martinec amp Rango 1986 41 If the hydrographs are not available or if their shape is distorted by reservoir operations the time lag can be estimated according to the basin size and by analogy with other comparable basins Generally the time lag in a basin increases as the snow line retreats In the WMO intercomparison test WMO 1986 most models calibrated the time lag However these results appear to be of little help to determine the proper values Contradictory time lags have been calibrated by different models However if the time lags for all models participating in the WMO intercomparison test are averaged for each basin the resulting values support the expected relation between L and basin size Basin W 3 8 42 km 3 0h Dischma 43 3 km 7 2 h Dunajec 680 km 10 5 h Durance 2170 km 12 4 h If there is some uncertainty L percentages in Equations 17 20 can be adjusted in order to improve the synchronization of the simulated and measured peaks of average daily flows It should be noted that a similar effect results from an adjustment of the recession coefficient 42 6 ASSESSMENT OF THE MODEL ACCURACY 6 1 Accuracy criteria The SRM computer program includes a graphical display of the computed hydrograph and of the measured runoff A visual inspection shows at first glance whether the simulation is successful or not SRM uses two well established accuracy cri
223. tabase by importing a DOS SRM datatle 128 Example 2 Adding physical data to a database from another digital source 129 Example 3 Developing a climate change scCenario 130 Example 4 Simulating climatechange using a climate change ecenarig 131 Example 5 Developing a custom climate change scenario 132 Example 6 Developing a normalized year ccccecececscseseeeeueueueeeseseeeeaeauaueueeterstsneeeeearananaes 134 AD OEN GIB Grapnical POS sail 138 Appendix C ee ie le inn VV INGO WS EEN 139 Appendix D WINSRM Database Schema ccccecececseseceeueeeeeueeeeeeeeeeeeaeaeauaueueeesteterseeeeeeeneanananeenens 140 BasiNDESCAPUON Ee EE 140 ER ge elen WEE 140 lant et Be Le ee E 141 PrysicalData Ee E 142 Simulation LETTRE 143 SimMResults RE 144 Climatescenarionaex Eeer EE Or ne 146 Cimatescenano Ee EE 147 APpendix E AGCICIO Mal tele VE 148 Appendix F Program Message Dialog ccccccecsececeeeeeeeeeee eset sees esse ee eeee eee ee esse ea eeeeeeaeaenerennegenenas 149 Appendix G SRM Computer Program Version A 150 Gal Te elen ET 150 EE ER lte State ni oa 150 G2 1 SYSTEM FEQUIFEMENUS Ee DEE 150 6 22 Kure ale MITO SRM ET 150 6 23 G ngur Microsd 151 6 24 OPDera e e Be geed TEE 151 Go Program TALUS TE 152 6 371 Screen arsplay EE 152 EE E TEX SC CENS E 153 EE Ee 153 6 3 4 _Dalaceniry TEE 154 Gro EE EIER ODON aer a Deet 156 326 Basi ele e TEE 156 EE Dar Basin Variables parameters iaa 156 G
224. table in its customized form there is only one SimResults table per database The table 1s recreated each time a simulation 1s performed 134 13 Click Run Climate Change The Run a Climate Change window appears Select Manual from the Scenario Definitions s drop down list the model uses the winter end date found in this scenario 14 Click the Use the climate changes defined in Results checkbox 15 Click on Step1 Winter Present Climate Because the checkbox Step 14 is checked the model skips the preliminary step that recreates the SimResults table Instead the table used 1s the one we created and edited in steps 4 10 16 Complete the climate change simulation by following Example 4 Steps 6 10 Example 6 Developing a normalized year This example illustrates the steps required to create a normalized year of temperature precipitation and snow depletion data using as a source a year round simulation selected from a WinSRM basin database See Section 8 5 to learn more about normalized data and how it is used in simulating climate change See Section 10 3 8 to learn how WinSRM derives long term monthly averages for temperature and precipitation and how it uses these long term averages to adjust corresponding observed daily values during climate change processing resulting in a set of normalized temperature and precipitation for the given year and the resulting snow depletion curve s deri
225. tation of daily flows is extended to 10 years without updating Martinec amp Rango 1986 6 2 Elimination of possible errors It is not possible to give threshold values of accuracy criteria which would determine whether a model run is successful or whether something must be changed With good data a value like R 0 80 might still be improved In unfavorable conditions with incomplete data a user may be satisfied even with lower R values Sometimes however the graphical display as well as the numerical criteria indicate that something went definitely wrong Before adjusting one or the other parameter it is recommended to check several probable sources of error first Examples in Table 5 are based on the actual experience of various users Runoff simulation went too high Dinwoody 1976 Dischma 1977 Runoff simulation suddenly deteriorated Dischma The simulated runoff hydrograph declined uncontrollably in September Illecillewaet basin Runoff simulation far below the measured runoff I llecillewaet After the start of snowmelt the simulated runoff kept decreasing in spite of snowmelt input small tributary basins of upper Rhine 1985 Frequent deviations from the measured runoff periodical lows of the hydrograph not simulated Felsberg 1982 Rainfall peaks inadequately simulated Illecillewaet Rainfall peak inadequately simulated even with the special rainfall peak program small tributary basins of u
226. te The climate part of the program can thus transform snow conditions CDC s and runoff of a year to CDC s and runoff of another year by using precipitation and temperature data of that year Figure 39 shows the measured depletion curves of the snow coverage in the Rio Grande basin near Del Norte in the extremely wet year 1979 and in the extremely dry year 1977 The resulting hydrographs are shown in Figure 13 By applying the precipitation and temperatures of 1977 to the year 1979 as a new climate 1t 1s possible to derive the CDC s and the runoff in 1977 as illustrated in Figures 40 and 41 The acceptable agreement between the originally simulated hydrograph of 1977 and the climate affected runoff of 1979 is the first available test of the climate part of a program for a hydrological model So far such test would be only possible by waiting for a climate scenario to materialize In the described example the climate scenario consisted of precipitation and temperatures which really occurred in the year 1977 Apart from this reassurance that the evaluations of the climate effect are fairly realistic this new capability provided by the WinSRM program improves the real time runoff forecasts The measured and predicted precipitation and temperatures in a current year can be considered as a changed climate for a selected historic year or for a normalized year as will be explained in Section 8 6 73 Snow cover APR MAY
227. te changed snow cover data The program s progress through these steps is monitored using another new screen display the Climate Change Progress Screen The steps defined by that screen are Step 1 Winter simulation normal climate determine total zonal melt e Run winter simulation with the normal climate Compute Xwinter_zonal_meltyormaL Where zonal melt 1 is M S melt from snowpack M 1 S melt from newsnow contributing P Cpr where M a T Cpr contributing precipitation 1 e that precipitation falling as rain that fell on the snowftree area a ripe snowpack or a combination e Determine 2Precipitation P Step 2 Winter simulation changed climate determine total zonal melt e Run winter half year simulation with the changed climate Compute winter zonal meltcrim 1 gt e Determine 2Precipitationc im P Step 3 Summer simulation normal climate save MDC s simulated runoff 17 2 e Run summer simulation with the normal climate save data needed to recreate modified depletion curves MDCexcL accumulated zonal melt curves AZMexc and the simulated runoff hydrograph Qsim MDCrexcL snow depletion vs Z a T newmelt AZMexcy X a T newmelt S vs time Step 4 Summer simulation changed climate save MDC om e Run summer simulation with the changed climate Save X a T ctm 2 newmelt crm Step 5 Derive CDCeLim wa using MG cm adjusted for Winter Change Sa Compute AHy Winter Change
228. te function key definitions that supersede the primary function key definitions during basin variable parameter data entry data display These alternate functions provide some additional tools useful in manipulating the complex structures required to store multiple day multiple zone data These functions are F5 Prior Zone F6 Next Zone As explained before these keys change the Variable Parameter data entry screen display to the same two month period for the next or prior elevation zone These keys are hot only for multi zone basins and only for zone specific variables and parameters F7 Duplicate Existing Zone When invoked this function prompts the user for a source and destination zone then copies all the daily values from the source zone to corresponding days in the destination zone This is particularly useful for some of the model s daily parameters which may change very little from zone to zone 167 F8 Repeat Prior Value Pressing F8 causes the value for the day preceding that marked by the cursor to be copied into the current day position The cursor then advances to the next day If the cursor is on the first day of the earlier of the two months displayed the prior day value used is out of sight that is the last day of month preceding the month displayed F9 Repeat Prior Value N Times After prompting the user for a repeat count the model enters the value from the day preceding the cursor into the next N days moving
229. tellite snow cover monitoring In Remote Sensing and Geographic Information Systems for Design and Operation of Water Resources Systems ed by M F Baumgartner G A Schultz A Johnson Proc 5 Scientific Assembly of AHS Rabat Morocco IAHS Publication no 242 83 91 Rango A amp van Katwijk V 1990 Development and testing of a snowmelt runoff forecasting technique Wat Resour Bull 26 1 135 144 Shafer B A Jones E B amp Frick D M 1981 Snowmelt Runoff Simulations Using the Martinec Rango Model on the South Fork Rio Grande and Conejos River in Colorado AgR STARS Report CP G1 04072 Goddard Space Flight Center Greenbelt MD USA Soil Conservation Service 1986 Urban Hydrology for Small Watersheds Tech Release 55 2nd Edition Wilson W T 1941 An outline of the thermodynamics of snow melt 7rans Am Geophys Union Part 1 182 195 WMO 1986 Intercomparison of Models of Snowmelt Runoff Operational Hydrol Report 23 WMO Geneva Switzerland WMO 1992 Simulated real time intercomparison of hydrological models Operational Hydrol Report 38 WMO Geneva Switzerland 121 11 2 Specific references for Table 1 1 Aschenbrenner B 1998 Snowmelt Runoff Model applications in the Otztal Austria Technical Report NN 1998 Presentation at the Fourth Snowmelt Runoff Model Workshop held at the University of Berne Switzerland 2 Azzouz A 1989 Martinec Rango Snowmelt Runoff Mo
230. ter storage in mountain basins from satellite snow cover monitoring 5th Scientific Assembly of the International Association of Hydrological Sciences Rabat Morocco pp 83 91 IAHS Publ No 242 April 1997 53 Rango A Martinec J 1994 Model accuracy in snowmelt runoff forecasts extending from 1 to 20 days AWRA Water Resources Bulletin Vol 30 3 pp 463 470 54 Rango A Roberts R amp Martinec J 1988 WMO Project on simulated real time intercomparison of hydrological models Technical report Hydrological Laboratory USDA Beltsville Maryland USA 1988 Individual report on SRM 55 Rango A 1985 Runoff Boise Unpublished Data U S Department of Agriculture Hydrology Laboratory Beltsville Maryland USA 1985 56 Rango A 1984 Runoff Madison Montana USA File Report 1984 57 Rango A amp Martinec J 1981 Accuracy of Snowmelt Runoff Simulation Nordic Hydrology Vol 12 pp 265 274 58 Rognes A 1992 Verification of SRM for Norwegian mountainous areas with snow cover input from NOAA AVHRR Technical report NN 1992 Presentation at the 1 Snowmelr Runoff Model Workshop at the University of Berne Switzerland 59 Rott H Nagler Th Dinning G Wright G Miller D Gould J Zaves R Ferguson R Quegan Sh Turpin O Clark Ch Johansson B Gyllander Q Baumgartner M Kleindienst H Voigt Dt amp Pirker O 2000 Hydrology of Alpine and High Lat
231. teria namely the coefficient of determination R and the volume difference D for a more objective assessment of how well the simulation has been carried out The coefficient of determination is computed as follows gt Q E OI R 1 5 21 gt Q E Ou i l where Q isthe measured daily discharge Q is the computed daily discharge Q is the average measured discharge of the given year or snowmelt season n is the number of daily discharge values Equation 21 also corresponds to the Nash Sutcliffe coefficient in which case Q is a long term average measured discharge applied to the respective years or seasons The deviation of the runoff volumes D is computed as follows D Ao 100 22 R where Vp is the measured yearly or seasonal runoff volume Vg is the computed yearly or seasonal runoff volume Numerical accuracy criteria are never perfect as illustrated by Figure 13 From the visual judgment both simulations look good because the fundamental difference between two extreme years is well reproduced However R 0 95 in 1979 while it amounts only to 0 48 in 1977 In spite of this unfavorable value the simulation or forecast in 1977 would certainly be useful for water management because it correctly reveals an extremely low runoff For the year 1977 with Q as a long term average substituted into Equation 21 Nash Sutcliffe instead of the average for the specific year a much more favor
232. the forward progression An or 4 moves the cursor to the next higher lower field relative to the current field G 4 3 Field editing keys Micro SRM uses type over as its standard data entry mode This means that any valid data character typed through the keyboard will replace that marked by the blinking cursor with the cursor advancing to the next character position in the field The model supports word processor like editing via the following keys Insert The Insert key toggles the data entry mode to insert with any typed characters being inserted at the point in the field marked by the reverse video cursor and all existing characters shifted rightward Insert mode is canceled by pressing Insert again or moving the cursor to another data field or screen 163 Delete The Delete key is logically the opposite of Insert The character marked by the cursor is deleted and all characters to the right of the cursor are shifted left one position Backspace The Backspace key deletes the character immediately to the left of the cursor with all characters at and right of the cursor shifted left one position CapsLock When CapsLock is set indicator on screen line 24 is visible any alphabetic character entered will appear in upper case Press CapsLock again to turn off NumLock When NumLock is set indicator is visible characters entered through the keyboard s number pad are treated as numbers WARNING If NumLock
233. tion Control Information dialog where an initial set of simulation parameters is described Once a simulation parameter table has been defined and indexed in the SimulationIndex table the new database can be Saved by clicking File Save As 10 4 1 2 Import an existing DOS SRM data file The simplest way to populate a WinSRM basin database is to import an existing SRM data file created by an earlier DOS version of the SRM model software When an existing SRM file is imported into a new empty database the basin zone and physical variable database tables described in the next section are initialized using imported values and a simulation parameter table for the simulation is created using the imported simulation parameters After the import finishes the database is adequately populated to support model processing of that simulation 117 Importing a DOS SRM file into an existing WinSRM database 1s identical to that for a new database with the follow exceptions e The program checks for inconsistencies between the basin database and the basin defined in the SRM file Units of measurement and the number of zones must be identical e Since the Basin and Zone data tables in the existing database are populated the program asks the user whether to overwrite that information using similar values present m each SRM file Appendix F Message 3 eSince WinSRM maintains a single set of physical variable data imported simulation files
234. tivities The success of the exercise can be verified by running the pre existing scenario that is already resident it the WinSRM database that you will be using l Start the WinSRM program Load the Rio Grande database that is included with the software by clicking File Open In the Open an SRM database dialog double click the entry RioGrand mdb Control will return to the main window populated with the new database information In the Basin Simulation grid click on the second simulation title RGWY79 making it the current simulation 131 3 Click on Data Climate Change Scenario Definition The Climate Change Scenario Definition for Simulation RGWY79 window appears See Window 16 Click New The Create a new climate change scenario dialog appears Enter the name Ex3 for your new climate change scenario 4 The definition dialog is initialized to an empty new state Notice the simulation period cannot be changed and that it is one year in length WinSRM performs climate change over a one year period only Notice also that the Winter End Date defaults to the halfway point in the year This value may be modified by the user 5 Begin creating your set of rules that will define your changed climate You estimate that temperature will be warmer by 4 C throughout the entire year in your changed climate a In Row 1 click the Variable Parameter drop down list and select Average Temperature b Do
235. to your PhysicalData table 4 Click on Data Variables Period of record to display the Edit basin variables window see Window 14 Experiment adding data to the data base by first typing values into the cells for several days Then sample the Copy Paste functionality with the following steps a Click on the data cell for runoff April 1 1975 Holding the left mouse button down drag the mouse down to April 15 An alternative to dragging 1s to click on April 1 press the keyboard Shift key then click on April 15 The block selected is highlighted after this process b Right click on the block selected A popup menu appears Click the Copy menu choice The keyboard alternative is to press and hold the Ctrl key then press C Ctrl C This copies the highlighted data to the Windows clipboard c Click on the data cell for runoff April 1 1981 Press the right mouse button then click the popup menu item_Paste to copy the data from the clipboard to the grid The keyboard equivalent is Ctrl V This technique described in step 4 cut and paste is very common in the Windows environment Typically a user will import digital information into a spreadsheet program then use cut and paste to move the information into corresponding WinSRM locations Example 3 Developing a climate change scenario This exercise results in a climate change scenario that the model developers have created during climate change research ac
236. tory Use the A EGA CFG configuration file Reroute any printed output to a disk file A SRMPRINT PRT C gt SRM gt SRM4 TOUT trace fl From the SRM subdirectory run SRM using the default config file and capture all keypresses during the run storing then in a trace file named C SRM TRACE FL C gt C SRM SRM4 C SRM SRM CFG TIN C SRM TRACE FL TDELA Y 2 Run SRM from the root directory using full pathnames for all file references Drive the model with keypresses stored in C SRM TRACE FL with a 2 second delay between commands C gt MSHERC SRM4 HERC CFG Load Microsoft s Hercules graphics driver and run SRM in the Hercules graphics mode HINT Create a batch file to run Micro SRM Store the batch file on the root directory of your hard disk and include the root directory in your PATH statement refer to your DOS manual for a description of PATH To create a batch BAT file type the following commands each terminated by Enter lt 2 C gt cd Make root dir the current dir C gt copy con SRM BAT Copy from keyboard to SRM BAT cd srm Line 1 change to model dir srm4 Line 2 run model cd Line 3 change to root dir lt F6 gt Press Function Key F6 then Enter lt to terminate copy con To run Micro SRM using a batch file simply type the batch file name followed by Enter lt 4 G 3 Program features G 3 1 Screen display types The model user interacts with Micro SRM through a combination of text data
237. ture and precipitation forecasts from the Swiss Meteorological Office the daily runoff was forecasted always for four subsequent days The runoff volume from April through September was forecasted and updated with the use of modified depletion curves Martinec amp Rango 1995 The climate part of the WinSRM program facilitates the real time runoff forecasts as explained in Section 8 6 7 2 Updating The model performance in the forecasting mode is naturally affected by the reduced accuracy and reliability of temperature and precipitation forecasts The propagation of errors can be avoided by periodic updating In the more recent versions of the computer program Versions 3 11 and 3 2 the computed discharge can be replaced every 1 9 days by the measured discharge which becomes known for the corresponding day so that each subsequent forecast period is computed by using a correct discharge value 56 y H SS a F E E a F gt AUGUST 1976 AUGUST 1976 Fig 24 Discharge simulation in the Dinwoody Creek basin 228 km 1981 4202 m a s 1 in Wyoming a without updating and b with updating by actual discharge on 1 August Even without this updating SRM prevents persistent large errors by a built in self adjusting feature which is efficient if Equation 7 is carefully assessed Figure 24a shows a model runoff simulation starting with computed discharge of only one half of the correct value Updating by actual discharge improves the
238. ture forecasts total 30 cm the snow coverage will drop to 55 The snow covered areas thus extrapolated are used for real time forecasts of daily flows The modified curve also indicates the water equivalent of snow 37 cm at the start of the snowmelt season for seasonal runoff forecasts If in another future year the cumulative computed snowmelt depth coincides for example with the snow coverage of only 36 the curve labeled 13 cm is valid The appropriate curve can thus be identified but with a certain time delay If the initial water equivalent of the snow cover can be evaluated from point measurements the proper curve can be selected at the start of the snowmelt season with no time delay The computer program also provides an option for plotting a modified depletion curve in which the total melt depth includes new snow that falls occasionally during the snowmelt period It appears in Figure 20 as the dashed line equidistant from the new snow excluded modified depletion curve While the new snow excluded MDC 1s used to evaluate the water equivalent of the seasonal snow cover at the start of the snowmelt period the new snow included MDC can be used to evaluate the shifting of the conventional depletion curves by changed temperatures as will be explained in the Section 8 The depletion curves of snow covered areas are dealt with in more detail in an earlier publication Hall amp Martinec 1985 Figure 22 shows a simulated runoff
239. uble click the starting date cell to set the starting date for this change to the simulation starting date Double click the ending date cell to set the change s ending date to the simulation ending date c Click on the Edit Action drop down list and select Add to each day Type 4 into the Edit Factor cell d Select Al from the Zone drop down list meaning this rule will apply to temperatures in every zone e In Row 2 select Degree Day Factor from the Variable Parameter drop down f Set the date period in row 2 by repeating step b g Select Shift Backward from the Edit Action drop down the increased temperature of the changed climate necessitates utilization of the warmer month s DD factors earlier in the season than would occur in the present climate h Enter 31 as the edit factor for this rule line When used with shifts the edit factor is the number of days to shift the parameter e g a shift backward of 15 days moves 4 16 to 4 1 4 17 to 4 2 etc 1 For Row 3 select Snow Runoff Coefficients from the Variable Parameter drop down Complete the line by repeating steps f g and h 6 Click Done to close the Climate Change Scenario Definition dialog and return control to the main window To preserve the newly created scenario click File Save Example 4 Simulating climate change using a climate change scenario This exercise uses the climate change scenario Ex3 developed in the previous
240. umulative daily snowmelt depths as computed by the model Consequently if SRM is run in a whole hydrological year the derivation of MDC from CDC starts with the summer half year and not earlier The decline of the modified depletion curves depends on the initial accumulation of snow and not on the climatic conditions as is the case with the conventional depletion curve 60 MEASURED COMPUTED 50 40 30 Q Im s 20 10 Fig 17 Runoff simulation in the catchment area of the hydroelectric station Sedrun Swiss Alps 108 km 1840 3210 m a s 1 Baumann et al 1990 51 e COMPUTED D 3 09 R 0 82 IW hat E NE W Te E Fig 18 Runoff simulation in the catchment area of the hydroelectric station Tavanasa Swiss Alps 215 km 1277 3210 m a s 1 Baumann et al 1990 DER Procedure for weekly forecasts of daily flows Assumption A family of modified depletion curves has been derived from the past snow cover monitoring and temperature measurements in the given basin Two of these curves representing the initial water equivalents Hw 20 cm and Hw 60 cm are plotted in Figure 19 Example 1 Snow accumulation in the basin unknown snow coverage measured by satellite on 15 May S 80 cumulative snowmelt depth from degree days and degree day ratios to date 30 cm Temperature forecast 30 degree days for the next week converted to meltwater depth M 15 cm by a degree day ratio
241. und simulations in No Snow mode t Clear Recent Files Window 4 WinSRM Main Window Options Menu 90 Pull down Choice Explanation Item Data When checked the model runs only year round Run year round simulations simulations and forecasts Snow is disregarded with in No Snow mode all melt coming from precipitation stored during the winter months for later melting as degree days this option is a research tool become available In normal mode this stored Later versions of WinSRM _ snow is set to zero when snow covered area reaches may not support no snow 100 or on the first day of the melt season whichever occurs first Cisse Recent rules Clear the list of recently opened files at the bottom of the File menu The Data menu pull down provides access to functionality for creating editing and displaying simulation parameter data basin variable data and climate change scenarios tables of rules for altering the variables from a current climate to approximate some proposed future climate WINSRM Snowmelt Runoff Model for Windows Fie Options Data Run Tools Graphics Reports Help Climate Change Scenario Definition Basin Defi Hame h Variables d Parameters Units of A BS SE Latitu Window 5 WinSRM Main Window Data Menu Pull down Choice Explanation Item Run Climate Change e Open window that manages climate change scenarios Scenario Definition l Open submenu
242. urface conditions within the zone Cs can be provided for each zone gt S WN 3 oras a basinwide parameter See SimulationIndex table water volume The rain runoff coefficient see Section 5 3 1 takes care of losses due to C Sinale surface conditions within the zone Cs can be provided for each zone or 9 as a basinwide parameter See Simulation ndex table water volume snowpack rain is retained by the snow 1 ripe snowpack rain from the entire zone area contributes to runoff Xcoeff X value for computing the recession coefficient See Section 5 3 6 Ycoeff Y value for computing the recession coefficient See Section 5 3 6 1 required 2 either Tmax Tmin or Tavg are required Rainfall Contributing Area see Section 5 3 5 is a switch that determines RCA Single how the model responds to rain falling on a zone s snow pack 0 dry 144 SimResults Table The SimResults table occurs once per database With one exception see 9 3 7 4 and Appendix A Example 5 this table is recreated each time a simulation is performed The records in the table represent one day for each zone for the simulation time period The table is initialized using corresponding information from the PhysicalData table and the simulation s simulation parameter table For climate change simulations values representing the changed climate are also included in separate fields within each record postfix CC The SimResults table i
243. used to specify what physical data to load Variables l into the data entry display window Open the data entry display window loaded with the Parameters f parameter data for the current simulation Open the data entry display window loaded with the Results S qe results table for the current simulation The Run menu pull down Window 6 directs the user to the appropriate processing path for the task desired Choosing Simulation Melt season or Simulation Year round will initiate a traditional SRM simulation The two are essentially identical The only distinction between the two is in the graphics plots available after completion of the simulation Choosing Melt season will enable plots 13 through 16 91 Appendix B which only have relevance during the melting season Two similar forecast processing modes are available under the Forecast menu In Forecast mode the model simulation is updated using measured runoff every n day n is a user defined value between 1 and 9 Selecting Climate Change displays another window the Run a Climate Change window Window 20 that directs the multiple user interactions required to perform a climate change simulation WINSRM Snowmelt Runoff Model for Windows Fil Options Data Run Tools Graphics Reports Help Simulation Melt Season Simulation Bazin Definition Forecast Year Round Simulation Hame pio Climate Change Window 6
244. ut which accelerates the basin response Program Versions 2 01 and later versions take this feature automatically into account whenever rainfall exceeds a preselected threshold Temperature extrapolated from station 800 m a s I to zone D 2380 mae by 0 65 C 100 m gt T 0 43 C while Terit is 0 75 C By decreasing Ten snowfall was converted to rainfall and the runoff peak was better simulated Only 1 precipitation station was available but the by zone option was switched on Consequently precipitation input took place only in 1 zone With 1 station select option basin wide 48 Table 5 cont 10 Runoff simulation went very high in a basin with Compact snow cover was assumed above the snow line so a large elevation range of 7400 m Kabul River 1976 Distorted runoff simulation outflow from a snow lysimeter Difficulties with the timing of rainfall runoff peaks WMO test for simulated operational forecasts lllecillewaet basin Forecasted runoff for hydroelectric stations Sedrun and Tavanasa too low in April J une 1994 In one run snow cover was not completely melted by the end of September Discrepancies in runoff simulations for Rio Grande near Del Norte Discrepancies in runoff simulations for Rio Grande near Del Norte Effect of climate change see Section 8 3 Decrease of runoff computed by an increased temperature Rio Grande near Del Norte Overestimation of runoff in the Rhine Felsberg basi
245. ved using the normalized climate The example assumes that multiple years of physical data are present in the current basin database An alternative manual method for normalizing your data replacing Step 4 1s described at the end of this example The alternative method demonstrates how long term averages derived externally from other sources are applied to the selected simulation year 1 Load an existing WinSRM database Click File Open or File followed by selecting a WinSRM database from the Recent Files list 2 Save the WinSRM database to a new mirrored database file File Save As After this step the database displayed by the model screens is the new mirror database 3 Select a simulation from the Basin Simulations grid The selected simulation must be a year round simulation for the year you want to normalize 4 Create a climate change scenario with temperature and or precipitation modified using the Normalize climate change function Click Data Climate Change Scenario Definition Click New Type in a new scenario name Click Ok Define a modifying rule for temperature A Select temperature from variable parameter dropdown menu B Identify the starting and ending dates winter starting date through summer ending date C Select Normalize from the edit action dropdown menu D Type 0 zero in the edit factor cell E Select All from the zone dropdown if temperature selected in sub
246. w Srow M 1 ontributing Ham Cpr IT Measured vs Computed Snownmelt Bum Climate Change Debot Gain Summary Disabled selection Print To E Printers File Report Viewer Help All print close Window 29 Output Definition Window 112 10 3 12 3 Buttons Click Close to close the Output Definition window and return to the WinSRM Main window This button caption toggles between View and Print depending on which option button is selected either Printer File or Report Viewer If there are reports not checked the button caption is All Click All to check all the reports When all the reports are checked the caption toggles to Reset which will clear all the check boxes Help Click Help to display the help text for the Output Definition Window Close Print Els EE 10 3 12 4 Available reports Available reports shown in the Output Definition window are shown in the following summary Report Name Description Current data description Lists the current user input data A listing of the observed temperature data used with the current Temperature values simulation The type of temperature information average max min 1s determined by the simulation s temperature type control value Degree day factors Parameters for the current simulation This report is for one zone or all Runoff coefficients zones depending on the parameters distribution modifiers Lapse Rate
247. xisting prior to the help request When help is requested from within another function SUMMARY FILEIO PLOT PRINT it is presented slightly differently through the use of a pop up window a transient area nondestructively superimposed over the current display image These function pop ups are only accessible after invoking the specific function F2 Summary This function is invoked by pressing clicking function key F2 from the Program Options Basin Definition or Variable Parameter display screens The Micro SRM summary screen provides the user with a snapshot of basin definition variables and zone specific parameter values as they are currently defined Beginning and mid month value day and day 16 if different for each month in the snowmelt period are displayed for seven model parameters Each zone in the basin is displayed separately and is accessed via the PgUp and PgDn keys A printed version of the summary may be selected through the Print Menu see below The following keys are hot within the summary function Fl Help info on this function PgUp PgDn Display the next prior zone s configuration ESCape Exit function and return to point where invoke 164 F3 FileIO The File Input Output function is invoked by pressing clicking function key F3 from Program Options Basin Definition or Variable Parameter display screens In response to the hot key a small pop up menu is displayed The user specifies one of four sub opti
248. year Winter October March in the Northern hemisphere 6 Estimate S in the selected year Syear as described in Section 8 1 7 Simulate the runoff with Syear Tyear and Ben to verify whether the estimated S and model parameters are realistic 8 Adjust Syear to AT to obtain the normalized Snorm 71 9 Run SRM with Prom Trorm and Snorm to obtain the normalized winter runoff representing the present climate 10 Adjust Snorm to Teri to obtain Sc m see Section 8 1 11 Run SRM with Perm Lem Scum to obtain the climate affected runoff The difference between the Norm run and the Clim run is the effect of the climate change on the winter runoff Summer April September in the Northern hemisphere 12 Simulate runoff with Syear Tyear and Pyear to verify whether the model parameters are realistic in summer Syear 1s obtained from satellite data 13 Run the SRM climate program see Section 8 4 with Thom Phorm in other words consider Thorms Pnorm as a new climate to derive Snormwa CD aen wa taking into account AHy obtained from winter runs 14 Continue climate run with Phorm Tnorm and Snorm to obtain the normalized summer runoff 15 Run the SRM climate program with Perm Ten to derive Serm CDCcim 16 Continue the climate run with Perm Term and Scum to obtain the climate affected runoff As mentioned for the winter half of the year the difference between the Norm run
249. year in question In the starting year the depletion curves of the snow coverage and the glacier area must be known apart from temperature and precipitation data The glacier melt volume and the carry over of the unmelted snow if any to the next year are computed From the glacier melt volume the reduced glacier area for the next year is estimated by statistically derived relations between the volume and area of glaciers Bahr et al 1997 With this new area together with temperature and precipitation data we can evaluate next year s CDC s the glacier melt volume and carry over of unmelted snow to the next year These computations are repeated year by year until the year for which the climate scenario 1s predicted In a cold year the carry over of unmelted snow will increase and the CDC s will probably not decline to the glacier area signaling no glacier melt Should the stochastic series indicate a sequence of cold years the amount of unmelted snow by the end of each hydrological year will be built up It cannot be determined which part of this snow will become part of a glacier However sooner or later this snow will be melted the CDC s will decrease to the last derived glacier area and the glacier melt will start again Each year the glacier melt volume will result in a reduced glacier area for the next year until the ultimate disappearance of glaciers Whether this will happen before or after the target year depends on the int

Snowmelt Runoff Model (SRM) User`s Manual

Contents

Download Pdf Manuals

Related Search

Related Contents