Home

User`s Manual - Agricultural and Biological Engineering

1. Spring Cover Soil loss ratio for cropstage period and canopy cover Cover Crop Sequence Resi After No and manmagement due Plant F SB 1 2 3 80 90 96 4Lf LB WC succulent blades 95 No till pl in killed WC 3000 11 11 17 23 18 16 96 2000 15 15 20 25 20 17 97 1500 20 20 23 26 21 18 98 1000 26 26 27 27 22 19 99 Strip till 0 25 row sp 3000 18 18 21 25 20 17 100 2000 21 21 25 27 21 18 101 1500 28 28 28 28 22 19 102 1000 33 33 31 29 23 20 CORN in Sodbased systems No till pl in killed sod 103 3 5 tons hay yld 1 1 1 1 1 1 104 1 2 tons hay yld 2 2 2 2 2 2 2 Strip till 3 5 ton hay 105 50 cover till strips 2 2 2 2 2 4 106 20 cover till strips 3 3 3 3 3 5 Strip till 1 2 ton hay 107 40 cover till strips 4 4 4 4 4 4 4 108 20 cover till strips 5 gt 5 3 5 5 5 Other tillage after sod n CORN after Soybeans 109 Sprg TP conv till HP 40 72 60 48 25 29 110 GP 47 78 65 51 30 25 37 111 FP 56 83 70 54 40 31 26 44 112 Fall TP conv till HP 47 75 60 48 25 113 GP 53 81 65 51 30 25 114 FP 62 86 70 54 40 31 26 115 Fall amp Sprg chisel or cult HP 30 40 35 29 23 29 116 GP 25 45 39 33 27 23 37 117 GP 20 51 44 39 34 27 23 37 118 FP 15 58 51 44 36 28 23 44 119 LP 10
2. Cover Manning s n range recommended ms 3 Bare sand 0 01 0 013 0 011 Bare clay loam eroded 0 012 0 033 0 02 Fallow no residue 0 006 0 16 0 05 Range natural 0 01 0 32 0 13 Range clipped 0 02 0 24 0 10 Grass bluegrass sod 0 39 0 63 0 45 Short grass prairie 0 10 0 20 0 15 Dense grass 0 17 0 30 0 24 Bermuda grass 0 30 0 48 0 41 aWeeping lovegrass bluegrass buffalo grass blue gramm grass native grass mix OK alfalfa lespedeza Part IV VFSMOD Appendices 162 3 3 Vegetation types for VFS s The following data on vegetation is taken from Haan et al 1994 Vegetation good stand Density Grass spacing Maximum Modified n stems m7 S cm height H cm Nm ems VEGETATION TYPICALLY RECOMMENDED FOR VFS Yelow bluestem 2700 1 9 Tall fescue 3900 1 63 38 0 012 Blue gramma 3750 1 65 25 0 012 Ryegrass perennial 3900 1 63 18 0 012 Weeping lovegrass 3750 1 65 30 Bermudagrass 5400 1 35 25 0 016 Bahiagrass 20 0 012 Centipedegrass 5400 1 35 15 0 016 Kentucky bluegrass 3750 1 65 20 0 012 Grass mixture 2150 2 15 18 0 012 Buffalograss 4300 1 5 13 0 012 VEGETATION NOT RECOMMENDED FOR VFS Alfalfa 1075 3 02 35 0 0084 Sericea lespedeza 650 3 92 40 0 0084 Common lespedeza 325 5 52 13 0 0084 Sudangrass 110 9 52 0 0084 a To convert densities for good stand to other stands mult
3. Axis Title Time min Runoff m 3 s 10 6 Minimum f 3 02666661 fo Maximum 46 5333333 A 92 Increment 5 5 Preview Done Cancel Once Plot editing is complete the Plot can be Copied to the Clipboard The Plot can then be inserted into another application such as a word processor For example the plot can be inserted in Word by Selecting Edit Paste Special Picture or Device Independent Bitmap If you desire only the data used to create the plot then use Paste In Powerpoint use Paste Special and Device Independent Bitmap The plot is copied to the Clipboard as a Windows Metafile wmf for optimum resolution The plot can also be printed by selecting the Print Plot button The plot can be previewed along with selecting the printer paper size and orientation Part III VFSMOD W WindowsTM User s Manual 118 To see 4 preview Use the Print Preview Button To change printers or printer properties press the Change Printer Button Selected Printer is HP LaserJet 4050 Series PCL 5 Change Printer Print Part HI VFSMOD W WindowsTM User s Manual 119 9 Calibration Mode Requirements Directory structure To run either the Hydrograph Calibration or the Sedigraph Calibration a filter strip project and a set of measured field data hydrograph or sedigraph are required to run the calibration mode The field data text files must be located in the inverse folder within the application di
4. Clay in sediment CL 40 00000A Outputs for pesticide trapping Sabbagh et al 2009 1 325 m3 Runoff inflow 45 053 Kg Sediment inflow 6 846 Phase distribution Fph 66 267 Infiltration dQ 97 536 Sediment reduction dE 44 727 Runoff inflow reduction 74 181 Pesticide reduction Reference Mufioz Carpena R 2012 Continuous simulation components for pesticide environmental assessment with VFSMOD 1 VFS pesticide residue between runoff events Technical Report University of Florida Agricultural Engineering Department URL Accessed June 15 2014 http abe ufl edu carpena vfsmod FOCUS VFSMOD Continuous Sim_ Report PesticideResidue_ MunozCarpena pdf Part III VFSMOD W WindowsTM User s Manual 114 7 Processing and Analysis of VFSMOD Results In this version limited analysis of the output files is available These options are available from the VFSMOD menu s Outputs option Viniiisiovite ee le File Source Area UH Fier Strip VFS Design Sensitivity Uncertainty Calibration Options Window Help Execution Outputs Filter Strip Output Viewer Graph Results Hydrology Sediment Balances Other 27 Currently the output files can be viewed by selecting the VFS Output Viewer menu and the user can create plots of the Hydrology and Sediment Balances for the simulation The user selects a results file for each of these options 5 Output Viewer m x VFSmod aft Ouiput Viewer
5. Save and Close gt a gt Close Help bA gt FIGURE 8 Editing measured data hydrograph for calibration Part III VFSMOD W WindowsTM User s Manual 125 Once both files project and measured data are selected the program displays an interface where the user selects the parameter s to calibrate 34 VFSmod Windows Editor v 4 0 7 File Source_Area_ UH Filter_Strip_ VF5 Design Sensitivity Uncertainty Calibration Options Window Help S Inverse Calibration Hydrograph Project Name sample prj Browse Bee ageph Messorod inverse meas_hyd txt Browse Select parameter s to be changed or calibrated No Change Calibrate New value Min C Vertical Saturated K VKS m s Average suction wetting front SAV m c Saturated water content 0S m3 m3 Initial water content OI m3 m3 Maximum Surface Storage ISM ml Filter fraction where ponding is checked SCHK Fiter width FWIDTH m Fiter length VL m Filter Manning s n IRNA s m 1 3 ENT C Average Filter Slope ISOA m m a Advanced Calibration Settings Run Results CARDINI FIGURE 9 Interface for Hydrograph calibration At least one parameter needs to be selected to run the calibration The interface lets the user to select three options for each see last page No Change Calibrate By default all the options are set to No The Change option lets the user to select a different value of the parameter
6. section for more details We are planning to implement other analysis options in future versions Part III VFSMOD W WindowsTM User s Manual 116 8 Using the Plot Windows The user can make graphs of various input and output parameters Each of the Plot windows offer option buttons to Copy the Plot to the Clipboard Edit the Plot or Print the Plot A plot of a runoff hydrograph is used as example to illustrate the various options 5 Hydrograph es fog ex peevenverscensonousesononscuncenvensenessnsonessossvesvasvaneueseovensevnsonvensanereusonensesncesvenvaneueveesensconsonvenesnessuconssnsssuesvenousevevenvenseonsonvenecseseucevenouescenconvusevevenvenseensenseneey Runoff Imes 10 63 13 0267 19 728 26 4293 33 1307 39 832 46 5333 Time min sessesssensssnssnenssonesnseseaneusnesnenseensensensenssnssouenssonesneensensusssonenscanennessensssussusnseaneenansaneussssnenseunsensensenssnssgusnscsseessunsuneusussuenscaneenssnsenssssensanousscaneaneeneese Copy Plot to Clipboard Print Plot Edit Plot Selecting the Edit Plot button the plot can be customized The Axis titles can be changed along with setting the minimum and maximum values of the range along with how many labels A Title for the Plot can also be added The effect of these changes can be viewed by selecting the Preview Button Part III VFSMOD W WindowsTM User s Manual 117 3 Graph Parameters Pa RoR Ex Graph Title X Axis Y Axis
7. 75 sediment reduction line 0 2 0 1 o 20 40 60 80 100 120 Filter length m Figure 16 Example of filter design results obtained with VFSMOD W Part I VFSMOD W Model Documentation 42 7 Potential Users and Applications of the Modelling System VFSMOD is a research model as such potential users are modelers and scientists involved in studies of sediment and other pollution from various sources and its control with the aim to gain a better understanding of the processes involved for a given scenario Results from this model can be used after calibration and field testing in extrapolation or prediction studies for decision making and design Suwandono et al 1999 Mu oz Carpena and Parsons 2002 Parsons and Mufioz Carpena 2002 An evaluation of the model from the user s perspective following modern criteria can be found in Mu oz Carpena and Parsons 1999 The GUI and integrated design procedures introduced with v2 x and above are intended to help extend the model user base to include others like engineers and environmental and natural resources experts involved in the design implementation and evaluation of VFS without requiring in depth computer knowledge Part I VFSMOD W Model Documentation 43 8 Known Limitations and Applicability of the Models 8 1 Known Limitations of the Model The handling of overland flow as sheet flow could pose problems when a filter is not properly maintained and co
8. both grass understory Good 41 61 71 Poor 67 80 85 Sagebrush with grass understory Fair 51 63 70 Good 35 47 55 Desert shrub major plants include Poor 63 77 85 88 saltbush greasewood creosote F 55 2 81 86 bush blackbrush bursage palo ae 1 verde mesquite and cactus Good 49 68 79 84 a Poor lt 30 ground cover litter grass and brush overstory Fair 30 to 70 ground cover Good gt 70 ground cover b Curve numbers for group A have been developed only for desert shrub 1 Average runoff condition and Ia 0 2S For range in humid regions use Table 2 2c Part IV VFSMOD Appendices 169 3 5 MUSLE Crop factor C Soil loss ratios CFACT to describe the effects of cropping management From 1992 GLEAMS User Manual Knisel et al 1992 Spring Cover Soil loss ratio for cropstage period and canopy cover Cover Crop Sequence Resi After No and manmagement due Plant F SB 1 2 3 80 90 96 4Lf LB Corn after C GS G or COT In Meadowless Systems 1 Rdl sprg TP 4500 31 55 48 38 20 23 2 3400 36 60 52 41 24 20 30 3 2600 43 64 56 43 32 25 21 37 4 2000 51 68 60 45 33 26 22 47 5 Rdl fall TP HP 44 65 53 38 20 6 GP 49 70 57 41 24 20 7 FP 57 74 61 43 32 25 21 8 LP 65 7
9. 37904 Volumetric Water Contents cm 3 cm 3 Source Area Storm Ri Initial Water Content OI 125 Saturated Water Content 0S 39 r Green Ampt Infiltration Parameters Maximum Surface Storage SM m 9 Sediment Transport Fraction of the filter where ponding is checked 0 lt SCHK lt 1 7 Flow through VFS Save Continue Editing Save and Close Close Help Detailed Hydrographs Part HI VFSMOD W WindowsTM User s Manual 101 Parameters in this window correspond to the regular soil input selection vfsmod w ver 5 8 or before Explanation for those parameters is as follow VKS saturated hydraulic conductivity K m s See Soils data Green Ampt parameters on page 161 SAV Green Ampt s average suction at wet front m See Soils data Green Ampt parameters on page 161 OS saturated soil water content 0 m m See Soils data Green Ampt parameters on page 161 Ol initial soil water content 0 m m SM maximum surface storage m SCHK relative distance from the upper filter edge where the check for ponding conditions is made i e 1 end filter 0 5 mid point 0 beginning When the shallow water table option is selected a new part of the soil input selection window is opened that includes selections that are only relevant for this case Notice that SAV and OI inputs from the previous window will be ignored they will be dimmed in the final version since the pro
10. 67 59 48 36 28 23 54 120 No till pl in crop resid HP 40 25 20 19 14 11 26 121 GP 30 33 29 25 22 18 14 33 122 FP 20 44 38 32 27 23 18 40 Part IV VFSMOD Appendices 173 Spring Cover Soil loss ratio for cropstage period and canopy cover Cover Crop Sequence Resi After No and manmagement due Plant F SB 1 2 3 80 90 96 4Lf LB BEANS after Corn 123 Sprng Tp Rdl conv till HP 33 60 52 38 20 17 p 124 GP 39 64 53 41 21 18 125 FP 45 68 60 43 29 22 126 Fall Tp Rdl conv till HP 45 69 57 38 20 17 127 GP 52 73 61 41 21 18 128 FP 59 77 65 43 29 22 Chisel or fld cult q Beans after Beans r GRAIN after C G GS COTS 129 In disked residues 4500 70 12 12 11 7 4 2 t 130 3400 60 16 14 12 7 4 2 131 50 22 18 14 8 5 3 132 40 27 21 16 9 5 3 133 30 32 35 18 9 6 3 134 20 38 30 21 10 6 3 135 Do 2600 40 29 24 19 9 6 3 136 20 43 34 21 11 7 4 137 10 52 39 27 12 7 4 138 Do 2000 30 38 30 21 11 7 4 139 20 46 36 26 12 7 4 140 10 56 43 30 13 8 5 141 In disked stubble Rdr 79 62 42 17 11 6 142 Winter G after fall TP HP 31 55 48 31 12 7 5 RDL 143 GP 36 60 52 33 13 8 5 144 FP 43 64 56 36 14 9 5 145 LP 53 68 60 38 15 10 6 Grain after Summer Fallow 146 With grain residues
11. A 1 05518 kg m 2 Using Rw Williams A 1 19323 kg m 2 based on Gleams Conc 29 18056 g L Using Rm Foster Conc 16 66349 g L Using Rw Williams Conc 18 84364 g L based on Gleams eoo0o ooo e e E o a e E e O e E a A A e a S ee E E e ER Part II VFSMOD and UH User s Manual These results are depicted in the next Figure 8 o q I rainfall 6 2 T S 2 z E 4 41 E ez T oO co 2 6 o 8 0 2 4 6 8 10 12 Time hours 2 5 Tips for running the model Here are some suggestions to running the model and answers to potential problems or questions a The finite element model becomes unstable or blows up This is due to a rapid change in boundary conditions quick slope and or roughness changes along the filter or inputs severe changes in rainfall intensity and or inflow from the adjacent field in your inputs For this type of conditions the kinematic wave formulation leads to a behavior termed kinematic shock The model s Petrov Galerkin PG finite element formulation was developed to improve the quality of the solution for these type of special sharp front problems and generally overcomes the instability problem Mufioz Carpena et al 1993b The time step is calculated based on a target Courant number CR for the simulation ikw file and an estimate of the less favorable conditions maximum incoming hydrograph peak flow and rainfall intensity In a few cases due to the dynamic nature of the problem thi
12. Cr CR 0 8 Order of shape functions NPOL 3 quadratic Petrov Galerkin flag KPG 1 Number of different filter segments NPROP 14 Saturated hydraulic conductivity K VKS 1 33 x105 m s Average suction at the wet front Sg SAV 0 379 m Water content at saturation 0 Os 0 311 Initial water content 0 OI 0 125 Surface storage S SM 0 0 m The flow inputs rainfall and incoming runoff from the field are shown later in the output The surface characteristics of the filter were shown as an example in section 7 1 3 1 6 1 2 Sediment transport files sample igr and sample isd Description symbol INPUT Value Units Sediment inflow concentration C CI 03400 g cm Particle size diameter NPART 4 d DP 0 0300 cm Particle fall velocity NPART 4 V calculated VF 3 0625 cm s Particle weight density NPART 4 y SG 1 6000 g cm of coarse particles d gt 0 0037 cm COARSE 100 0 Porosity of deposited sediment POR 43 4 Filter main slope calculated S SC 0 0564 Part II VFSMOD and UH User s Manual 66 Description symbol INPUT Value Units Filter media spacing S SS 2 20 cm Filter media height H H 15 0 cm Grass modified Manning coefficient VN 0 0120 s cm 3 Manning coefficient for bare soil n gt VN gt 0 04 san Surface changes feedback ICO 1 YES 1 6 2 Outputs 1 6 2 1 Calculated
13. Exposure analysis modeling system User s guide for EXAMS II version 2 94 EPA 600 3 89 084 U S EPA Athens GA 1990 Campolongo F J Cariboni A Saltelli W Schoutens 2005 Enhancing the Morris Method In Hanson K M Hemez F M eds Sensitivity Analysis of Model Output Proceedings of the 4th International Conference on Sensitivity Analysis of Model Out put SAMO 2004 Los Alamos National Laboratory Los Alamos 369 379 Carsel R F Mulkey L A Lorber M N Baskin L B The pesticide root zone model PRZM A procedure for evaluating pesticide leaching threats to groundwater Eco logical Model 1985 30 1 2 49 70 Chu S T 1978 Infiltration during unsteady rain Water Resour Res 14 3 461 466 Cooley K R 1980 Erosivity R for individual design storms In CREAMS A Field Scale Model for Chemicals Runoff and Erosion from Agricultural Management Sys tems Vol III Chapter 2 USDA SEA Conservation Report No 26 pp 386 397 Cukier R I C M Fortuin K E Schuler A G Petschek and J H Schaibly 1973 Study of the sensitivity of coupled reaction systems to uncertainties in rate coefficients I Theory J Comput Phys 59 3873 3878 Dillaha T A J H Sherrard and D Lee 1986 Long Term Effectiveness and Mainte nance of Vegetative Filter Strips VPI VWRRC Bull 153 Virginia Polytechnic Insti tute and State University Blacksburg Dillaha T A R B Reneau S Mostaghimi and D
14. Source Runoff mm 2725 2 20 ANB 2 10 Relalive Sensiliv ily ZSCSSTTAN TT AR fs a 76 78 80 82 84 8 88 90 92 94 Curve Number 2 05 2 00 Curve Number Filter Length 20 m Rain 151 1 mm Base value 75 Range 75 to 95 in increments of 1 Mean Rel Sen 2 122 Std Dev Rel Sen 0 0521 Ov Rel Sen 2 5 Dismiss FIGURE 19 Plot selected Output Relative Sensitivity Part III VFSMOD W WindowsTM User s Manual 137 Output Sensitivity Results This output file contains all of the analyses for each of the above graphs and the statistics This is useful for further analyses using other application packages such as spreadsheets Part III VFSMOD W WindowsTM User s Manual 138 11 Uncertainty Analysis Screens Uncertainty analysis can be done on a number of the input parameters for both the UH model and VFSMOD These are done using base values from a specific UH and VFSMOD project Select the input parameters you would like to analyze and leave the others unchecked For each of the selected input parameters select a probability distribution and specify the parameters to define the distribution For the UH uncertainty analysis Curve Number CN Soil Erodibility Factor K Crop Factor C and the Practice Factor P can be selected The user selects the parameters to consider using the Check boxes and selects the probability distribution Currently the normal log no
15. distributions for the key outputs of interest and assign confidence intervals and other estimates on the final filter strip designs note the program supplies basic statistics and the actual simulated data to allow the users to use other outside analysis tools to complete this analysis users are welcome to contact us for suggestions Part III VFSMOD W WindowsTM User s Manual 89 4 Main Window EN vesmod windows Editor v 2 2 2 l File Source_Area_ UH Filter_Strip_ VFS Design Sensitivity Uncertainty Options Window Help Application of VFSMOD is done via project files The project files consist of a list of filenames identified by keywords indicating the type of input or output file To Open a project file select the File Menu and Open a VFS Project File Z Source_Area_ UH Filter_Strip_ VFS Design Sensitivity Uncerte Scenes Current Filter Strip Project Filter Strip Project From Disk Batch of Filter Strip Projects Open Execution Panel From the main window you can also execute VFSMOD This option is available from the VFSM menu This menu contains submenus for Execution and Analysis Under the Execution submenu the current project a project from disk or multiple projects from disk can be executed The Analysis submenu can be used to view output files from VFSMOD in addition to graphing and comparing some of the outputs generated by VFSMOD Part III VFSMOD W WindowsTM User s Manual 90 Other menu select
16. q mm h 0000 1366 2783 0434 2038 13 3 ITs 0501 7981 8278 24 26 27 26 5 25 23 2s 7677 4836 0228 5477 2776 4442 2627 Part II VFSMOD and UH User s Manual 75 x97 03 10 16 223 6 29 36 42 49 preys 62 68 74 61 87 94 00 07 Pele 20 26 33 239 46 392 lt 98 65 71 78 84 lt 91 IT 04 10 S 23 30 Wwwwwhs nH NNHNHNHNHNHNHNHNHNHNHNHNHNRRHPH HEHE HHH HHH RHE DS O4O593 9 OV ONS OS OVS O42 a g O49 ASO OOS 6S SO OO OL OVO SOS OOS 2627 2298 1981 1686 1417 1180 0972 0795 0645 0519 0415 0330 0261 0205 0161 0125 0097 0075 0058 0044 0034 0026 0020 0015 0011 0009 0006 0005 0004 0003 0002 0002 0001 0001 0001 0000 0000 Time to ponding Duration of rainfall excess Time correction to match hyetograph file sample2 hyt a ee Mm ONM A DO e0o00o0o0oo0oco0 coo C00 00 00 CC CCOCORRRENN WA GN 2 743 File output sample2 hyt SCS 10 MIN HYETOGRAPH No Time hr 0 167 ID 500 667 833 000 167 AND aS WHE RRODWOOO 80 000 RoR RR RY RY bY 9159 5479 2642 1361 2059 4933 0018 7230 6414 PELs 9899 SLI 8800 4790 eL1 579 9024 7004 5415 4171 3201 2449 1868 1421 1078 0815 0615 0463 0348 0261 O19995 0146 0108 0081 0060 0044 0033 0024 h 000 3 257 RD 0 000 440
17. 0 03205 0 0000 0 000 80 0 Sandy loam 60 25 0 02549 0 0325 0 000 98 0 Coarse sandy 60 25 0 01914 0 0325 0 000 160 0 loam Loamy very fine 84 8 0 03726 0 0325 0 025 90 0 sand Loamy fine sand 84 8 0 02301 0 0000 0 025 120 0 Loamy sand 84 8 0 01624 0 0325 0 025 135 0 Loamy coarse 84 8 0 00982 0 0325 0 025 180 0 sand Very fine sand 90 5 0 04401 0 0325 0 050 140 0 Fine sand 90 5 0 02173 0 0000 0 050 160 0 Sand 90 5 0 01481 0 0325 0 050 170 0 Coarse sand 90 5 0 00827 0 0325 0 050 200 0 3 3 2 Modifications to USLE to handle storm events USLE was developed for extended periods for example yearly To attempt to use USLE for storm events others have modified EI to represent a storm event and used this in place of R in the original equation Williams 1975 The erosion index EI is a measure of total raindrop energy of a storm One approach for computing EI is to examine 30 min rainfall intensities and compute erosion indices for these periods referred to as El3 In this approach one sums EI over each rainfall period to obtain a rainfall erosivity factor for the Part I VFSMOD W Model Documentation 27 storm In the CREAMS model Cooley 1980 used e 916 331 log i 33 where e is the energy in ft ton acre in 1 ft ton acre in 26 38 J m or 26 38 N m i is the hourly intensity in in h in h 0 007 mm s and log is base 10 The E e r over the storm where r was the increment of rainfall duri
18. 1 070 6 36 6 6 6 773 33 866 6 290 31 595 2 968 6 36 6 16 6 777 33 884 6 614 30 304 4 742 6 36 6 14 6 775 33 876 5 752 29 073 6 429 6 56 6 2 19 796 98 982 19 622 98 262 1 167 6 56 6 6 19 788 98 938 19 325 97 072 3 627 6 56 6 16 19 795 98 974 19 073 96 101 4 808 6 56 6 14 19 791 98 954 18 829 95 166 6 497 6 66 6 2 27 432 137 162 27 256 136 492 1 134 6 66 6 6 27 431 137 153 26 956 135 4604 3 142 6 66 6 16 27 428 137 1408 26 671 134 387 5 076 6 66 6 14 27 432 137 158 26 421 133 535 6 873 sults Graphs e Plot RDR E 2 0 75 S Plot SDA z 05 r1 0 25 aie E Jee Plot PDR 3 0 Filter Length m Copy Plot to Clipboard Print Plot Edit Plot Each graph can be copied to the Clipboard and pasted as a table with the data or as an image in other applications such as spreadsheets text processors etc See Using the Plot Window section The output file from the design analysis simulations can easily be imported into a spreadsheet for more detailed analysis The file format is comma separated variable csv Part III VFSMOD W WindowsTM User s Manual 144 13 Troubleshooting vfsmod w As you encounter problems you can e mail us for help assistance In most cases you should send us copies of the files giving problems along with a detailed description so we can recreate the problem You c
19. 4 Large aggregate 0 0300 3 0625 1 60 5 Sand 0 0050 0 2 0 0200 3 7431 2 65 6 Silt 2 0 0002 0 005 0 0029 0 0076 2 65 7 User selected DP model SG Part II VFSMOD and UH User s Manual 62 COARSE of particles from incoming sediment with diameter gt 0 0037 cm coarse fraction that will be routed through wedge unit fraction i e 100 1 0 CI incoming flow sediment concentration g cm POR porosity of deposited sediment unit fraction i e 43 4 0 434 DP sediment particle size diameter ds cm read only if NPART 7 SG sediment particle density y g em read only if NPART 7 Note COARSE and DP are related so that their values need to follow the these rules COARSE DP gt 0 5 gt 0 0037 0 5 0 0037 lt 0 5 lt 0 0037 1 4 6 3 File example 4 1 0 0 034 434 0013 2 65 1 4 7 filename iwq water quality transport model 1 4 7 1 Structure of the file IQPRO IKD VKOC VKD OCP CCP IDG NDGDAY DGHALF FC DGPIN DGML DGT DGTHETA 1 1 4 7 2 Definition TIWQPRO Flag for type of water quality problem 1 runs pesticides based on Sabbagh et al 2009 2 runs simple solute transport under construction 3 runs the multi reactive transport under construction IKD Flag for reading VKOC VKD and OCP If IKD 0 then reads Kd if IKD 1 then Koc and OCP are read VKOC Adsorption coefficient L Kg VKD Linear sorption coefficient L Kg
20. 6 VFS Project Window The files used by VFSMOD are identified in the project window There are options to Save the project Edit any of the input files and Browse to Select different input files In addition any of the input or output filenames can be changed from this window Other options buttons provide shortcuts to the VFSMOD menu entry these include the buttons Run This Project executes the current project Graph a Sediment Runoff Balance produces bar graphs comparing sediment and runoff in and out of the filter strips the user selects the output summary file View Output Files opens a text window with a user selected output file Changes had been made to the current Windows GUI to incorporate the new WQ feature when configuring and running a VFSM project as well as running filter design for pesticide reduction The figure below illustrates these changes VFSmod Windows Editor v 5 1 File Source Area UH Filter Strip VFS Design Sensitivity Uncertainty Calibration Options Window Help a z e m a Ee owe Eh Eh Files Graph a Sediment Runoff ater and Sediment Balances output sample asm Balance Overall Summary output sample osp View Output Files ater Quality Summary output sample owg As a feature if the user changes the project file name then the output file names are changed to the same first level name This feature can be overridden by changing the output fi
21. 64 Rows U D Slope lt 12 0 7 0 7 1 0 1 0 1 0 1 0 1 0 65 Rows U D Slope gt 12 0 9 0 9 1 0 1 0 1 0 1 0 1 0 Till Plant limes 33 59 times factor of 66 Rows on contour 0 7 0 85 1 0 1 0 1 0 1 0 1 0 66 Rows U D slope lt 7 1 0 1 0 1 0 1 0 1 0 1 0 1 0 Strip Till 0 25 of row spacing 68 Rows on contour 4500 60 12 10 9 8 23 69 3400 50 16 14 12 11 19 27 70 2600 40 22 19 17 17 14 12 30 71 2000 30 27 23 21 20 16 13 36 72 Rows U D Slope 4500 60 16 13 11 9 23 73 3400 50 20 17 14 12 11 27 74 2600 40 26 22 19 17 14 12 30 75 2000 30 31 26 21 20 16 13 36 Vari Till 76 Rows on Contour 3400 40 13 12 11 11 22 77 3400 30 16 15 14 14 13 12 26 78 2600 20 21 19 19 19 16 14 34 Corn after WC of ryegrass or wheat stubble WC reaches stemming stage 79 No till pl in killed WC 4000 T 7 7 T 6 m 80 3000 11 11 11 11 9 7 81 2000 15 15 14 14 11 9 82 1500 20 19 18 18 14 11 Strip till 0 25 space 83 Rows U D slope 4000 13 12 11 11 9 84 3000 18 17 16 16 13 10 85 2000 23 22 20 19 15 12 86 1500 28 26 24 22 17 14 87 Rows on contour 4000 10 10 10 10 8 88 3000 15 15 15 15 12 9 89 2000 20 20 19 19 15 12 90 1500 25 24 23 22 17 14 91 Tp conv seedbed 4000 36 60 52 41 24 20 92 3000 43 64 56 43 31 25 21 93 2000 51 68 50 45 33 26 22 94 1500 6l 73 64 47 35 27 23 Part IV VFSMOD Appendices 172
22. 8 65 m Filter Strip Width input 3 87 m Mean Filter Mannings Roughness input 0 400 Ratio of Filter Length to Source Flow Length 25 46 Total Rainfall 25 15 mm Total Rainfall on Filter 0 842 m3 Total Runoff from Source mm depth over Source Area 9 74 mm Total Runoff from Source 1 324 m3 Total Runoff out from Filter mm depth over Source Filter 4 53 mm Total Runoff out from Filter 0 767 m3 Total Infiltration in Filter 1 399 m3 Runoff Delivery Ratio RDR 0 579 Mass Sediment Input to Filter 45 03 kg Concentration Sediment in Runoff from source Area 34 00 g L Mass Sediment Output from Filter 0 16 kg Concentration Sediment in Runoff exiting the Filter 0 21 g L Mass Sediment retained in Filter 44 87 kg Sediment Delivery Ratio SDR 0 004 Effective Filter Length 8 65 m Wedge Distance 0 05 m a Used for design see Part I Section on page 37 Part II VFSMOD and UH User s Manual 70 2 UH for Input Preparation User s Manual 2 1 Installing and running UH UH is installed by default when installing VFSMOD See Section 1 2 on page 53 for details When running UH from the command line DOS and UNIX versions the name of the input file set to process is selected at the command line In this way different problems can be run from the same directory without overwriting previous results As an example one could run from the VFSMOD directory uh sample2 In this example the input file sample2 inp included in the
23. Appendices 176
24. File samples Close Window General Dutputs Detailed Sediment Transport Flow through VFS osm Hydrographs ohy oa1 oa2 File output sample osp UFSMOD v2 4 5 057 Files for this simulation File 1 code ikw inputs sample ikw File code irn inputs sample irn File i code iro inputs sample iro File code iso inputs sample iso File i code osm output sample osm File code ohy output sample ohy File code igr inputs sample igr File code og1 output sample og1 File code og2 output sample og2 File 16 code osp output sample osp File 11 code isd inputs sample isd Summary of Buffer Performance Indicators Part III VFSMOD W WindowsTM User s Manual 115 B Summary File sample osp D Eo Filter Length 8 65 m Mannings Roughness 4 Amount ime3 Runoff In Rainfall Infittration Runoff Out Component Copy to Clipboard Print Plot Edit Plot B gt Summary File sample osp ter Length 8 65 m Mannings Roughness 4 Sediment Delivery Ratio 2 Amount 4c N Q o Oo Sediment In Sediment Retained Sediment Out Component Copy to Clipboard Print Plot Edit Plot All of the output files are in ASCII format and can be imported into other applications for further processing When designing a filter using the Design option a csv file is produced and the results can be read either in text format or in graphs for RDR SDR and PDR see the Design Screens
25. III VFSMOD W WindowsTM User s Manual 121 Parameter Unit Name Spacing for grass steam cm SS Roughness grass Manning s coefficient s cm1 3 VN Height of grass cm H Roughness bare surface Manning s coefficient s cm1 3 VN2 Coarse sediment fraction d gt 0 0037 cm g g COARSE Incoming Flow sediment concentration g cm3 Cl Porosity of deposited sediments 100 POR Sediment particle class diameter d50 cm DP Sediment particle densit g cm3 SG Description of the calibration interface Once a project has been run VFSMOD W can search for the optimal model parameter s to match the simulated results and the field data measurements through the calibration menu option The calibration menu includes Hydrograph and Sedigraph however these options can not be run simultaneously Before using this feature make sure that that you have set correctly the information requested in the menu Options as shown below i e pointing to the root directory of the application typically c vfsmod w EJ FSmod Windows Editor v 4 0 7 File Source_Area_ UH Filter_Strip_ FS Design Sensitivity Uncertainty Calibration Options Window Help st yfsmod w Initialization Information DER Set Up vfsmeod w Configuration File Accept and Save Close User Name Paue User Affliation Meu ti ODt ti lt S User Address fiquser myemaiaddess UseremalAddess UO precy fa Saving o wismod w Browse J7 Ch
26. Manual cceeccecsseesseceneceeeceeeesaeeeeeeeeeeensees 70 2 1 Installing and running UH recien honk Ane eae RW ei ei HR MR a ane 70 2 2 Using the project file for input and OUtDUt 0 cee ceceseesecesceeeceeceeeeeceseeeeceseeeeeeneeseeenees 71 2 3 UM inp t Pil OS sc oi oes ee eset aod eed E A A a e 72 2 3 1 filename inp parameters for generating inputs for VFSMOD cesses 72 2 4 Sample applications isores i reti ar cae ccde ca E RR othiele heated E 73 25 Tips for running the model sssi setst asees pau seeorbiuvesecosetesaconestivbornelb ieee 78 Part HI VFSMOD W Windows User s Manual c ccescsssesssessssssssesesseessesseesseesseessee 86 1 VESMOD Model Description eccdcetissssaaievssettvadyaarenaicatss as dovels nt niircedsdetarnees 86 2r Installation Information sssseseesessesessesseesessesetsessesresssesseseesessestesesseseeseseesesseseese 86 3 Usine VESMOD artene raer R E E tua EN 89 4 M in Wind Wresni eei aa a i ia a iiis 90 4 1 vismod w Options Filei c i228 a at he a Re eee 91 5 UH Proje t WindOWrroetisisiroid e tina n E E A E a 93 5 1 UH Input File Editing essin annn eia e i e EE EE EEES 94 6 VES Project Window osmoncrioniornesedisii etal esi eh Pe a a 96 6 1 Overland Flow Inputs iKW cccccccccesseeseessecsecseeesecsecssecnecesesaeceeeeeceeeeeeeaeeaeeeseeaeecaesseeaees 98 6 2 VFS Infiltration Soil Properties 180 cccccessesseceseeseceecesceeceeeceseeeeeeseeseecaeeseeeeeseneeeerene
27. OCP of organic carbon CCP clay content in incoming sediment Part II VFSMOD and UH User s Manual 63 The following factors are used only if pesticide mass balance residue calculation is requested IDG 1 4 IDG Flag to calculate degradation 1 4 1 EU FOCUS k kref T 0 2 US EPA k kref 3 k kref T 4 k kref 0 for other values ignore and no more lines needed NDGDAY Number of days between runoff events from PRZM if used DGGHALF Pesticide half life days at reference values of temperature and water content i e 20 C and field capacity i e from PRZM Note DGHALF Ln 2 DGKREF FC Ofc VFS topsoil field capacity m m gt DGPIN Total pesticide mass liquid and solid phase entering the filter per unit area of the source field mg m e g from PRZM if used or other field simulation or data plus residual in filter measured or calculated by VFSMOD from last event in series i e OWQ file Note this is converted to total mass entering at the filter as mi DPIN SLENGTH SWIDTH from IRO file DGML dml surface mixing layer thickness cm DGML 2 cm recommended i e from PRZM DGT I Daily air temperatures C for period between events I 1 NDGDAY from PRZM MET file if used DGTHETA I Top soil water content m3 m for period between events I 1 NDGDAY from measurements THETAFAO calculations or PRZM runs for grassed area 1 4 7 3 File Example Example with no degradation requested 1 1
28. OI cumulative infiltration m cumulative infiltration when the end to ponding in reached m rainfall period index to show if time step is in the same rainfall period LO L Node number for flood checking indicates that the end of field source area runoff is reached 1 surface ponding indicator at beginning rain period 1 ponded overriding indicator to force maximum infiltration when overland flow is present 1 force infiltration capacity initial soil water content saturated soil water content cumulative precipitation in m for time step cumulative precipitation in m for last rainfall period total cumulative precipitation in m time s and rainfall rate m s over the VFS for each period cumulative excess rainfallat the node without considering BCRO relative distance from de upper filter edge where the check for ponding conditions is made i e 1 end filter 0 5 mid point 0 beginning Maximum rainfall intensity for the storm Green Ampt s average suction at wet front m maximum surface storage m cumulative surface storage m Chu s 1978 tp and tp coefficients Total cumulative rainfal m Saturated hydraulic conductivity m s Part IV VFSMOD Appendices 159 2 3 Sediment transport CDEP CI COARSE ICOARSE DF DEP DFS DP F FWID GAMMAW GAMMASB GSI GSSI GS2 GSO H NFUP NODEX J PART 1 PART 2 PART 3 POR QSED J RS RSS SE SS SC VLCM VM VMS VN VN2 XP
29. This is a good question Although this clearly assumes a certain type of distribution or at least the d50 can represent realistically the population of sediment particles entering the filter this is an accepted approach in sediment transport studies most of which come from river dynamics Please keep in mind that you are in fact not characterizing the incoming sediment based on the d50 but most importantly also based on the partitioning between fine and coarse sediment The Hayes approach is just a further elaboration of this principle where the incoming sediment population is divided into several classes ranges of particles each with its representing particle characteristics and then each class routed in turn to the filter The results are then aggregated at the end of the simulation This allows you to obtain more detail on the outgoing sediment distribution tto However it does require a lot more information on the incoming characteristics of your sediment We are now also incorporating this approach into VFSMOD which we feel is granted only when moving sediment adsorbed pollutants through the filter v My problem here is that when I calculate a dp from the input sediment distribution that includes 5 classes like Foster does it primary sand silt and clay small and large aggregates I get pretty large dp values like 100 um or more The soil I am working with is a silty clay loam and the large aggregates contribute a lot to the dp large
30. aggregates assumed to be 500 um like Foster suggests Can you tell me if some particle size distribution is assumed with the equations you use The Hayes et al 1984 paper discusses this but I must confess to not fully understanding everything they did The publications from Kentucky that deal with the theory are reports that I have not been able to find in our library Can these still be obtained We can recommend a book recently published e Design Hydrology and Sedimentology for Small Catchments by C T Haan B J Barfield and J C Hayes Chapter 9 pp 359 375 and Appendix 9C describes well most of this equations Notice that the strength of VFSMOD W is that it contains a hydrodynamic approach runoff rainfall infiltration rather than the average conditions for the event used in this reference Part II VFSMOD and UH User s Manual 84 w Sediment trapping efficiencies simulated by VFSMOD W seem to be much higher than we have measured measured are 70 80 in our farm field buffers and I have to put in small dp values seemingly too small for consistency with the Foster equations for aggregates to get near to agreement with our measurements Also I am surprised that trapping efficiencies can go to 99 when the incoming sediment is 10 primary clay particles which I would not expect to settle out in a buffer of 15m length You need to revise your sediment characteristics as well as soil infiltration capacity High simulated ef
31. and should be replaced by any other name you would like to identify the case study with max 25 characters with the only condition that all six files must be in the inputs subdirectory The name of the input file set to process is selected at the command line and the output file set is created automatically using the name given as input In this way different problems can be run from the same directory without overwriting previous results 1 4 1 filename ikw parameters for the overland flow solution 1 4 1 1 Structure of the file LABEL FWIDTH VL N THETAW CR MAXITER NPOL IELOUT KPG NPROP SX IPROP RNA IPROP SOA IPROP IPROP 1 NPROP IWQ 1 4 1 2 Definition LABEL a label max 50 characters to identify the program run FWIDTH width of the strip m VL length of the filter strip m N number of nodes in the domain integer must be an odd number for a quadratic finite element solution but the program checks and corrects if needed THETAW time weight factor for the Crank Nicholson solution 0 5 recommended CR Courant number for the calculation of time step from 0 5 0 8 recommended See Section 6 for more details MAXITER integer maximum number of iterations alowed in the Picard loop NPOL integer number of nodal points over each element polynomial degree 1 Part II VFSMOD and UH User s Manual 57 IELOUT KPG NPROP SX RNA D SOA I IWQ integer flag to output elemental information
32. are several simplified equations available such us Kirpich s 1940 t 0 01952 y 19 where is in minutes The design peak flow TR55 method m s is calculated in UH as Ip AQF 20 where q is the unit peak flow m stha lmn A is the watershed area ha Q is the runoff volume mm and F is the ponding factor that accounts for the percentage of the watershed with ponding or wetland conditions that will delay the overland flow Part I VFSMOD W Model Documentation 19 ponding area F p 0 1 00 0 2 0 97 1 0 0 87 3 0 0 75 5 0 0 72 qu is calculated from t values using the following equation SI units q 4 3046 x 19 07 108 C logt 6 21 where log is the logarithm to the base 10 t is in hours Cy C and C are constants obtained from the following Table based on the ratio P and the 24 hour design storm type for the area Types I IA II IT remember 0 2 S based on NRCS curve number method Storm I P Co C Cp I 0 10 2 30550 0 51429 0 11750 0 20 2 23537 0 50387 0 08929 0 25 2 18219 0 48488 0 06589 0 30 2 10624 0 45695 0 02835 0 35 2 00303 0 40769 0 01983 0 40 1 87733 0 32274 0 05754 0 45 1 76312 0 15644 0 00453 0 50 1 67889 0 06930 0 0 IA 0 10 2 03250 0 31583 0 13748 0 20 1 91978 0 28215 0 07020 0 25 1 83842 0 25543 0 02597 0 30 1 72657 0 19826
33. area flow path rer length SLENGTH m 34 0 00000316 000008045 00002054 Peak flow of incoming 0001215 hydrograph BCROPEAK 2192E 02 000292 m s 0003567 0005072 0005777 Plot Hydrograph 0005521 Save Continue Editing Save and Close Close Help SWIDTH Source area width m SLENGTH Source area flow path length m NBCROFF integer number of time steps of the incoming field hydrograph BCROPEAK Peak flow of the incoming field hydrograph m s BCROFF I J incoming field hydrograph flow rate time s vs gin m s The hydrograph can be viewed using the Plot Hydrograph button Part III VFSMOD W WindowsTM User s Manual 110 5 Hydrograph oe a g t a Lg te HSS SSeS OSS SSeS SoS SSS SSS A n SoS oS S50 52 4520520 a E a a a l e l 13 0267 19 728 26 4293 33 1307 39 832 46 5333 Time min Copy Plot to Clipboard Print Plot Edit Plot 6 7 VES Water Quality Input File iwq hi VFSmod Windows Editor v 5 1 6 A x File Source_Area_ UH Filter_Strip_ VFS Design Sensitivity Uncertainty Calibration Options Window Help B gt vism Project sampleP prj fo feiss Working Directory CAvismodPesticides Filter Stip Pract Files samplePpi Gave Help IV Include Water Quality 1 Pesticides Select ok Input Overland Flow Inputs Jinputs samplePikw Edit Browse Files Infiltration Soil Properites B vfsm Editing C vf
34. available and wi is the weight of a particular measurement which denotes the measurement error and is set equal to s 2 where s is the standard deviation of the measured data Lambot et al 2002 To perform the inverse calibration of parameter vector VFSMOD is coupled with the Global Multilevel Coordinate Search GMCS algorithm Huyer and Neumaier 1999 This algorithm combines global and local search capabilities with a multilevel approach To refine the minimization of the objective function the GMCS is combined sequentially with the Nelder Mead Simplex NMS algorithm Nelder and Mead 1965 Fig 1 Further details about application of GMCS NMS to inverse modeling of soil hydraulic properties are given in Lambot et al 2002 and Ritter et al 2003 The coefficient of efficiency Ceff Nash and Sutcliffe 1970 compares the variance about the 1 1 line perfect agreement to the variance of the observed data and it ranges from 8 to 1 Thereby Ceff 1 implies that the plot of predicted vs observed values matches the 1 1 line Legates and McCabe 1999 Only those data related to the parameters to be calibrated are shown in the final report If more than one parameter is selected the results will be shown as they are listed in the interface For instance if the vertical hydraulic conductivity and the filter width are selected for calibration the first number displayed in Estimated parameters Global Search MCS and Estimated
35. d 3 particle diameter d cm 100 particles lt d 8 particle diameter d cm d50 0 0037 0 2 FIGURE 15 Particle Size Distribution Cumulative Frequency Graph Please be sure to maintain these two parameters in the sediment module within these limits For best results it is important that the vfsmod simulation length extends for at least the last time step given in the measured data file If needed the simulation length can be increased in the input rain file irn Part III VFSMOD W WindowsTM User s Manual 132 10 Sensitivity Analysis Screens Sensitivity analysis can be done on a number of the input parameters for both the UH model and VFSMOD Set the ones you would like to analyze and leave the others unchecked B Source Area UH Parameter Selection iSe 2C Source Area Sensitivity Parameter Selections Set VFS Parameters Base Project Files sample2 lis UH Parameters Base Values Min Value Max Value Increment fee s ff Soil Erodibility K 1 1 1 D l Crop Factor C ir fit i jot Practice Factor P ee F Run Multi dimensional Sensitivity Analysis Do Simulations Load a different Cancel Help Base Project For the UH model sensitivity analysis can be done for Curve Number CN Soil Erodibility Factor K Crop Factor C and the Practice Factor P The user selects the parameters to consider using the Check boxes and enters the minimum maximum and an increment for th
36. e e Ce R Q Q eee CR QAQ Q e e e eee CEE 6 2011 v4 1 1 R Munoz Carpena J E Parsons WO Wihe reiclem USA NCSU A USA carpena ufl edu john _parsons ncsu edu PROGRAM TO CALCULATE OVERLAND FLOW AND SEDIMENT FILTRATION THROUGH A VEGETATIVE FILTER STRIP OF AN INFLOW HYDROGRAPH FROM AN ADJACENT FIELD DURING A STORM EVENT VFSMOD HANDLES THE CASE OF VARYING SURFACE COVER AND SLOPES AT THE NODES AND TIME DEPENDENT INFILTRATION FOR THE DOMAIN Reading inputs from inputs sample igr Reading inputs from inputs sample isd Reading inputs from inputs sample ikw Reading inputs from inputs sample irn Reading inputs from inputs sample iso Reading inputs from inputs sample iro eee Sion Ome Wiis Cis wiles il 2a S RUNNING i 7 NS SIND GWATSIMO D VA oil oil WGy BO lal During the run a set of new files is created in the OUTPUT directory SAMPLE OGI SAMPLE OG2 SAMPLE OHY SAMPLE OSM SAMPLE OSP The content of both input output files is explained in detail in the following section 1 2 2 Installing together with the Windows Graphical Interface Windows 9x NT 2000 XP See Part III of this document describing the MS Windows version of the system 1 2 3 Installing on a UNIX system a Create a directory named VFSMOD mkdir VFSMOD mv vfsmodux tar gz VFSMOD cd VFSMOD b Expand the contents of the file vfsmodux tar gz on the new directory Part II VFSMOD and UH User s Manual
37. eeceeeeteeneeereeeeeees 150 1 1 Program VF SM OD ss wich caseesveiteex cases E E E end Es EE bah cezes save dete nah ineess teh seenee 150 1 2 FINPUT LISHI D e i iiaii i E E E RE E REE E E 151 1 3 INI A B X XM X0 Q0 QM SSE NODEX ececeeseecesceseceseeesececeseeceaeeaeeeeeeeeeeneeeseenes 151 1 4 GRASSIN ICOARSE COARSE LISFIL cccccccessessecseceeeeecerceseeeeceseeneeseeeneeeneeeeeenees 151 1 5 INPUTS N NBAND NRAIN RAIN NBCROFF BCROFF TE QMAX VL FWIDTH PGPAR VKS NCHK LISFIL cccecccesseseesseseeceecesecaecaeeeenseneeeseseeeneeeseenes 151 1 6 QUAD riiete eae n asda chsh ces Richins EOE EN n lei AE EEEE E E oleh BI RMA seh i 152 1 7 FORMA A NBAND PGPAR ccecccseseseseceecescesecesceseeeeceseecsecseeesecaeeeaecneceaeeaesnseeneeneeenes 152 1 8 ELEM EK PGPAR renerne n belive cdc telecast ee hee 152 1 9 SHAPE XIS PSI DPSI WE PGPAR cececseescesseeseesceseecsecseeaeceaeeseceeneeeeeeeeeneeeseeaeeeaees 152 1 10 ASMA PE N BAND NED E E A e EE EE nEn aiT 152 1 11 BCAA NBAND R E R E dds E E O ee 153 1 12 FACTOR AN NBA N D a E a e ae 153 1 13 GASUB TIME DT L R RAIN NEND TRAI ce eececccesseeccesceesecsceeeceeceenseenseeseeneeeneeses 153 1 14 FORMB BO0 X0 Q0 N BCRO PGPAR cccccesseessesseseeeseeeecseceecaecaeeeecesenseensenseeneesseeaes 153 1 15 MODIFY QM B BCRO PGPAR ceeccesssesceseeseeeeeeseeesecaeecaecaaecaeceeceaecaeeeesresreneeeneeaes 154 1 16 SOLVE A B XSN NBAND 12k Sesic teh AN ivi eas Bote wine esta Kee aia
38. field inflow IRO file or rain series IRN file 04 2014 Changes in vfsm v4 2 4 Added error handling for no of days in pest degradation calculations max 366 file IWQ and steps in hydrograph file IRO and rainfall file IRN max 200 Added error handling when no incoming flow or sediment is provided in IRO or IRN files so that it does not produce NaN in output files 02 2014 Added additional cases for the water table scenarios gasubwt f Fixed calculation of pesticide degradation rate Kref Ln 2 t_halflife inputs f Fixed unit conversion for gas constant in pesticide degradation equation outmass f Added a check for high sorption pesticides when all pesticide in filter is sediment bonded outmass f Various static improvements in IWQ output file outmass f Part III VFSMOD W WindowsTM User s Manual 148 Part III VFSMOD W WindowsTM User s Manual 149 Part IV VFSMOD Appendices 1 APPENDIX 1 Description of the model subroutines The source code is distributed with the model This section is intended to be used with the source code which contains more detailed documentation 1 1 Program VFSMOD The main program is the driver for the program subroutines as discussed in the previous section This is done on the following steps a print banner get I O file names and open the files b initialize matrices c read inputs for sediment problem d get inputs and parameters for h
39. low Your slope too steep Keep also in mind that the Froude number that is calculated is also an estimate best guess since we don t know a priory the time series of velocity or water depth in the filter these are results from the model m We are getting a warning about large Froude numbers during the simulation What can cause this I reviewed your input file There was a mistake in the ikw file with the slope units These should be given in fractional numbers not in For example 5 3 slope is input in that file as 0 053 The numbers in the spreadsheet you sent me indicated that the filed contained the percentage form n We are getting a warning about the filter inundated during the simulation during the simulation What can cause this The peak flow rate of the incoming hydrograph for the size of plot you are working with is very large Notice that when this is converted to an estimated water depth using Manning s we are talking about close to 8 of water on the surface of the filter This violates most overland sheetflow conceptualizations The model will run but is giving you a warning that the kinematic wave will produce approximate results only errors in excess of the 10 15 that the original work by Woolhiser and Ligget suggested for this formulation o Are there any rules to set the beginning of the inflow and rainfall files iro and irn You should pick the beginning of the storm in your iro file as t 0 and then s
40. models based on conservative tracer and pesticide leaching Pest Manage Sci 2006 62 6 537 550 Haan C T B J Barfield and J C Hayes 1994 Design Hydrology and Sedimentology for Small Catchments San Diego Academic Press Haan C T B Allred D E Storm G J Sabbagh and S Prabhu 1995 Statistical Proce dure for Evaluating Hydrologic Water Quality Models TRANS ASAE 38 3 725 733 Haan C T D E Storm T Al Issa S Prabhu G J Sabbagh and D R Edwards 1998 Effect of Parameter Distributions on Uncertainty Analysis of Hydrologic Models TRANS ASAE 41 1 65 70 Hayes J C 1979 Evaluation of design procedures for vegetal filtration of sediment from flowing water unpublished Ph D dissertation Univ of Kentucky Lexington KY USA Hayes J C B J Barfield and R I Barnhisel 1979 Filtration of sediment by simulated vegetation II Unsteady flow with non homogeneous sediment Transactions of ASAE 22 5 1063 1967 Hayes J C B J Barfield and R I Barnhisel 1982 The use of grass filters for sediment control in strip mine drainage III Empirical verification of procedures using real veg etation Report No IMMR82 070 Int for Mining and Mineral Res Univ of Ken tucky Lexington KY Hayes J C B J Barfield and R I Barnhisel 1984 Performance of grass filters under Jackson S Hendley P Jones R Poletika N Russell M Comparison of regulatory method estimated drinking water ex
41. output files The source code is written in standard FORTRAN77 so that compilation should be straight forward following the included makefile and using the proper set of files for each platform Windows 9x NT 2000 XP or UNIX Binaries for a few computer platforms can also be found at the internet site 1 2 1 Installing for a DOS command prompt window under Windows 9x NT 2000 XP a From the Start Menu Start a Command Prompt DOS window b Change to the drive and directory where you want to install c Create a directory named VFSMOD d Expand the contents of the file vfsmodpce zip This should create the following directory structure vfsmod docs inputs output src_uh src_vfsm e The executable files VFSM EXE and UH EXE can be found in the parent directory VFSMOD Part II VFSMOD and UH User s Manual 53 f Run the sample case named SAMPLE by typing VFSM SAMPLE at the DOS prompt Please note that the second part of the command issued SAMPLE refers to a set of files located in the subdirectory INPUTS You could run a different problem by selecting a different set of input files with the condition that they are located in the subdirectory INPUTS In this example if you issue the DIR command within the INPUTS directory you should see the following files SAMPLE IGR SAMPLE IKW SAMPLE IRN SAMPLE IRO SAMPLE ISD SAMPLE ISO After you execute the command you should see a screen as follows C eU Cee C Cee eE
42. parameters Local Search NMS represents the calibrated value for the hydraulic conductivity and the second one would represent the value for the filter width The final report includes the parameter set with confidence limits estimated by the GMCS and the NMS Information related to the optimization covariance and correlation matrices model adequacy RMSE Ceff number of iterations of model execution and total duration of the whole process is also displayed Part III VFSMOD W WindowsTM User s Manual 130 P Untitled Notepad File Edit Format View Help terations 112 Final OF 1 182011806024628100 Estimated parameters Global Search mcs 0 o1600000000000 Estimated parameters Local Search nms 0 o1600000000000 xopt 90 ConFLimits 95 CI 1 60000e 002 Inf Inf Parameters covariance matrix Inf Parameters correlation matrix NaN Model adequacy p_adeq 9 055841 FIGURE 14 Details of the calibration results as shown when the Calibration Results button is clicked Tips to run the calibration component For the Sedigraph option based on the particle size distribution cumulative frequency graph see next figure the values of COARSE and DP d 50 must meet the following requirements For COARSE 0 5 then d50 gt 0 0037 cm See part a in next figure For COARSE 0 5 then d50 lt 0 0037 cm See part b in next figure Part III VFSMOD W WindowsTM User s Manual 131 particles lt
43. program Part IV VFSMOD Appendices 150 1 2 FINPUT LISFIL This subroutine writes the program banner reads the name of the file set to be processed from the command line string creates I O file names accordingly and opens I O files 1 3 INI A B X XM X0 Q0 QM SSE NODEX The main program matrices are set to zero before the beginning of the simulation 14 GRASSIN ICOARSE COARSE LISFIL This subroutine reads in the main parameters of the sediment sub model calculates some of the additional parameters needed and echo this information into the output files This is done in the following steps a read parameters from the igr and isd input files b choose particle diameter cm fall velocity cm s and particle density g cm from the internal data base or if the particle class is not in the database calculate values using Fair and Geyer method 1954 based on Stokes note all units in SI c if particle is fine d lt 37 mm don t run the wedge part COARSE 0 D0 d output some input values here and leave the rest for the INPUTS subroutine e print heading for tables in output files 1 5 INPUTS N NBAND NRAIN RAIN NBCROFF BCROFF TE QMAX VL FWIDTH PGPAR VKS NCHK LISFIL This subroutine reads in the main parameters of the overland flow and infiltration sub models calculates some of the additional simulation parameters needed and echo this information into the output files This is done in the following steps a
44. read parameters from the ikw input file assign nodes to the X values for each surface segment and calculate elemental Manning s a s b calculate filter main slope Sc and roughness for sediment calculations c check if N is compatible with type of shape function selected and if not print message d read rainfall distribution from irn e read soil inputs from iso and calculate Green Ampt parameters f get downslope node for flood checking SCHK g read runoff inflow at upper side of strip BC in m s from iro h find the bandwidth for the matrix number of elements and number of nodes i calculate convergence and wave form parameters CR C FR FK j calculate the PG Parameters from the Courant number values k set the order of the integration rule 1 output parameters Part IV VFSMOD Appendices 151 1 2 output hydrological and numerical parameters l 1 output nodal information if selected ielout 1 1 2 output values for sediment transport read previously in GRASSIN m issue a warning if any of the criteria CR FR FK is not met n print heading for tables in output files 1 6 QUAD Get the Gaussian Quadrature points for orders 1 through 5 1 7 FORMA A NBAND PGPAR This subroutine assembles the system matrix A as a banded matrix This procedure involves the calculation of element matrices EK subroutine ELEM and their accumula tion in the banded system matrix A subroutine ASSM Fina
45. roughness in each of the sections entry 1 2 3 and exit After solving the sediment transport problem for a time step new values of roughness and or slope are selected as nodal values for the finite element grid in zones A t and B t whereas C t and D t remain unchanged Figure 3 Changes in surface saturated hydraulic conductivity values K are considered negligible The new surface parameters are fed back into the hydrology model for the next time step Surface changes are accounted for in this way The time step for the simulation is selected by the kinematic wave model to satisfy convergence and computational criteria of the FE method based on model inputs Mufioz Carpena et al 1993a b At the end of the simulation the model outputs include information on the water balance volume of rainfall field inflow filter outflow and infiltration hydrograph sediment balance field inflow filter outflow and deposition sedimentograph filter trapping efficiency and sediment deposition pattern within the filter 2 5 Solution procedure The VFSMOD main program calls the subroutines along the solution procedure The Part I VFSMOD W Model Documentation 12 backbone of the model is the numerical solution to the overland flow equation and the infiltration and sediment transport models are called upon to solve the equation for each time step at the time of assembling the matrix system The numerical method is based on a
46. shrub with l to 2 inch sand or gravel mulch and basin borders Urban districts Commercial and business 85 89 92 94 95 Industrial 72 81 88 91 93 Residential districts by average lot size 1 8 acre or less town houses 65 77 85 90 92 1 4 acre 38 61 75 83 87 1 3 acre 30 57 72 81 86 1 2 acre 25 54 70 80 85 1 acre 20 51 68 79 84 1 Average runoff condition Ia 0 2S Part IV VFSMOD Appendices 164 Cover Description Curve Numbers for hydrologic soil group Average T percent Cover type and hydrologic condition impervious A B C D area 2 acres 12 46 65 77 82 Developing urban areas Newly graded areas pervious areas only 77 86 91 94 no vegetation Idle lands CN s are determined using cover types similar to those in table 2 2c a areas are directly connected to the drainage system impervious areas have a CN of 98 and pervious areas are considered b c d Composite CN s to use for the design of temporary measures during grading and construction should be computed using figure 2 3 or 2 4 TR55 document based on the degree of development impervious area percentage and the CN s for the newly graded pervious areas Type Description Final infiltration of Ks mm h minimum runoff potential deep sands loess 8 to 12 low runoff potential shallow sands loess 4 to 8 medium runoff potential shallow sands or claye
47. simulation parameters file sample ohy Parameter Symbol Value Units Petrov Galerkin parameters PGPAR 0 0433 0 0031 0 3165 0 1451 Space step DX 0 155 s Time step DT 1 40 Number of elements in system NELEM 28 Number of time steps NDT 2568 Estimated maximum flow rate QMAX 0 000735 m s Estimated maximum flow depth HMAX 0 000735 m Celerity of the wave C 0 08816 0 01389 m s Courant time step DTC 1 753 s Froude number FR 0 143 Kinematic wave number FK 1892 1 6 2 2 Hydrological outputs files sample ohy and sample osm The hydrographs included in the next figure show the volume reduction infiltration and peak delay increase of roughness by vegetated surface produced by the filter over the incoming field hydrograph input Part II VFSMOD and UH User s Manual 67 0 003 5 106 10 10 1 5105 2105 EE Rainfall 2 5105 Inflow o e nm Outflow Flow rate m s Rainfall m s 0 001 0 1000 2000 3000 Time s The water balance for the simulation was as follows Volume from rainfall 0 8423 m Volume from up field hydrograph 1 3240 m Volume from outflow hydrograph 0 7674 m Volume infiltrated 1 3990 m 1 6 2 3 Sediment transport files sample igl sample ig2 and sample osm The sedimentograph and mass balance at the filter is included in the next two figures Both graphs show a significant load reduction due to deposition at the wed
48. the measured sediment characteristics are not known as is often the case but the soil texture is one can estimate the dp parameter from soil texture Tables on how to do this are provided in the documentation of the User s Manual 5 7MB x In the Infiltration soil properties file there is the parameter maximum surface storage of which I don t really know how it is defined Does this refer to ponding How is it measured in the field Surface storage represents the amount of excess rainfall that must be filled at the surface an average over the area considered before runoff can begin Remember excess rainfall is the amount not infiltrated into the soil during the infiltration storm event To set this value consider the regularity of your area 0 0 5 cm would correspond to a fairly well graded soil surface To get more background on the role of this parameter I recommend Chapter 4 by Skaggs and Khaleel pg 147 152 in Hydrologic Modeling of Small Watersheds eds C T Haan et al 1982 ASAE Mon 5 ASAE St Joseph USA y There is the parameter fraction of the filter where ponding is checked of which I also cannot find the definition This is a fairly insensitive parameter except for very sandy soils It represents where along the filter the user wants for ponding at the beginning of the event to be checked for The idea is that two different mechanisms can produce ponding in the filter during an Part II VFSMOD and UH User s Manua
49. uplands of the Piedmont and Coastal Plain areas of North Carolina USA and under controlled field conditions in Guelph Canada The model is targeted at studying VFS performance on an event by event basis Though a research tool the model can assist planners and regulators to determine the relative effectiveness of filter strips in a given situation This version of the model uses inputs that are easily obtainable and a program to generate inputs for the model is provided Extensive model documentation is provided on line also in PDF format from the model s web site http abe ufl edu carpena vfsmod 2 Installation Information This package consists of three programs to assist users in evaluating and developing design specifications for vegetative filter strips for trapping sediment and enhancing infiltration The programs are the graphical user interface GUI vfsmod w exe version 4 1 xx a program to estimate rainfall hyetographs runoff hydrographs and storm based erosion losses from typical source areas UH uh exe version 2 4 or later and the vegetative filter strip model VFSMOD vfsm exe version 2 4 or later The GUI was developed to assist users in executing the Vegetative Filter Strip Model VFSMOD and UH Development of the graphical user interface program GUI was started in March 2000 Since that time we have continued to improve the interface and add new features to the system As such we expect there will be a numbe
50. 0 02633 0 50 1 63417 0 09100 0 0 II 0 10 2 55323 0 61512 0 16403 0 30 2 46532 0 62257 0 11657 0 35 2 41896 0 61594 0 08820 0 40 2 36409 0 59857 0 05621 0 45 2 29238 0 57005 0 02281 0 50 2 20282 0 51599 0 01259 Ill 0 10 2 47317 0 51848 0 17083 0 30 2 39628 0 51202 0 13245 0 35 2 35477 0 49735 0 11985 0 4 2 30726 0 46541 0 11094 0 45 2 24876 0 41314 0 11508 0 50 2 17772 0 36803 0 09525 To obtain the coefficients rather than interpolating in the previos Table a set of fourth order polynomials were obtained see next Table Part I VFSMOD W Model Documentation 20 Storm Coef A B C D E Co 68 0317 74 693 24 9255 3 9797 2 5222 l C 82 907 105 222 42 167 6 7479 0 8657 C3 11 1619 26 314 16 1126 2 9776 0 0456 Co 144 547 136 68 41 8526 6 2829 2 3645 IA C 130 64 134 907 45 773 6 585 0 6384 C3 55 230 47 9565 13 503 2 1954 0 2644 Co 11 312 12 1681 6 5688 1 0577 2 5021 Il C 16 6125 16 337 6 4981 1 1784 0 5476 C3 43 015 50 4334 19 740 3 2996 0 3427 Co 11 505 14 2182 7 8919 1 3836 2 4007 m C 64 177 85 7116 38 206 6 7419 0 8899 C3 65 9007 85 806 39 0036 6 8946 0 2078 a CAUP HBU PV CU P DU P E i 0 1 2 Part I VFSMOD W Model Documentation 21 2 Storm type o CO a 0 1 A C2
51. 0 18 16 21 39 On moderate slopes 4500 70 9 8 7 7 18 40 60 12 10 9 8 18 41 50 14 13 11 9 19 42 40 17 15 13 10 20 43 30 21 18 15 13 21 44 20 25 22 19 16 22 45 Do 3400 60 13 11 10 10 8 20 46 50 16 13 12 12 9 24 47 40 19 17 16 14 11 25 48 30 23 21 19 17 14 26 49 20 29 25 23 21 16 27 50 10 36 32 29 24 20 30 51 Do 2600 50 17 16 15 15 13 10 29 52 40 21 20 19 19 15 12 30 53 30 25 23 22 22 18 14 32 54 20 32 29 28 27 22 17 34 55 10 41 36 34 32 25 21 37 56 Do 2000 40 23 21 20 20 15 12 37 57 30 27 25 24 23 19 15 39 58 20 35 32 30 28 22 18 42 59 10 46 42 38 33 26 22 47 On slopes gt 12 lines 33 Brie nee factor of 3 13 13 1a 10 fro 10 10 Disk or harrow after spring chisel or field cultivation lines 33 59 times 6l On moderate slopes 1 1 1 1 1 1 1 0 1 0 1 0 1 1 62 On slopes gt 12 1 4 1 4 1 2 1 0 1 0 1 0 1 0 Ridge Plant lines 33 59 times factor of 63 Rows on contour 0 7 0 7 0 7 0 7 0 7 0 7 0 7 Part IV VFSMOD Appendices 171 Spring Cover Soil loss ratio for cropstage period and canopy cover Cover Crop Sequence Resi After No and manmagement due Plant F SB 1 2 3 80 90 96 4Lf LB
52. 1 or not 0 integer flag to choose the Petrov Galerkin solution 1 or regular finite element 0 integer number of segments with different surface properties slope or roughness real X distance from the beginning on the filter in which the segment of uniform surface properties ends m Manning s roughness for each segment s m 3 slope at each segment unit fraction i e no units water quality transport problem selection flag 0 or not present do not run problem 1 run problem iwq file required 1 4 1 3 File example 3 87 14 0 6182 0 4 1 2364 0 4 1 8546 0 4 2 4729 0 4 3 0911 0 4 3 7093 0 4 4 3275 0 4 4 9457 0 4 5 5639 0 4 6 1821 0 4 6 8004 0 4 7 4186 0 4 8 0368 0 4 8 6550 0 4 0 Unit9 g8 ul83 91 8 655 57 0 5 0 8 350 3 1 1 0 052778 0 032639 0 071528 0 075000 0 031944 0 019444 0 029885 0 028947 0 041667 0 134028 0 079167 0 074306 0 040972 0 062346 Which corresponds to a filter on dense uniform bermuda grass with slope as follows 1 4 2 filename irn storm hyetograph 1 4 2 1 Structure of the file Part II VFSMOD and UH User s Manual 58 NRAIN RPEAK RAIN LJ I 1 NRAIN J 1 2 1 4 2 2 Definition NRAIN integer number of rainfall periods including period to end simulation RPEAK maximum rainfall intensity for the storm m s RAIN LJ time s and rainfall rate intensity m s over the VFS for each period The last time step cor
53. 1 4 4 filename iso soil properties for the infiltration model Case 1 No water table present 1 4 4 1 Structure of the file VKS SAV OS OI SM SCHK 1 4 4 2 Definition VKS saturated hydraulic conductivity K m s SAV Green Ampt s average suction at wet front m OS saturated soil water content 0 m m Ol initial soil water content 0 m m SM maximum surface storage m SCHK relative distance from de upper filter edge where the check for ponding conditions is made i e 1 end filter 0 5 mid point 0 beginning Part II VFSMOD and UH User s Manual 60 1 4 4 3 File example 1 33e 5 0 37904 0 311 0 125 0 0 1 00 Case 2 With shallow water table present 1 4 4 4 Structure of the file VKS SAV OS Ol SM SCHK WTD ITHETATYPE PAR I IKUNSTYPE PARK J 1 4 4 5 Definition VKS SAV OS See Case 1 above for definition OL SM SCHK WTD water table depth m ITHETATYPE an integer to select the soil water characteristic curve type with values 1 van Genuchten 2 Brooks and Corey IKUNSTYPE an integer to select the unsaturated hydraulic conductivity curve type with values 1 van Genuchten 2 Brooks and Corey and 3 Gardner s PAR parameters of the soil water retention curve PARK J parameters of the unsaturated hydraulic conductivity 1 4 4 6 File example 3 89E 05 1 72 0 39 0 25 0 1 0 7 2 0 15 78 286 0 4 2 6 7471 78 286 In this example a shallow water table is present 0 7 m
54. 147 30 30 Example with degradation requested 1 1 read create iwq amp owg files 0 0 396 Kd proc 0 Kd L Kg 1 Koc Koc L Kg OC 25 Clay content in sediment 1 IDG 3 27 995 0 26 6 097 2 ndgday dgHalf d FC m3 m3 dgPin mg m2 dgML cm 95 86 6 3 dgT i i 1 ndgday Celsius 0 265 0 264 0 265 dgTheta i i 1 ndgday 1 5 Model file outputs The program writes output into ASCII files Each aspect of the model is written to different files The model outputs include information on the water balance volume of rainfall field inflow filter outflow and infiltration hydrograph sediment balance field inflow filter outflow and deposition sedimentograph filter trapping efficiency and Part II VFSMOD and UH User s Manual 64 sediment deposition pattern within the filter The output files contain summaries of the main state variables in the program Note that these files are created in the output directory at run time every time the model is run and that the actual file names are given by substituting filename by the name of the set selected at the command line If you wish to keep the results from different simulations it is advised that you create a new set of input files with a different name for each case study The inputs and outputs included in these files are labeled in a verbose form to be self explanatory a filename ohy This file contains information related to the hydrology side of the
55. 200 10 70 55 43 18 13 11 u 147 500 30 43 34 23 13 10 8 148 750 40 34 27 18 10 7 7 149 1000 50 26 21 15 8 7 6 150 1500 60 20 15 12 7 5 5 151 2000 70 14 11 9 7 5 5 Part IV VFSMOD Appendices 174 Spring Cover Soil loss ratio for cropstage period and canopy cover Cover Crop Sequence Resi After No and manmagement due Plant F SB 1 2 3 80 90 96 4Lf LB 152 With row crop residues 300 5 82 65 44 19 14 12 153 500 15 62 49 35 17 13 11 154 750 23 50 40 29 14 11 9 155 1000 30 40 31 24 13 10 8 156 1500 45 31 24 18 10 8 7 157 2000 55 23 19 14 8 7 5 158 2500 65 17 14 12 7 5 4 POTATOES 159 Rows with slope 43 64 56 36 26 19 16 Contoured rows ridged when canopy cover is about 160 10 43 64 28 18 13 10 8 a Symbols B soybeans C corn conv till plow disk and harrow for seedbed cot cotton F rough fallow fld cult field cultivator G small grain GS grain sorghum M grass and legume meadow at least 1 full year pl plant RdL crop residues left on field RdR crop residues removed SB seedbed period sprg spring TP plowed with moldboard WC winter cover crop insignificant or an unlikely combination of variables b Dry weight per acre after winter loss and reductions by grazing or partial removal 4500 lbs represents 100 to 125 bu corn 3400 Ibs
56. 2002 as well as a component to simulate fecal pathogen filtering from runoff Zang et al 2001 For an updated list of the latest applications please visit the model web page at http abe ufl edu carpena vfsmod Part I VFSMOD W Model Documentation 31 4 Sensitivity and Uncertainty Analysis Procedures for UH and VFSMOD Built In VFSMOD W The sensitivity of a model output to a given input factor has been traditionally expressed mathematically as derivatives of the model output with respect to the input variation sometimes normalized by either the central values where the derivative is calculated or by the standard deviations of the input and output values Haan et al 1995 These sensitivity measurements are local because they are fixed to a point base value or narrow range where the derivative is taken These local sensitivity indexes are classified as one parameter at a time OAT methods i e they quantify the effect of a single parameter by assuming all others are fixed Saltelli et al 2005 Local OAT sensitivity indexes are only efficient if all factors in a model produce linear output responses or if some type of average can be used over the parametric space Often the model output responses to changes in the input factors are non linear and an alternative global sensitivity approach where the entire parametric space of the model is explored simultaneously for all input factors is needed The advantage of the global ov
57. 54 gzcat vfsmodux tar gz tar xvf This should create the following directory structure vfsmod docs inputs output src_uh src_vfsm c An installation script setup is included in the VFSMOD directory To compile and install the program simply type setup The script will compile the source code and copy the executable files vfsm and UH to the VFSMOD directory If your FORTRAN compiler name is not f77 you will need to edit the makefile found in the src directory You can also clean the executable and object files by typing setup clean d Run the sample case named sample by typing vfsm sample at the UNIX prompt Please note that the second part of the command issued sample refers to a set of files located in the subdirectory inputs You could run a different problem by selecting a different set of input files with the condition that they are located in the subdirectory inputs Note that you must have all the six input files in order to run the program In our example if you issue the s command within the inputs directory you should see the following files sample igr sample ikw sample irn sample iro sample isd sample iso After you execute the command you should see a screen similar to the one given above During the run a new set of files is created in the output directory sample ogl sample og2 sample ohy sample osm sample osp The content of both input output files is explained in detail in the following section 1 3 Using th
58. 62 3 3 Vegetation types for VFS S 00 cccccesessecseessecseceeceseeseceseeseceseeeeseaseseecaeeseecaesaeeeserenseeerenes 163 3 4 NRCS SCS Curve Nam beis ne a Rice EE T loess Mi Rat Ao 164 3 5 MUSLE Crop factor C icc sce ssinscpencastesesvleueths anodes traps ecdtdestd tian esbeanasitees NE 170 3 6 Contour factor P values for MUSLE equation in UH ce ececseesecsceeteceeenteeneeeteeeens 176 3 7 References for Soils and Vegetation data ccccceccecscesseessesseesececesseeeeceecneceeeeeesneeeseeses 176 Part I VFSMOD W Model Documentation 1 Introduction Runoff carrying sediment from nonpoint sources has long been recognized as a major pollutant of surface water Sediment bound pollutants such as phosphorous and some pesticides are also a major pollution concern Several management practices have been suggested to control runoff quantity and quality from disturbed areas One such management practice is vegetative filter strips VFS which can be defined as Dillaha et al 1989 areas of vegetation designed to remove sediment and other pollutants from surface water runoff by filtration deposition infiltration adsorption absorption decomposition and volatilization These bands of planted or indigenous vegetation separate a water body from a land area that could act as a nonpoint pollution source Vegetation at the downstream edge of disturbed areas may effectively reduce runoff volume and peak velocity primarily because
59. 728 867 021 194 388 WoN Ww GW WD hb 000 2 743 h Rainfall mm Rain30 mm 0 199 241 287 1338 396 460 933 UH v1 06 3 2002 Part II VFSMOD and UH User s Manual 76 9 yes here 10 1 500 11 1 667 12 1 833 13 2 000 14 2 167 15 24333 16 2 500 17 2 667 18 2 833 19 3 000 20 B67 21 9 2333 22 3 500 23 3 667 24 3 833 25 4 000 26 4 167 27 4 333 28 4 500 29 4 667 30 4 833 31 5 000 32 Ta LEI 33 Ie 333 34 5 500 35 5 667 36 5 833 37 6 000 Computed Total Rain Actual Total Rain raimax30 I30 616 713 x826 967 138 S357 650 070 740 067 832 564 067 740 070 650 J357 1 38 z967 828 ETAS 616 3533 460 396 338 287 241 2199 RRR RRR RH RH RH Lt DS NH WH WH ORD G WH WN NH NH KPH EHH 102 102 36 72 608 1862 158 509 934 463 146 078 460 877 35 639 36 463 36 463 1323 71 ble 877 9 460 078 146 463 934 509 1158 862 608 388 194 021 867 728 BPO ONDA GAA Wao A AK AA OO ODN 600 mm 600 mm 463 mm 927 mm h RAINFALL ENERGY FACTOR R FOR EROSION CALCULATIONS a Foster et al 1977 E 3738 632 ft tonf acre 25 049 MJ ha volro 63 323 mm qpeak 46 011 mm h Factors in Rm Rst 182 695 Rro 226 905 Rm Foster 170 764 N h b Williams 1975 Watershed area 0 500 ha V 316 613 m3 Qp 0 064 m3 s Rw Williams 97 514 N h c GLEAMS daily CREAMS Rain 102 60 mm R_GLM 64 79 From Gleams
60. 75 to 99 bu 2600 Ibs 60 to 74 bu and 2000 Ibs 40 to 59 bu with normal 30 percent winter loss For RdR or fall plow practices these four productivity levels are indicated by HP GP FP and LP respectively high good fair and low productivity In lines 79 to 102 this column indicates dry weigth of the winter cover crop c Percentage of soil surface covered by plant residue mulch after crop seeding The difference betweenn spring residue and that on the surface after crop seeding is reflected in the soil loss ratios as residues mixed with the topsoil d The soil loss ratios given as percentages assume that the indicated crop sequence and practices are followed consistently One year deviations from normal practices do not have the effect of a permanent change Linear interpolation between lines is recommended when justified by field conditions See also footnote 7 e Cropstage periods are as defined on p 18 Agriculture Handbook 537 The three columns for cropstage 3 are for 80 90 and 96 to 100 percent canopy cover at maturity f Column 4L is for all residues left on field Corn stalks partially standing as left by some mechn anical pickers If stalks are shredded and spread by picker select ratio from Table When residues are reduced by grazing take ratio from lower spring residue line g Period 4 values in lines 9 12 are for corn stubble stover removed h Inversion plowed no secondary tillage For this practice residues mus
61. 8 65 45 32 26 22 9 RdR sprg TP HP 66 74 65 47 22 568 10 GP 67 75 66 47 27 23 62 11 FP 68 76 67 48 35 27 69 12 LP 69 77 68 49 35 74 13 RdR fall TP HP 76 82 70 49 22 14 GP 77 83 71 50 27 23 15 FP 78 85 72 51 35 27 16 LP 79 86 73 52 35 17 Wheeltrack pl RdL TP 4500 z k 31 27 25 5 5 18 26 18 3400 36 32 30 22 18 30 19 2600 43 36 32 29 24 19 37 20 2000 51 43 36 31 24 20 47 21 Deep offset disk or plow 4500 10 45 38 34 20 23 22 3400 10 52 43 37 24 20 30 23 2600 5 57 48 40 32 25 21 37 24 2000 61 51 42 33 26 22 47 25 No till plant in crop resi 6000 95 2 2 2 2 14 26 due 6000 90 3 3 3 3 14 27 4500 80 5 5 5 5 15 28 3400 70 8 8 8 8 6 19 29 3400 60 12 12 12 12 9 8 23 30 3400 50 15 15 14 14 11 9 27 31 2600 40 21 20 18 17 13 11 30 32 2600 30 26 24 22 21 17 14 36 Part IV VFSMOD Appendices 170 Spring Cover Soil loss ratio for cropstage period and canopy cover Cover Crop Sequence Resi After No and manmagement due Plant F SB 1 2 3 80 90 96 4Lf LB Chisel shallow disk or fld cult as only tillage 33 On moderate Slopes 6000 70 8 8 7 7 17 34 60 10 9 8 8 17 35 50 13 11 10 9 18 36 40 15 13 11 10 19 37 30 18 15 13 12 20 38 20 23 2
62. 92 37 14386 87 16548 71 18780 59 21407 38 23885 81 26558 78 9 88 8 20 8 70 9 20 9 70 10 10 10 60 11 00 11 30 11 70 11 90 12 10 s FldROmm FldROm3 VFSROmm VFSROm3 VFSINFm3 FldSEDkg FldSEDcone VFSSEDkg VFSSEDcone SDR RDR 10538 32 5283 96 6570 51 8037 12 9701 96 11445 35 13525 39 15123 39 16983 22 19594 29 21576 43 23583 94 9 17 1 21 7 80 8 39 8 97 9 43 9 98 10 07 10 23 10 72 10 76 10 76 926 874 893 908 921 931 94 914 904 915 903 888 998 994 995 996 996 998 998 999 299 999 999 999 The format of this file is space separated that is easily imported into another analysis package such as spreadsheet To import in a spreadsheet one selects space separated data Part III VFSMOD W WindowsTM User s Manual 134 treat multiple spaces as one The first five lines denote general information on the parameter and the base project file This information includes the parameter range total rainfall for the event and the filter strip length along with the base project filenames The tabular information presents the event level outputs and starts on the 7 line Each line contains the results for one of the simulations The columns are retv the return value for that simulation 0 indicates that simulation had no errors CN the curve number input UHk soil erodibility K input UHc crop factor input UHp practice factor input iso
63. Coefficient value C0 C1 C2 Storm type IA 8 o CO o Ct o z A C2 Q on oO T gt E S Ke E oO Sr Sanne are Ae mes Ae gusaka gt setae o t 7 0 1 0 2 0 3 0 4 0 5 I P Coefficient C0 C1 C2 Coefficients C0 C1 C2 oe Storm type II ee ee o co ao Cf A C2 p EEEE bend Agree A 4 A ee ese AET TENE oeigeses es ae a 02 03 04 05 lP c Storm type III e o co m Ci A C2 h ra A 4 h eise Faeeeces Figure 8 Coefficients predicted by proposed polynomials used in NRCS peak flow calculation Part I VFSMOD W Model Documentation 22 3 2 3 Time correction for hydrograph to match hyetograph 3 2 3 1 Option 1 based on NRCS abstraction method Following the NRCS definition for abstraction and curve number we have P I F Q Time mti De i D Figure 9 Precipitation partition in NRCS method Since we can calculate the initial abstraction as Ia 0 2 S and S 25400 CN 254 22 as shown in out file we could find the time when this initial abstraction ends ti by interpolating in the constructed NRCS 10 min hyetograph hyt file Since the starting time for runoff coincides the time rainfall excess begins a time shifting is needed in the hydrograph to match the rainfall as toff ti 23 and all the hydrograph times will be corrected as t t toff 24 3 2 3 2 Option 2 based on NRCS abstraction and Unit Hydrograph Based on the u
64. Dp Particle Class diameter and SS Media Element Spacing parameters Selection of each parameter is done with the Check boxes and setting the distribution Currently the normal log normal triangular and uniform are available After selecting the distribution the Set Parameters button opens the window to enter the parameters defining the distribution For the normal and log normal distribution the mean Part III VFSMOD W WindowsTM User s Manual 139 and standard deviation are entered The peak and maximum and minimum values specify the triangular distribution The minimum and maximum values determine the range for sampling the uniform distribution On either of the screens the number of simulations is also specified These will typically range in the thousands although the user can specify any number On a Pentium IH 1 GHZ processor based desktop each simulation takes from 10 15 seconds up to as much as minute a VFSMOD Uncertainty Parameter Selections p d O x Uncertainty Selections for Buffer Area I kag Siiuleiarintarnetion Current Project Files sample prj After Selecting the Distribution Click on the Button to set the Parameters Base Values Distribution parameters v Green Ampt Ksat VKS em h 4 788 Set Parameters Green Ampt Theta Initial cm 3 cm 3 125 Set Parameters Particle Class Diameter cm i Set Parameters Media Element Spacing 83 cm 2 2 Select Distribution v Set Parameter
65. EAMS method organic matter Part II VFSMOD and UH User s Manual 73 The acceptable values for soiltype are Soil Types Case Sensitive Clay Silty clay Sandy clay Silty clay loam Clay loam Sandy clay loam Silt Silt loam Loam Very fine sandy loam Fine sandy loam Sandy loam Coarse sandy loam Loamy very fine sand Loamy fine sand Loamy sand Loamy coarse sand Very fine sand Fine sand Sand Coarse sand 2 3 1 3 File example file sample2 inp Clay 1 25 00 Waa 8 G OAS O O NOOO OZ 2 4 Sample application Table 1 Parameter values for the sample run Parameter P CN A storm type D L Y soiltype K CFACT PFACT Value 25 85 0 5 3 6 100 0 02 Clay 0 25 1 0 1 0 Parameter Description amount of storm precipatation in mm NRCS SCS Curve Number for the source area Area of the upstream portion in ha storm type 1 I 2 II 3 HI 4 Ia storm duration h Length of the source area along the slope m Slope of the source area expressed as a fraction See Table for Acceptable Soil Types Soil Erodibility If K lt 0 then K is computed based on texture and organic matter See REF C factor See Table in Appendix 3 P factor See Table in Appendix 3 Part II VFSMOD and UH User s Manual 74 IEROTY 1 Select the method to compute the storm R factor in MUSLE not present or 1
66. Equation USLE The Universal Soil Loss Equation USLE was developed in the 1950 s by Wischmeier and Smith 1978 as an empirical equation to address erosion from areas characterized by overland flow The equation was derived from thousands of site years of observed erosion rates around the world The equation is given by A RKLSCP 30 A soil loss average over the slope length R combined erosivity of rainfall and runoff see section 3 3 2 K soil erodibility factor determined as the soil loss from a unit plot with dimensions 22 m 73 feet on a 9 slope tilled up and down slope with tillage periodically to prevent surface crusting and weeds LS topographic factor based on the lenght and slope and is computed as where slope length in m and n slope length exponent which is 0 5 for slopes gt 4 0 4 for slopes between 3 and 4 and 0 3 for slopes lt 3 Part I VFSMOD W Model Documentation 25 S Slope factor calculated as S 65 45 4 56s 0 065 where s sin slope angle C cover management factor ratio of soil loss from the particular cover management to that of the unit plot dimensionless P conservation practice factor ratio of soil loss from the practice to that of the unit plot dimensionless The unit plots were defined as 22 m 73 feet in length 9 slope tilled up and down the slope periodically to prevent surface crusting and weeds The values L S C and P are refere
67. Figure 3 The algorithm calculates the g g value for each time step and compares it with the sediment inflow load If g j gt g all sediment is transported through the first part of the filter wedge g g 7 and the sediment is filtered at the suspended sediment zone lower part of the filter If g lt g deposition at the wedge occurs and the fraction not deposited is filtered at the lower part of the filter g g g 7 The calculation procedure utilizes a modified Manning s open channel flow equation continuity equation and Einstein s sediment bed load transport function Flow values at the filter entry and points 1 and 2 in Figure 3 di 7 q2 respectively are needed for these calculations After the downside of the wedge two zones C t and D t form the suspended load zone or effective filter length L t Figure 3 On zone C t sediment has covered the indentations of the surface so that bed load transport and deposition occurs but the soil slope S is not significantly changed All bed load transported sediment is captured before reaching zone D t so only suspended sediment is transported and deposited in this zone until the flow reaches the end of the filter with sediment load g The sediment trapping algorithm for the suspended load zone follows Tollner et al 1976 equation based on a probabilistic approach to turbulent diffusion for non submerged flow Flow values at point 3 and filter exit g3 and qour respect
68. J B BCRO BCROFF 200 2 BCROFFQ C DPSI 1 DR DT DTC DX DR DX1 EK I J EPS FK FR FWIDTH HMAX KNOISE MAXITER MFLAG N NBAND NDT NELEM NL NMAX NPOL PGPAR I PSI I QK MAXEQN QM N QO N QMAX QOUT R RNA I SOA I SR SWIDTH system matrix square of dimensions nxn ie A right hand side vector of dimensions Ixn ie b time interpolated water depth at the first node of the system m boundary condition at the upstream node time s vs depth m inflow hydrograph m3 s read from input file ROFFKW IN celerity of the wave m s derivative of the i th shape basis function at XI duration of the simulation s increment of time s Courant time step for the numerical solution space step m duration of the rain s distance between nodes in element entry in element stiffness matrix convergence criteria set to 10 8 in the program kinematic flow number Froude number width of the strip m maximum flow depth at steady state condition filter for numerical oscillation in FLOW subroutine maximum number of iterations alowed convergence flag 0 no convergence 1 convergence actual number of nodes in the domain bandwidth for the A matrix number of time steps actual number of elements in the domain order of the integration rule over each element maximum number of equations and variables that can be solved number of nodal points over each element polynomial degree 1 Pe
69. KINEROS A Kinematic Runoff and Erosion Model Documentation and User Manual USDA ARS ARS Pub no 77 Yonts C D Wilson R G Hein G L Control of pesticides and nitrates in surface irriga tion runoff water Paper No 96 1041 American Society of Agricultural Engineers St Joseph MI 1996 Part I VFSMOD W Model Documentation 53 Zhang Q C G Okoren and K R Mankin 2001 Modeling Fecal Pathogen Transport in Vegetative Filter Strips American Society of Agriculture Engineers Sacramento Cal ifornia Paper of ASAE no 01 2194 ASAE St Joseph Part I VFSMOD W Model Documentation 54 Part HI VF SMOD and UH User s Manual 1 VFSMOD user s manual 1 1 Obtaining VE SMOD VFSMOD documentation source code and binaries for a number of platforms can be obtained in digital format through internet at the following URL site USA http abe ufl edu carpena vfsmod The files are in ZIP tar gz compressed format All necessary files to compile and run a sample application are included Please select Windows 9x NT 2000 XP vfsmodpc zip for the command line version or vfsmod w install zip for the graphical user interface or UNIX vfsmodux tar gz versions as needed If you do not have an internet connection you can contact the authors for assistance 1 2 Installing and running VFSMOD VFSMOD and UH source code is distributed both in Windows 95 98 NT 2000 XP and UNIX versions along with make files and sample input and
70. Lee 1989 Vegetative filter strips for agricultural nonpoint source pollution control Transactions of ASAE 32 2 491 496 Engman E T 1986 Roughness coefficients for routing surface runoff J Irrigation and Drainage Eng ASCE 112 1 39 53 Part I VFSMOD W Model Documentation 48 Foster G R and L F Huggins 1977 Deposition of sediment by overland flow on con cave slopes In Soil Erosion Prediction and Control Special Publication No 21 Soil Conservation Society of America Ankeny IA pp 167 182 Foster G R 1982 Chapter 8 Modeling the erosion process In Hydrologic Modeling of Small Watersheds Editors C T Haan H P Johnson and D L Brakensiek ASAE Monograph No 5 American Society of Agricultural Engineers St Joseph MI pp 297 380 Foster G R R A Young and W H Neibling 1985 Sediment composition for nonpoint source pollution analyses Trans of ASABE 28 1 133 146 Fox G A Mufioz Capena R Sabbagh G J Influence of flow concentration on input fac tor importance and uncertainty in predicting pesticide surface runoff reduction by veg etative filter strips J Hydrol 2010 doi 10 1016 j jhydrol 2010 01 020 Fox GA Sabbagh G J Comment on Major Factors Influencing the Efficacy of Vege tated Buffers on Sediment Trapping A Review and Analysis J Environ Qual 2009 38 1 1 3 Fox GA Sabbagh GJ Chen W Russell M Comparison of uncalibrated Tier II ground water screening
71. Loamy sand 16 67 0 0135 0 2794 0 437 0 363 0 506 26 86 0 0613 0 46 Sand 65 4 0 0097 0 2536 0 437 0 374 0 500 Rawls and Brakensiek 1983 P Saxton and Rawls 2006 assuming MO 2 5 Note Values in parenthesis are mean values For an alternative source of Green Ampt soil parameters see also McCuen et al 1981 References for Table McCuen R H W J Rawls and D L Brakensiek 1981 Statistical Analysis of the Brooks and Corey and the Green Ampt parameters across soil textures Water Resour Res 17 4 1005 1013 Part IV VFSMOD Appendices 161 Rawls W J and D L Brakensiek 1983 A procedure to predict Green Apmt infiltration parameters Adv in Infiltration pp 102 112 ASAE Pub no 11 83 Sabbagh G J GA Fox A Kamanzi B Roepke and J Z Tang Effectiveness of vegeta tive filter strips in reducing pesticide loading Quantifying pesticide trapping efficiency Journal of Environmental Quality 38 2 38 2 762 771 Saxton K E and W J Rawls 2006 Soil water characteristic estimates by texture and organic matter for hydrologic solutions Soil Sci Soc Am J 70 1569 1578 3 2 Manning s roughness coeficient n There are several publications dedicated to the stimation of this important parameter for overland flow routing see Arcement et al 1989 A summary of the most common values used in overland flow routing can be taken from Engman 1986 as
72. N 2 upwinding Petrov Galerkin finite element method approximation for the spacial derivatives and a time weighting finite difference approximation for the time derivatives The non linearity of the equation q q h is taken care of by using the Picard iterative scheme inside every time step lagging 2 3 of the power of A in q 5 3 2 3 m 1 m 1 for the iteration level m such as A h b h 9 In this program the core of the time step solution is taken care of following this steps 1 2 3 4 5 6 7 Form the system matrix A of constant coefficients Perform LUD decomposition over this matrix A Form the system matrix BM of constant coefficients Form r h s of equation vector b BM x for each time step Solve for A b to get a x for that time step Repeat 4 amp 5 until convergence of that time step Repeat 3 amp 6 until completion of desired number of time steps The transport model supplies information to build the BM and b for each time step Part I VFSMOD W Model Documentation 13 dt The general procedure is structured into subroutines as illustrated in the next diagram El a 4 FACTOR I k UPDATE MODIFY O SOLVE Picard Iteration is E lt OCF EINSTEIN STEP3 Only POINTS a 100 times GRASSED KWWRITE P OUTMASS Figure 5 VFSMOD model structure After solving the sediment transport problem for a given time
73. NTRCAP COARSE This program solves STEP3 of the sediment transport problem after Barfield et al 1979 and Hayes et al 1984 The outputs from this part of the problem are f sediment fraction trapped in the deposition wedge X7 t Y t X t sediment wedge geometry DEP depth of deposited sediment at lower section of the filter 7 sediment trapping effi ciency The procedure is as follows a if sediment transport capacity g 2 is greater than the fine sediment load fraction Lsim diameter gt 0 0037cm all sediment goes through the wedge to the lower part of the filter ntrcap 1 b if transport capacity is lower than the fine sediment load fraction then the FINE frac tion goes through the wedge and the COARSE fraction is filtered at the wedge c apply open channel flow theory and Einstein s bed load transport equation in B t find df Ry Se Newton Raphson method d find advancement of sediment front and outflow concentration d 1 if top of vegetation has not been reached calculate the triangular wedge geometry d 2 trapezoidal wedge geometry e check if strip has been filled up If so set flag NFUP 1 change sediment wedge geom etry to a rectangle of hight H and length VL and bypass GRASSED in the future Also in this case avoid suspended sediment zone calculations f on the assumption that the trapped sediment is uniformly distributed on the bed of the filter s lower section area calculate DEP depth
74. OINTS J XT X1 YT Coefficient for reducing suspended sediment deposition in D t sediment inflow concentration g cm3 of coarse particles gt 37 microns in incoming sediment Flag 0 all particles fine lt 37 microns don t run the wedge part depth of flow at D t cm depth of sediment deposited in suspended sediment zone D t cm depth of flow at B t cm particle size diameter cm fraction trapped in the deposition wedge width of the strip Cm FWIDTH 100 water weight density g cm3 sediment weight density g cm3 sediment load entering before field deposition gsI g s 1cm 1 sediment load entering the filter after field depos gsi g s 1cm 1 sediment load entering downstream section gsd g s lcm 1 sediment load exiting the filter gso g s 1cm 1 filter media height cm Flag 0 strip is not filled up 1 strip is filled up node number for X1 X2 X3 points sediment class diameter cm sediment class fall velocity cm s sediment class weight density g cm3 porosity of deposited sediment overland flow rate for X1 X2 X3 points cm2 s hydraulic radius of the filter cm hydraulic radius of the filter at B t cm equilibrium slope at B t spacing of the filter media elements cm filter main slope filter length in cm depth averaged velocity at D t cm s depth averaged velocity at B t cm s filter media Manning s n 0 012 for cylindrical media s cm 1 3 Filter Manning s roughness
75. RO 154 1 17 CONVER N X XM MEFLAG 0 cececcecsseeeseceeceececesceseceseeseecsecseeesecsaesaecnecaeeneeneserseneeaes 154 1 18 UPDATE N X XO rerin a ici eit eis ieee Wie eas ee nR 154 1 19 FLOW N X TEOT Jaen Ries n oo hn ees 154 1 20 GRASSED TIME N QIN NODEX ICOARSE COARSE cccceceesseeseeeseeseeseeseeeseeaes 155 1 21 OCECNPLACE ic ocscrsci E E Roe E acted cage havea eateadonck ost O 155 1 22 EINSTEIN GS2 NTRCAP COARSE cccccessssesescesceeeceseeeeeeaeeseecsecaaesaecnecnseneenseerenes 155 1 23 STEP3 GS2 TIME NTRCAP COARSE eececsceesesseesseeseceeceseeseceseeeeceseeeeeeneeneeeneeaeeeaees 156 1 24 POINTS N XPOINTS NODEX VBT cccccecceseesseesceseeeeceseeneecseeseesaecaeceaeceeneserenseeats 156 1 25 KWWRITE N L M QTEMP X BCRO FWIDTH ccccecceseesseeseceeceseceeceeceseeseeneeeseenes 157 1 26 OUTMASS VL FWIDTH TRALLISFIL 00 00 iaa es a a E Aa 157 APPENDIX 2 Model parameters and variables csceescesseeseeeeeeeeeeeceseeneeess 158 2 1 Overland TOW ress Sees oe tote a a edb toc ae te e Cee a Se oe ae ales 158 2 2 Infiltration 22 2 eka bean ear AR ee oe es 159 2 3 Sediment transport v 24 ees Hote sAv O Ree ee gace ene ee ee 160 APPENDIX 3 Soils and Vegetation data c ccccccsccsseceseceesceeeecsseceeeeeeeeeaeens 161 3 1 Soils data Green Ampt parameters ccccceeccesseesecsceesececeseeeeeeeceseeeeeeseeeeecaeeaeecseeseeaees 161 3 2 Manning s roughness Coeficient 1 nescence iei a 1
76. State Univ Raleigh Mu oz Carpena R C T Miller and J E Parsons 1993a A quadratic Petrov Galerkin solution for kinematic wave overland flow Water Resour Res 29 8 2615 2627 Mu oz Carpena R Fox GA Sabbagh G J Input factor importance and uncertainty in predicting pesticide surface runoff reduction with vegetative filter strips J Environ Qual 2010 doi 10 2134 jeq2009 0300 Mufioz Carpena R J E Parsons and J W Gilliam 1993b Numerical approach to the overland flow process in vegetative filter strips Transactions of ASAE 36 3 761 770 Mufioz Carpena R J E Parsons and J W Gilliam 1999 Modeling hydrology and sedi ment transport in vegetative filter strips and riparian areas J of Hydrology 214 1 A 111 129 Mufioz Carpena R and J E Parsons 1999 Evaluation of VFSmod a vegetative filter strips hydrology and sediment Paper of ASAE no 992152 ASAE St Joseph Mufioz Carpena R and J E Parsons 2002 A normalized design procedure to meet sedi ment TMDL with vegetable filter strips In Watershed Management to Meet Emerg ing TMDL Environmental Regulations Proc 11 13 March Fort Worth Texas USA eds A Saleh B Wilson pp 256 261 St Joseph Michigan ASAE Mufoz Carpena R Parsons J E 2004 A design procedure for vegetative filter strips using VFSMOD W T ASAE 2004 47 6 1933 1941 Mufioz Carpena R 2012 Continuous simulation components for pesticide environmental ass
77. VEFSMOD VEGETATIVE FILTER STRIPS BE MODELING SYS MODEL DOCUMENTATION amp USERS MANUAL Rafael Mu oz Carpena Agricultural amp Biological Engineering University of Florida 287 Frazier Rogers Hall P O Box 110570 Gainesville FL 32611 0570 John E Parsons deceased October 2005 Biological amp Agricultural Engineering North Carolina State University Raleigh NC 27695 VFSMOD W Vegetative Filter Strips Modelling System MODEL DOCUMENTATION amp USER S MANUAL version 6 x Rafael Mu oz Carpena Agricultural amp Biological Engineering University of Florida 287 Frazier Rogers Hall P O Box 110570 Gainesville FL 32611 0570 carpena ufl edu John E Parsons deceased Biological and Agricultural Engineering North Carolina State University Raleigh NC 27695 UF FLORIDA Last Updated June 24 2014 Disclaimer VFSMOD W 5 x Vegetative Filter Strip Modelling System VFSMOD was initially developed in the Department of Biological and Agricultural Engineering by Dr Rafael Mufioz Carpena under the direction of Dr John E Parsons The model and associated documentation is supplied as is with no warranty explicit or implied The model is provided to as an educational and research tool This version is the fifth moving the model from a research tool to one available for general users As with any model the results are totally dependent on the user s ability to wisely select input parameters that represe
78. W Model Documentation 28 where R storm modified R see below for explanation of units V volume of runoff m Q peak discharge rate m s Using R in place of R in the USLE is referred to as the modified USLE or MUSLE The units of soil loss for this version are Mg for the total watershed area and not per unit area as in the original USLE This assumes that the soil erodibility K units are Mg h ha N 3 4 Computational Structure of UH The program UH generates the necessary inputs from the upslope source area for vfsmod The inputs for UH are discussed in the User s Manual along with a sample input and data set Figure 12 shows the computation structure of UH First the input data describing the source area is read Next UH computes the total runoff from the source area using the SCS Curve Number method The time of concentration peak runoff rate and time of peak is computed by the SCS TR55 method Next SCS unit hydrograph theory is used to estimate the runoff hydrograph An idealized rainfall hyetograph is generated from the SCS storm type MUSLE is then used to estimate the sediment loss from the source area for the storm The sediment loss is partitioned into silt and clay based on the soil particle distribution in the top soil The average concentration in runoff then estimated based on the total runoff and distribution of soil particles in the sediment loss Finally the results are used to create input files for vf
79. Wischmeirer R_GLM 110 27 N h Converted to Metric ERODIBILITY K AND PARTICLE SIZE SELECTION Table for computing Ksoil from GLEAMS and KINEROS i Soil Type Sand Silt Tex F SETG En Per F D50 Part II VFSMOD and UH User s Manual 77 1 Clay 20 5 30 0 01287 0 0650 0 075 23 2 Silty clay 10 45 0 01870 0 0650 0 075 24 3 Sandy clay 50 10 0 01714 0 0650 0 075 66 4 Silty clay loam 15 50 0 02606 0 0650 0 050 25 5 Clay loam 35 30 0 02360 0 0650 0 050 18 6 Sandy clay loam 553 20 0 02778 0 0650 0 050 91 7 SAIE Ons Soe 0 05845 0 0650 0 025 19 8 Silt loam 20 60 0 04259 0 0650 0 025 27 9 Loam 45 35 0 03618 0 0325 0 025 con 10 Very fine sandy loam 60 25 0 03877 0 0350 0 000 35 11 Fine sandy loam 60 25s 0 03205 0 0000 0 000 80 12 Sandy loam 60 25 0 02549 0 0325 0 000 98 13 Coarse sandy loam 60 25 0 01914 0 0325 0 000 160 14 Loamy very fine sand 84 8 0 03726 0 20325 0 025 90 15 Loamy fine sand 84 8 0 02301 0 0000 0 025 120 16 Loamy sand 84 8 0 01624 0 0325 0 025 139 17 Loamy coarse sand 84 8 0 00982 0 0325 0 025 180 18 Very fine sand 90 Ja 0 04401 0 0325 0 050 140 19 Fine sand 90 Ds 0 02173 0 0000 0 050 160 20 Sand 90 Dis 0 01481 0 0325 0 050 170 21 Coarse sand 90 Ja 0 00827 0 0325 0 050 200 For the selected soil type Sandy clay K 0 041 kg h N m 2 d50 66 00 um MISCELLANEOUS CALCS L 1 447 S 0 182 cfact 1 00 pfact 1 00 FINAL CALCS A 1 84779 kg m 2 Using Rm Foster
80. al 1971 He developed a regression equation based on data collected from 55 midwestern soils using percentages of organic matter primary particles sand silt clay and permeability In GLEAMS this relationship was further simplified to K TF 12 0 OM SF PF 31 where K soil erodibility factor in tons acre EI TF texture factor OM percentage organic matter SF structure factor PF permeability factor TF SF and PF are given in the following Table for the primary soil types K is converted to SI units kg N h m by multiplying by 0 1317 So K in SI units kg Part I VFSMOD W Model Documentation 26 N h m7 is given by K 0 1317 TF 12 0 OM SF PF 32 TABLE 1 Factors for computing K by soil type from GLEAMS based on data from Wischmeier et al 1971 Soil Type Sand Silt Texture Structure Permeability D50 Factor Factor Factor Clay 20 30 0 01287 0 0650 0 075 23 0 Silty clay 10 45 0 01870 0 0650 0 075 24 0 Sandy Clay 50 10 0 01714 0 0650 0 075 66 0 Silty clay loam 15 50 0 02606 0 0650 0 050 25 0 Clay loam 35 30 0 02360 0 0650 0 050 18 0 Sandy clay loam 55 20 0 02778 0 0650 0 050 91 0 Silt 5 85 0 05845 0 0650 0 025 19 0 Silt loam 20 60 0 04259 0 0650 0 025 27 0 Loam 45 35 0 03618 0 0325 0 025 35 0 Very fine sandy 60 25 0 03877 0 0350 0 000 35 0 loam Fine sandy loam 60 25
81. alibration Results button on the main calibration window ad VFSmod Windows Editor v 4 0 7 File Source_Area_ UH Filter_Strip_ V F5 Design Sensitivity Uncertainty Calibration Options Window Help S Inverse Calibration Hydrograph Project Name sample prj Browse Hogan ord inverse meas_hyd txt s Results Inverse Calibration ii flora agen ufl edu SharedDocs c wfsmod terations 112 gt Final OF 1 182011806024628100 gt Estimated parameters Global Search MCS 0 01600000000000 a Estimated parameters Local Search NMS D 0 01600000000000 3 Xopt 90 ConfLimits 95 C1 D 1 60000e 002 Inf Inf 3 Parameters covariance matrix o Inf Parameters correlation matrix Close 1 1 Graph Graph Fit gt FIGURE 13 Text and graphical results after the calibration is completed Part III VFSMOD W WindowsTM User s Manual 129 Description of the global calibration algorithm used The inverse simulation of the flow or sediment parameters is carried out by gt minimizing an objective function OF b that represents the error between measured and simulated values such that it can be defined as a nonlinear least squares problem by N gt gt OF b Y wlr t Yb i 1 where the right hand side represents the deviations between observed Yo and predicted Ys hydrographs or sedigraphs using parameter vector t is the time N is the number of measurements
82. an e mail problems and any suggestions or questions via the web site http abe ufl edu carpena vfsmod You can also e mail your problems directly to carpena ufl edu Current Issues Hints Problems and Workarounds 1 2 3 4 5 6 7 Download the zip file containing the VFSMOD package vfsmod w install zip to your temp directory and unzip into a subdirectory After setup is complete you can delete the subdirectory You can delete the zip file but you may want to keep this in case you need to re install the program During setup you may receive a message that setup needs to update your system If you receive this message then allow setup to update your system After setup updates your system reboot and execute setup again In Windows 98 the MSDOS command window that vfsm exe and UH exe executes within is not automatically closed You should close this manually On some systems if you choose to install the package in drv Program Files then the execution menu may not work correctly for UH and VFSMOD We have seen this on Windows NT 4 0 systems The default install directory is c vfsmod To avoid this problem we recommend you use this directory If you have a previous version of vfsmod on your computer you should uninstall prior to installing this version With this version on Windows NT 2000 and XP you will need Administrator privileges to install A few system files are copied into the Windows System
83. anic soil J Hydrol 295 124 139 Sabbagh G J Fox G A Kamanzi A Roepke B Tang J Z Effectiveness of vegetative filter strips in reducing pesticide loading Quantifying pesticide trapping efficiency J Environ Qual 2009 38 2 762 771 Saltelli A 1999 Sensitivity analysis Could better methods be used J Geophys Res 104 24013 24013 Saltelli A M Ratto S Tarantola and F Campolongo 2005 Sensitivity analysis for chemical models Chem Rev 105 2811 2827 Saltelli A S Tarantola and F Campolongo 2000a Sensitivity analysis as an ingredient of modeling Stat Sci 15 377 395 Saltelli A K Chan and E M Scott 2000b Sensitivity Analysis Wiley Series in Proba bility and Statistics Saltelli A S Tarantola F Campolongo and M Ratto 2004 Sensitivity Analysis in Practice A Guide to assessing Scientific Models John Wiley amp Sons Chichester Skaggs R W and R Khaheel 1982 Chapter 4 Infiltration In Hydrologic modeling of small watersheds Ed by C T Haan H P Johnson and D L Brakensiek ASAE Monograph No 5 American Society of Agricultural Engineers St Joseph MI pp 121 168 Sobol I M 1990 Sensitivity estimates for nonlinear mathematical models Matematicheskoe Modelirovanie 2 112 118 translated as I M Sobol 1993 Sensi tivity analysis for non linear mathematical models Math Modeling Comput Stutter M I Langan S J Lumsdon D G Vegetated buffer
84. ar and fewer iterations to convergence will be needed for each time step Initial testing of the program showed that the sediment predictions do not change greatly but the user is advised to assess this point for each particular application h Sometimes when assigning a high intensity in irn file from the beggininng of the simulation this can result in kinematic shock and the numerical solution blows up Part II VFSMOD and UH User s Manual 80 This sometimes can be avoided by adding intermediate steps of rainfall with the same intensity at the beginning of the period For example the irn file that produces shock contains one rainfall period of 64 mm h starting at t 0 s and ending at 3h 36 min i e 4 0 000017817 Nrain rpeak m s 0 000017817 13000 000017817 13001 0 13603 0 The shock in this case can be avoided by adding an intermediate step after the t 0 s value and close to it i e 5 0 000017817 Nrain rpeak m s 0 000017817 300 000017817 13000 000017817 13001 0 13603 0 i When optimizing particle size or specific density for predefined particle classes NPART lt 7 in isd file with the inverse calibration component the results don t seem to be right By default the automatic calibrator overrides the NPART setting and forces it to NPART 7 so changes in DP are considered by the model However the user must be careful to consider the specific density value given for SG
85. arameter is computed as Part I VFSMOD W Model Documentation 32 _ 00P i T PO 39 Once the most sensitive inputs are identified the model users can concentrate on determining the best or most appropriate values for a given desing scenario In addition these parameters can also be used to evaluate the uncertainty in the model outputs based on these most sensitive inputs This approach involves selecting probability distributions for each sensitive input based on based on previous literature and field research After the probability distributions are identified for each of the inputs then these distributions are sampled for typical inputs and the simulated outputs are used to determine a probability distribution for each output parameter Two possible methods were presented for generating the general probability distributions of the output variables of interest Haan et al 1995 and Haan et al 1998 The first method was First Order Approximation FOA Morgan and Henrion 1990 In this method the mean or expected value of the output is estimated as E O Model P 40 and the variance is estimated as _ eo O00 Var O by A Var P 2 5 De rar Cov P P 41 i l i 1j itl where O is the output parameter of interest P is the base input parameter values for the selected input variable P is the input parameter 7 n is the number of input parameters Var is the variance and Cov is the covariance If the inpu
86. at the value for bare surface Manning s is just the value you would give if there where no vegetation And the vegetation Manning s is then the value that your vegetation adds to the total Manning s coefficient Or is it the total value In the buffer segment properties per segment there is also a value required for the Manning s coefficient Does this value refer to any of those two other Manning s coefficients bare surface and grass that are asked in the buffered vegetation characteristics Or is it a total value of the Manning s coefficient for every segment There are indeed 3 Manning s coefficients used RNA in filename ikw gives the value for each grass segment typically depending on the grass density in that segment or sometimes an average value across the filter for all segments is used VN and VN2 in filename igr give the values for base soil and modified value for cylindrical media grass stems at a mesoscopic scale respectively The first one of these last two represents the value into which the filter transitions when the edge closer to the field starts filling up with sediment sediment wedge and this reaches the maximum value H Notice the value for each time step is average weighed between good grass and bare soil as the wedge thickens for those segments of the filter affected The modified Manning s values depend on the filter type and typical values are given in the VFSMOD W manual appendixes Notice the units are different to th
87. atibility with previous versions of the program command line and windows GUI c provide a framework for the addition of future other water quality components options 2 and 3 when the Include Water Quality option is selected User selection of new pesticide component From a user perspective the main difference with previous versions of the program is the addition of a new flag at the end of the IKW input file that enables the new water quality component Sample IKW file showing new water quality CWQ flag WQ on CWQ 1 WQ off CWQ any number character or missing bme C Documents and Settings ABE User My Documents desktop backup oct 2007 VFSMOD CODE HELP with HELLLP help5 vfsmod w FlagIkw bmp When the new flag is missing or contains a vale different than 1 the program executes like in previous versions 4 0 or lower without the water quality component With CWQ 1 the program requires a new input file in the inputs directory with extension WQ described below As before the program can be executed from the command line in two ways When the program is executed followed by the name no extension of a set of files input files with same name For example typing vfsm sample will execute the program for a set of files sample ikw sample irn sampe iro sample iso sample isd A second way of Part III VFSMOD W WindowsTM User s Manual 97 executing the program is specifying a project file wi
88. below the surface Brooks and Corey is selected for the soil moisture characteristic curve 8 0 15 ago 78 286 A 0 4 The unsaturated hydraulic conductivity based on Brooks and Corey is selected n 6 7471 QBC 78 286 and he 1 agc 1 4 5 filename igr buffer properties for sediment filtration model 1 4 5 1 Structure of the file Part II VFSMOD and UH User s Manual 6l SS VN H VN2 ICO 1 4 5 2 Definition SS spacing of the filter media elements cm VN filter media grass Manning s nm 0 012 for cylindrical media s cm 1 3 H filter media height cm VN2 bare surface Manning s n for sediment inundated area and overland flow s m 1 3 integer flag to feedback the change in slope and surface roughness at the ICO sediment wedge for each time step 0 no feedback 1 feedback See also additional info on this parameter on the Tips to Run the Model section 1 4 5 3 File example 2 2 0 012 15 0 04 1 1 4 6 filename isd sediment properties for sediment filtration model 1 4 6 1 Structure of the file NPART COARSE CI POR DP SG 1 4 6 2 Definition NPART integer incoming sediment particle class according to the USDA 1975 and Foster et al 1985 particle classes NPART Particle class Diam range cm dp cm V_ cm s Ys cm s 1 Clay lt 0 0002 0 0002 0 0004 2 60 2 Silt 1 0 0002 0 005 0 0010 0 0094 2 65 3 Small aggregate 0 0030 0 0408 1 80
89. calibration mode Part III VFSMOD W WindowsTM User s Manual 123 2 VFSmod Windows Editor v 4 0 7 File Source_Area_ UH Filter_Strip_ VF5 Design Sensitivity Uncertainty Calibration Options Window Help Look in vfsmod w z K fim My Recent Documents O documentation inputs Diinverse output patterns on Desktop E la ae File name sample My Network Files of type Project Files pri X Places J Open as read only FIGURE 5 Selction of a project file to run the calibration mde 2 VFSmod Windows Editor v 4 0 7 File Source_Area_ UH Filter_Strip_ VF5 Design Sensitivity Uncertainty Calibration Options Window Help S Inverse Calibration Hydrograph Project Name fsample pr Browse Hydrograph Measured py Data Fie tt Browse Edit Select parameters to be changed or calibrated No Change Calibrate e C C Vertical Saturated K VKS m s Average suction wetting front SAV m Saturated water content 0S m3 m3 Initial water content 01 m3 m3 Maximum Surface Storage SM ml Filter fraction where ponding is checked SCHK rT Filter width FWIDTH m as a G 2 a a Filter length VL m Filter Manning s n RNA s m 1 3 Gi D S ee a ee ee ee S E D E D S ee ee D Average Filter Slope ISOA m m Advanced Calibration Settings Results FIGURE 6 Opening a field data file to run the calibration m
90. ce Option 1 Pesticides Sabbagh et al 2009 allowed the use in VFSMOD of the following empirical pesticide trapping efficiency equation AP a b AQ c AE din F 1 e C Where AP is the pesticide removal efficiency 5 AQ is the infiltration defined as the difference between flow entering to VFS i e inflow runon plus precipitation minus the runoff from the VFS AF is the sediment reduction Fph is a phase distribution factor i e ratio between the mass of pesticide in the dissolved phase relative to the mass of the pesticide sorbed to sediment and C is the clay content For five model development studies Sabbagh et al 2009 reported regression parameter a 24 8 b 0 5 c 0 5 d 2 4 and e 0 9 The foundation of the proposed empirical equation is a physically based parameter for the potential for a pesticide to partition between the soil and water phase as quantified through the linear sorption coefficient Kd Kd Koc OC 100 where OC is the percentage organic carbon in the soil The phase distribution parameter Fph is determined using the following equation where Qi and Ei are the volume of water L and mass of sediment kg entering the VFS Fph Qi AQ Ei Option 2 TaRSE and Option 3 Other Part III VFSMOD W WindowsTM User s Manual 112 These options are not available now but leave the space ready for adding the Generic Pollutant Transport Component and other algorithms for specif
91. cified B Setup Design Parameters for Filter Strips Base Projects gt Base UH Project File sample2 lis Show UH Change UH Hide Project Project Base FS Project File sample prj Se aon Hide Select the options for analyzing the effectiveness of a proposed VFS design Design Storm Duration hrs 6 Return Period T years 5 10 25 50 100 Design St Amount Oceanis valine mm fO fo fo f fo p fo Design Storm Start mm 20 End mm 700 Inc mm 10 Range Lower Upper Increment Base M FS Length m 8 655 I Grass eg Media Spacing cm 2 2 Design Results Filename Specify a short name without ext DesignRes OK Run View Design Results Cancel Results can be seen in text format by pushing the View Results button There is also an available option to show graphical results of RDR SDR and PDR by selecting the View Graphical Results button Part III VFSMOD W WindowsTM User s Manual 143 B gt General Output Viewer General Output ewer C vismod w output DesignResT est csv Close Input Design Parameters a baselow basehigh baseinc Filter Length amp Media Spacing 2 14 4 2 2 2 2 B StrmDuration 6 Amounts 36 56 66 Base Info Rainfall mm Buffer Length m gt base uh sample2 lis ba 25 8 655 retu Rainfall InPNum InpValue FldROmm F1dROm3 YFSROmm UFSROM3 YFSINF 6 36 6 2 6 777 33 883 6 599 33 046
92. coefficient for bare sediment inundated soil s cm 1 3 Position for the 3 locations where de flow is read from the hydrol ogy model q1 q2 q3 at the 3 last faces of the filter cm X2 XPOINTS 2 cm width of sediment wedge from field edge width of sediment wedge in the field cm height of sediment deposition wedge at the initial triangular stage Part IV VFSMOD Appendices 160 3 APPENDIX 3 Soils and Vegetation data 3 1 Soils data Green Ampt parameters The model developers encorage the users to obtain the soil inputs for the model based on sail samples taken on site If that is not possible or the model is applied to study the effect of soil type on the effectiveness of the VFS the following table gives values for the Green Ampt parameters as suggested by Rawls and Brakensiek 1983 3 3 Soil Texture USDA K m s x10 6 Sam Porosity x 0 m m Clay 0 167 0 0639 1 565 0 475 0 427 0 523 0 306 0 3163 0 50 Sandy clay 0 333 0 0408 1 402 0 430 0 370 0 490 Clay Loam 0 556 0 0479 0 9110 0 464 0 409 0 519 Silty Clay 0 278 0 0613 1 394 0 479 0 425 0 533 1 028 0 2922 0 52 Silty clay loam 0 556 0 0567 1 315 0 471 0 418 0 524 Sandy clay loam 0 833 0 0442 1 080 0 398 0 332 0 464 Loam 3 677 0 0133 0 5938 0 463 0 375 0 551 4 306 0 0889 0 46 Silt loam 1 89 0 0292 0 9539 0 501 0 420 0 582 Sandy loam 6 067 0 0267 0 4547 0 453 0 351 0 555
93. d Corey equation for hydraulic conductivity curve is K K S pe ABC K K koe ABC where n Brooks and Corey shape parameter 77 3 2 A Note There are 3 parameters the user needs to define gc and 7 The Gardner equation for hydraulic conductivity curve is K h K e cpxh where Agpn Gardner shape parameter Note There are 2 paramters the user needs to define K and agpy Part III VFSMOD W WindowsTM User s Manual 105 Example of the new iso file with shallow groundwater option 3 89E 05 1 72 0 39 0 25 0 1 0 7 2 0 15 78 286 0 4 2 6 7471 78 286 In this example a shallow water table is present 0 7 m below the surface Brooks and Corey is selected for the soil moisture characteristic curve 0 0 15 gc 78 286 A 0 4 The unsaturated hydraulic conductivity based on Brooks and Corey is selected n 6 7471 pc 78 286 and since there is no value in the fifth line of the file h 1 agc An example of the same iso file without shallow groundwater will be 3 89E 05 1 72 0 39 0 25 0 1 6 3 VFS Buffer Vegetation Characteristics igr B vfsm Editing C vfsrnod w inputs sample igr S E Buffer Vegetation Properties file igt _inputs sample igr Vegetation Properties Spacing for grass stems 5S cm 39 Pepi m Manning s n 012 s m Height of grass H cm Mso Roughness Bare surface Manning s u n Vn2 s m 1 3 Feedback the change in slope and rou
94. d equations To construct hyetographs for any duration D h and storm type equation 3 7 in Haan et al 1994 pg 49 was transformed to P P tmigt t D 2 POnig PD 2 ji Pp P t at D 2 P taig D 2 oa mi where 7 18 9 995 for storm type I 7 960 for storm type IA and 12 0 for storm type II and MI This modification from the original formula results from the fact that to construct a hyetograph for a duration lt 24 h the interval should be centered around the steepest part of the curve i e around tnig for each one of the storm types An example of the hyetographs obtained for the different storm types for the event in the included sample file UH in 25mm in 6 hours can be seen in Figure 7 Part I VFSMOD W Model Documentation l IA Il amp I IJ ikg ttre Loew ew ek a Ufs Ifa Ifa Ifa Ifa yo on E a ak f i il ik k 0 1 2 3 4 5 6 Time h Figure 7 Rainfall hyetographs generated for different storm types 3 2 Generation of Runoff Hydrographs 3 2 1 Computation of Total Runoff using NRCS Curve Number method SI units Runoff from the source area is computed using the NRCS SCS Curve number method USDA NRCS 1984 _ P 0 2S Q P 0 85 14 where Q total runoff in cm P total precipitation in cm and P gt 0 2S S represents the antecedent moisture and is computed by _ 25400 3 CN 254 15 where CN curve number for the source area The initial abstrac
95. d running uncertainty and sensitivity analyses on vfsm alone along with uh and vfsm 2 Changed a number of the menu headings to more verbose self explanatory titles 3 Fixed a number of bugs when low runoff and or sediment yields occur many of these fixes are in vfsm and uh see their change logs for details 7 31 2005 vfsmod w exe version 3 00 xx uh exe version 2 4 xx vfsm exe version 2 4 xx 1 Numerous bug fixes and improvements have been made 1 28 2007 vfsmod w exe version 4 1 xx New release 1 Release of version 4 1 0 of vfsmod w eautomatic inverse calibration engine global sensitivity and uncertainty analysis modules for analysis with SIMLAB v2 2 Part III VFSMOD W WindowsTM User s Manual 146 eupdated versions of the vfsm uh programs redesign many GUI input forms for consistency and ease of use eupdated program built in help file with new calibration component redesign forms 2 Revision v2 4 6 of uh exe and vfsm exe with a number of minor fixes see the changes files with the source code distribution 7 15 2008 1 Minor release version 4 1 1 of vfsm The input runoff time series in the iro file was not written properly by the GUI number of lines was not refreshed Fixed Some cosmetic changes and typos fixed 10 15 2008 vfsmod w exe version 5 0 xx New release Water quality capabilities added 1 Major release version 5 0 0 of vfsmod w A component for water quality is add
96. d water balance the sediment trapping efficiency of the filter for the simulation case and the final geometry of the filter e filename osp Summary of the filter performance parameters and comparisons between source and filter areas f filename owgq This file is only created during run time when CWQ 1 in the input file IKW In this case the file will be created in the output directory of the application The water quality component is specified in the first line of this input file The parameters listed and results in the rest of the file depend on the type of water quality component selected Part II VFSMOD and UH User s Manual 65 1 6 Sample application A sample application case is shown by using input data collected at a NC State University experimental site Raleigh NC USA The input and output files can be found in the sample case included in the distribution package obtained from the internet sites 1 6 1 Inputs 1 6 1 1 Hydrological inputs files sample ikw and sample iso Description symbol INPUT Value Units Source area flow path length L SLENGTH 34 0 m Source area width ws SWIDTH 4 0 m Filter length L VL 8 655 m Filter width w FWIDTH 3 87 m Filter mean Manning s coefficient calculated n VN 0 40 s m Duration of the simulation DR 3603 s Number of nodes N 57 Number of different filter segments NPROP 14 Courant number
97. diment Delivery Ratio Filter Source ee Output Sensitivity Results Runoff Delivery Ratio Filter Source Enence Current Results File Dismiss Sensitivity File C vfsmod woutput igrS Ssens sen From this screen selected storm outputs are available for analyses Various plots of the outputs versus the inputs along with some statistics are available For example the Curve Number was varied from 76 to 95 and produced Source Runoff from 85 mm to 138 mm Selecting Plot Selected Output produces Part III VFSMOD W WindowsTM User s Manual 135 s Sensitivity Plot for Curve Number ljal x Curve Number 140 130 Filter Length 20 m Rain 151 1 mm 120 Base value 75 Range 75 to 95 in increments of 1 Sampled Mean 109 321 Source Runall mm Sampled Std Dev 16 8597 76 78 80 62 84 86 88 90 92 94 96 Ov 15 4 Curve Nurnber Dismiss Other plotting and summary options include Plot Selected Output Absolute Sensitivity This produces a plot of slope of the output versus the input example Slope of Source Runoff versus CN Plot Selected Output Relative Base Sensitivity This produces a plot of Output BaseOutput Input BaseInput BaseInput BaseOutput versus the Input Plot Selected Output Relative Sensitivity This produces a plot of slope of the output versus the input times Input Output versus the Input Output Sensitivity Results This pro
98. directories Since the vfsm and uh executables are written in Fortran and run at the Command window level all filenames should not contain any spaces Spaces in the filenames will cause unpredictable results For example my Project prj will cause problems Use something like my_Project prj this should work fine Part III VFSMOD W WindowsTM User s Manual 145 14 VFSMOD W Change History 11 18 2000 1 Added buttons in VFSMod Output Viewer for the remaining Output files ohy ogl 0g2 2 Made Nprop global and now the number of segments in the segment properties window updates when the user deletes and adds segments 2 5 2001 1 The first level of the output filenames default to the same as the project name The user can override this by changing the output filenames 2 In the igr files a check is made when the user changes the VL buffer length The new buffer length is checked against the segment properties If these are unequal a warning message box is displayed and the View Edit Segment properties window is opened 3 5 2003 vfsmod w exe version 2 00 uh exe version 1 06 vfsm exe version 1 06 1 Major revisions for the entire system Adding a number of buttons on pages to duplicate menu selections 2 Added Sensitivity Analysis 3 Added Uncertainty Analysis 4 Added Design Analysis 5 20 2003 Interim release vfsmod w exe version 2 20 xx uh exe version 2 06 vfsm exe version 2 06 1 Enable
99. distribution package would be read from the JVPUTS subdirectory After you execute the command you should see a screen as follows e e ACCAR e CeCe March 2005 v2 4 2 R Munoz Carpena J E Parsons UFL USA NCSU USA carpena ufl edu john _parsons ncsu edu PROGRAM GENERATE RAINFALL AND RUNOFF INPUTS FOR VESMOD Opening inputs sample2 inp Opening output sample2 out Opening output sample2 hyt Opening inputs sample2 iro Opening inputs sample2 irn Opening inputs sample2 isd During the run a set of the VFSMOD inputs is created in the INPUTS subdirectory sample2 irn sample2 iro sample2 isd sample2 iso Two more output files are created in the OUTPUT subdirectory that summarize the calculations performed sample2 out and sample2 hyt The content of these files is produced in verbose mode and is self explanatory Note that two more files are needed to run VFSMOD filter characteristics files ikw and igr and they are not created by UH but the user needs to set them up from field data To continue the example given above one could copy the sample files included in the distribution package sample ikw and sample igr into sample2 ikw and sample2 igr in the INPUTS subdirectory VFSMOD is now ready to be run by issuing the command Part II VFSMOD and UH User s Manual 71 vfsm sample2 2 2 Using the project file for input and output Another way to complete the exam
100. dition to these algorithms the GMCS NMS Global Multilevel Coordinate Search combined with a Nelder Mead Simplex is a powerful available alternative Lambot et al 2002 Ritter et al 2003 This consists in the sequential combination as described by Lambot et al 2002 of the global optimization algorithm developed by Huyer and Neumaier 1999 and the classical Nelder Mead Simplex Nelder and Mead 1965 In this work we have integrated the GMCS NMS within the VFSMOD W and its graphical user interface to allow the model users to perform automatic inverse optimization of the hydraulic and sediment transport parameters of VFSMOD W when Part I VFSMOD W Model Documentation 38 experimental data are available The inverse simulation of the flow or sediment transport parameters is carried out by minimizing an objective function OF that represents the error between measured and simulated values such that it can be defined as a nonlinear least squares problem by OF b F wY t Yt BY 49 i where the right hand side represents the deviations between observed Y and predicted Y time series hydrographs or sedimentographs using the parameter vector t is the time N is the number of measurements available and wi is the weight of a particular measurement which denotes the measurement error and is set equal to s 2 where s is the standard deviation of the measured data Lambot et al 2002 To perform the inverse calib
101. drologic and sedimentological conditions experienced by the VFS at the time of the study For example some studies report little reductions in low to moderately sorbed pesticides by VFS Yonts et al 1996 other researchers report significant reductions in similarly sorbed pesticides by VFS Tingle et al 1998 Review papers have concluded that a significant effect of VFS length on pesticide trapping was not uniformly observed in all of the studies primarily due to the fact that the removal depended largely on the pesticide properties nature of the runoff event and antecedent moisture content Reichenberger et al 2007 The most common approaches for attempting to predict VFS effectiveness are statistical analyses that attempt to relate physiographic characteristics of the VFS i e slope vegetation area ratio and VFS length to sediment and or contaminant removal Neitsch et al 2005 Liu et al 2008 Fox and Sabbagh 2009 These statistical approaches showed poor predictive power with little confidence in being able to accurately predict VFS reduction given the wide range of conditions likely to be experienced by the VFS Recent research has proven that simple physiographic characteristics of the VFS are not explicitly driving contaminant reductions Rather it is the hydrologic impacts of the physiographic characteristics on the VFS system that drive sediment and contaminant removal Fox and Sabbagh 2009 Sabbagh et al 2009 and Poleti
102. duces an output file with all of the statistics Here are examples for each of these al Sensitivity Plot for Curve Number k Ejjel x Curve Number 140 130 Filter Length 20 m Rain 151 1 mm z 2 7 Base value 75 3 g 110 Range 75 to 95 in increments of 1 5 100 3 Sampled Mean 109 321 90 Sampled Std Dev 16 8597 80 76 78 80 82 84 86 88 90 92 94 96 Ov 15 4 Curve Number FIGURE 16 Plot Selected Output Absolute Sensitivity Part III VFSMOD W WindowsTM User s Manual 136 a Absolute Sensitivity Plot for Curve Number Source Runoff mm Absolute Sensitivity N N N O N o Oo Oo N D N in 76 78 80 82 84 86 88 90 92 94 Curve Number Curve Number Filter Length 20 m Rain 151 1 mm Base value 75 Range 75 to 95 in increments of 1 Mean Abs Sen 2 728 Std Dev Abs Sen 0 1253 Ov Abs Sen 4 6 Dismiss FIGURE 17 Plot Selected Output Absolute Sensitivity a Relative Base Sensitivity Plot for Curve Number Source Runoff mm Relalive Sensiliv ily 76 78 80 82 84 86 88 90 92 94 96 Curve Number l Curve Number Filter Length 20 m Rain 151 1 mm Base value 75 Range 75 to 95 in increments of 1 Mean Rel Sen 2 362 Std Dev Rel Sen 0 0598 Cv Rel Sen 2 5 Dismiss FIGURE 18 Plot Selected Output Relative Base Sensitivity a Relative Sensitivity Plot for Curve Number
103. e the residual moisture content and is the Genuchten equation for the soil water ll lt 1 aygh i 1 h h yg van Genuchten enpirical parameter n and m van Genuchten shape parameters m 1 1 n h air entry pressure Note There are 4 parameters the user needs to define 6 6 yg and n The Brooks and Corey for the soil water characteristic curve is OCh 0 0 8 o gch S Jageh h lt ORC Q h 0 S 1 kaet ORC where Apc h inverse of air entry pressure A Brooks and Corey shape parameter Note There are 4 parameters the user needs to define 6 6 agcand A Part III VFSMOD W WindowsTM User s Manual 104 By default the program assumes that h is equal to 1 ayg or 1 agc However if the user wants to specify a fixed h it can be done by addinga value Vhe at the end of the file If the numberis not present the program calculates h according to the rules above IKUNSTYPE is an integer to select the unsaturated hydraulic conductivity curve type with values 1 van Genuchten 2 Brooks and Corey and 3 Gardner s The parameters for these curves will be Equation IKUNSTYPE PAR1 PAR2 can Genuchten 1 VGM Brooks and Corey 2 BCETA BCALPHA Gardner 3 GDALPHA The van Genuchten equation for the hydraulic conductivity curve is K 0 K S 1 1 sm h lt h K 9 K h gt h Note There are 2 parameters the user needs to define K and n The Brooks an
104. e average flow conditions calculated as described above c filename og2 This file contains the flow characteristics at the singular points 1 3 in and out as defined in Part I of this manual of the filter for the simulation period for the same 100 steps described above d filename osm This file contains a summary of the most relevant input parameters and output results including a sediment and water balance the sediment trapping efficiency of the filter for the simulation case and the final geometry of the filter e filename osp Part III VFSMOD W WindowsTM User s Manual 113 Summary of the filter performance parameters and comparisons between source and filter areas f filename owq This file is only created during run time when CWQ 1 in the input file IKW In this case the file will be created in the output directory of the application The water quality component is specified in the first line of this input file The parameters listed and results in the rest of the file depend on the type of water quality component selected Currently only a pesticide component is available in this version An option for pesticide partitioning and degradation between events is also included details are provided in Mufioz Carpena 2012 Contents of the OWQ file with no degradatibetween events on selected Parameters for Water Quality Type of problem Pesticide trapping BAYER Particion coefficient Kd 35 000000 L Kg
105. e buttons Run This Project executes the current project Graph Hyetograph produces bar graphs of rainfall intensity versus time the user selects the output rainfall hyetograph file irn Graph a Runoff Hydrograph produces a runoff hydrograph the user selects the runoff file iro View Output Files opens a text window with a user selected output file Part HI VFSMOD W WindowsTM User s Manual 93 For Tips on running troubleshooting VFSMOD see section in Part I T Inputs Outputs inop inputs for the source area for UH irn rainfall hyetograph input for vismod runoff hydrograph from the source iro area input for vfsmod sediment properties for the sediment isd filtration submodel summary of the inputs and outputs out from UH detailed summary of of MUSLE calculations and the runoff hyt hydrograph 5 1 UH Input File Editing Input Filename amp gt uh Editing inputs sample2 inp ee t fase finputs sample2 inp 25 Rainfall rm 85 Curve Number Rainfall Event and Runoff Source Area Slope as a fraction o2 0 Erosion Parameters 6 Storm Duration h II Storm Type 100 Length m along the slope 5 Area ha Soil Erodibility K Metric t ha h ha MJ mm typical 0 01 0 07 Percent organic matter for example 1 Crop Factor C Rainfall Factor R Williams Recommended Save Soil Type dp Particle Class Diameter fern 2 200 10 4 or 1 fo
106. e coastal regions of Alaska Type II is used to represent most of the remaining areas of the US Type III is used for storms along the Gulf coast southern Florida and coastal areas of the eastern US Soil Erodibility Factor K This is the USLE soil erodibility factor If K lt 0 then K is computed based on texture and organic matter See Universal Soil Loss Equation USLE on page 24 C Factor The USLE Crop factor See MUSLE Crop factor C on page 170 P Factor The USLE P Practice factor See Contour factor P values for MUSLE equation in UH on page 176 Soil Type for the surface soil layer See Definition on page 73 dp Particle Class Diameter ranges from 15 to 200 um Tables based on soil types are used when the user specifies 1 See Sample application on page 66 Rainfall Factor The rainfall factor for the modified storm version of USLE Select the method to compute the storm R factor in MUSLE not present or 1 selects Foster s Method 2 selects Williams method and 3 selects the CREAMS GLEAMS method The User can change the name of the input file in the Input Filename window The inputs can be saved using the Save button In this case the window remains open for further editing This is helpful to create multiple inputs The Close and Save will save the inputs and close this window The Help button gives this help screen Part III VFSMOD W WindowsTM User s Manual 95
107. e focus was on immobilization of the pesticide by the VFS due to the assumption that the most significant loading threat was due to surface runoff in the immediate runoff event The previous research also proposed a procedure linking a VFSMOD W with the proposed empirical trapping efficiency equation Sabbagh et al 2009 and Poletika et al 2009 For data sets with sufficient information the linked numerical and empirical models significantly improved predictions of pesticide trapping over conventional equations based solely on physiographic characteristics of the vegetated filter strip R2 0 74 with a slope not significantly different than 1 0 and intercept not significantly different than 0 0 Others Poletika et al 2009 Mu oz Carpena et al 2010 and Fox et al 2010 further evaluated VFSMOD W which included the empirical pesticide trapping Part I VFSMOD W Model Documentation 10 efficiency equation The integrated numerical model was capable of predicting runoff volume sediment and chemical reductions by the VFS under both uniform and concentrated flow in good agreement with the measured reductions Poletika et al 2009 A new component to calculate in filter pesticide distribution and degradation between events is now available for continuous simulations in long term environmental assessments conducted by regulatory agencies US EPA EU FOCUS Details are provided in Mufioz Carpena R 2012 2 3 2 Solute Transport I
108. e is on at least 5 of the surface throughout the year b Hydraulic condition is based on combination factors that affect Infiltration and runoff including a density and canopy of vegetative areas b amount of year round cover c amount of grass or close seeded legumes d percent of residue cover on the land surface good gt 20 and e degree of surface roughness Poor Factors impair infiltration and tend to increase runoff Good Factors encourage average and better than average infiltration and tend to decrease runoff Part IV VFSMOD Appendices 167 Runoff curve numbers for other agricultural lands Table 2 2c From USDA NRCS 210 VI TR 55 2nd Edition June 1986 Cover Description Curve numbers for hydrologic soil group Hydrologic Cover Type Condition A B C D Poor 68 79 86 89 Pasture grassland or range continu Fair 49 69 79 84 ous forage for grazing Good 39 61 74 80 Meadow continuous grass pro 30 58 71 78 tected from grazing and generally mowed for hay Brush brush weed grass mixture FOU 48 67 uh with brush the major element Fair 35 56 70 77 Good 39 48 65 73 Woods grass combination orchard Poor 57 73 82 86 or tree farm 4 Fair 43 65 76 82 Good 32 58 72 79 Poor 45 66 77 83 Woods Fair 36 60 73 79 Good d30 55 70 77 Farmsteads buildings lanes drive 59 74 82 86 ways and surrounding lots a Poor lt 50 gr
109. e parameter The base value shown is the value in the base project files These values are used in some of the analysis screens In addition to setting the values the user can load a different base project and once the inputs are set do the simulations If the user would like to also do the analysis for the VFSMOD parameters they can switch to the VFSMOD screen Part III VFSMOD W WindowsTM User s Manual 133 E gt Filter Strip VFS Parameter Selection Filter Strip Sensitivity Parameter Selections Base Project Files sample prj Base Values 4 788 VFS Parameters V Green Ampt Ksat VKS cmh m Green mpt Theta Initial cm 3 cm 3 125 Particle Class Diameter dp cm M 1 gt Model Selects dp based on Soil Type 2200 Min Values a788 ps fae ff ao o p Mfo SetUH Parameters Max Values 4 788 Increment m j2 o pz fo D hd nsional Sensitivit t 1 Do Simulations Load areren Base Cancel Help roject Similar to the UH screen only selected parameters are available for sensitivity analysis Currently the parameters are the saturated vertical conductivity and initial water content for the Green Ampt infiltration submodel for the filter strip and the Dp Particle Class diameter and SS Media Element Spacing parameters Selection of each parameter is done with the Check boxes and setting minimum and maximum values along with an increment for the sensitivity analysis Once th
110. e procedure discussed by Vieux et al 1990 a find dx for integration rule b begin vector formation element by element b 1 do integration point loop b 2 plug the element vector into the b vector c plug in the boundary condition b 1 0 1 16 SOLVE A B X N NBAND Solve the LUD transformed matrix A using a backward and forward substitution with A X b since A L U then L U X L U X L Y b solving L Y b forward substitution U X y backward substitution 1 17 CONVER N X XM MFLAG This subroutine checks for convergence as max X 1 xm lt 10 8 max X A If there is convergence return MFLAG 1 otherwise MFLAG 0 1 18 UPDATE N X X0 Refresh values of the X vector this is x x 1 19 FLOW N XT QT Calculate the flow vector at each iteration or time step by using Manning s equation Part IV VFSMOD Appendices 154 1 20 GRASSED TIME N QIN NODEX ICOARSE COARSE This subroutine is the driver for the sediment transport problem on grass filter strips Notice that all units are CGS system cm g s including Manning s n to follow the origi nal method as described by the authors Tollner et al 1977 Barfield et al 1979 Hayes et al 1979 1984 Wilson et al 1981 For computational purposes the filter is divided into the following sections notice the change in properties as sediment is deposited A t top flat face of sediment wedge B t d
111. e project file for input and output Versions 1 04 and later now allow the user to create project files These files contain the list of input and output files for the model This enables the user to mix and match inputs from multiple simulation scenarios Each line of the project file contains a keyword denoting the type of input and output file and the filename A project file sample prj for the sample inputs in UNIX and Windows 9x NT 2000 XP contains the following line Part II VFSMOD and UH User s Manual 55 UNIX Windows 9x NT 2000 XP ikw inputs sample ikw ikw inputs sample ikw igr inputs sample igr igr inputs sample igr im inputs sample irn im inputs sample irn iro inputs sample iro iro inputs sample iro isd inputs sample isd isd inputs sample isd iso inputs sample iso iso inputs sample iso iwq inputs sample iwq iwq inputs sample 1wq og1 output sample og1 og1 output sample og1 og2 output sample og2 og2 output sample og2 ohy output sample ohy ohy output sample ohy osm output sample osm osm output sample osm osp output sample osp osp output sample osp owq output sample owq owq output sample owq The project file in this example sample prj would be saved in the VFSMOD directory where the executable vfsm or VFSM EXE is To execute the model with the project file the following would be entered vfsm sample prj In this example the input files would be read from the inputs subdirectory and the output files would be created in t
112. e regular Manning s ones While the first two values RNA and VN are used in overland flow calculations VN2 is only used for sediment deposition calculations based on the original work at the University of Kentucky on sediment trapping Part II VFSMOD and UH User s Manual 86 Part I VPSMOD W Windows User s Manual 1 VFSMOD Model Description VFSMOD is a field scale mechanistic storm based model designed to route the incoming hydrograph and sedimentograph from an adjacent field through a vegetative filter strip VFS and to calculate the outflow infiltration and sediment trapping efficiency The model handles time dependent hyetographs space distributed filter parameters vegetation roughness or density slope infiltration characteristics and different particle size of the incoming sediment Any combination of unsteady storm and incoming hydrograph types can be used As an aid to set up the model inputs the distribution package includes a utility uh that creates synthetic model inputs based on the NRCS SCS design storm for a given location and soil type The utility implements the NRCS SCS curve number unit hydrograph and Modified Universal Soil Loss Equation MUSLE concepts to produce ready to use input files for VFSMOD These inputs are rainfall hyetograph field inflow hydrograph and field sediment inflow and characteristics The model has been field tested for different soil and climatic conditions in the
113. e simulations are complete the user can do some analysis using the VFSMOD Selected storm outputs are saved in files for each parameter selected For example if Curve Numbers are selected then the suggested name for the output of the sensitivity parameters is UHCNsens sen CO lolx Fle Edit Block Convert Options View Help OR Gre Oloroy sMOly tlt la a unCNsens jep sen Jep Parameter 72 65 85 2 Rainfall mm and Buffer Length m gt base uh sample20c lis base vfs sampleOc prj 80 8 655 Input sensitivity parameters a Base O 72 48 7 6 4 788 1 125 2 2 23 03 1151 566 22 97 1149 131 0 65 48 7 6 4 788 i 125 2 2 14 75 737 344 14 65 733 088 O 67 48 7 6 4 788 1 125 2 2 16 92 846 082 16 83 842 178 0 69 48 7 4 788 1 125 2 2 19 24 961 847 19 15 957 820 O 71 48 7 6 4 788 1 125 2 2 21 71 1085 634 21 62 1081 559 0O 73 48 7 6 4 788 1 125 2 24 34 1216 834 24 26 1213 802 O 75 48 7 6 4 788 1 125 2 2 27 16 1357 888 27 08 1354 619 O 77 48 7 6 4 788 1 125 2 2 30 08 1503 985 30 02 1501 926 O 79 48 7 6 4 788 1 125 2 2 33 24 1662 017 33 18 1659 850 O 81 48 7 6 4 788 1 125 2 2 36 60 1830 102 36 53 1827 523 O 83 48 7 6 4 788 1 125 2 2 40 16 2007 972 40 08 2005 109 O 85 48 7 6 4 788 1 125 2 2 43 90 2195 140 43 82 2192 298 output uhCNser 4 512 6 333 5 981 6 104 6 153 5 109 5 346 4 136 4 244 4 656 4 941 4 920 11378 57 6046 01 7358 90 8850 20 10534 94 122
114. e to the upslope edge of the filter strip If the user changes VL the length of the filter strip then a check is made of the segment properties to ensure that the last point in SX is equal to the new buffer strip length If it is not then the View Edit Segment Properties screen is opened and a warning message box is shown reminding the user to fix the segment properties data Part III VFSMOD W WindowsTM User s Manual 100 0 10057 0 20114 n 0 30171 2 Lg gt Fa wi 0 40228 0 50285 1 81755 3 6351 5 45265 7 2702 9 08775 Distance m Copy Plot to Clipboard Print Plot Edit Plot 6 2 VES Infiltration Soil Properties iso After opening the iso file the user sees the regular soil input selection window Notice that a new option has been added for shallow water table Fig iirsmrod wincouseatr y St E nnn tell Teh File Source Area UH Filter Strip VFS Design Sensitivity Uncertainty Calibration Options Window Help gt vfsm Project sampleliso prj ming Dees YCWvenaiy Filter Strip Project File samplet iso prj Save Help f Include Water Quality 7 vfs od w in Input Overland Flow Inputs See p aa R Files Infiltration Soil Properties file iso Jinputs sa pleisomod iso Infitation Soil Prope Cs Table Present Buffer Vegetation Pro Vertical Saturated K VKS m s or cmh 00000000389 0014004 Storm Hvetogreph Average Suction at the Wetting Front Sav m
115. eck here to tum on Debug Option usually not checked only on for this run TT Associate Project files pri and lis with vfsmod w option not yet available To Register Email the vfsmod w cfg File to carpena ufl edu Thank You FIGURE 2 Checking the root directory before running the calibration mode Part III VFSMOD W WindowsTM User s Manual 122 The recommended calibration procedure is to first calibrate the flow hydrograph component of the model i e run menu item Hydrograph first then modify the original project with the calibrated flow parameters and then use the modified project to run the menu item Sedigraph In this way the balances for sediments will be related to the optimized water balances KJ VFSmod Windows Editor v4 0 7 File Source_Area UH Filter_Strip VFS Design Sensitivity Uncertainty Me Options Window Help aa Hydrograph Sedigraph FIGURE 3 Hydrograph and Sedigraph options within the calibration mode After the user has selected one option i e Hydrograph the program requests to select the name of the project to be run it is located in the root directory of VFSMOD W i e c vfsmod w and also the measured data file hydrograph in this case x File Source_Area_ UH Filter_Strip_ VF5 Design Sensitivity Uncertainty Calibration Options Window Help vismod w x Please Open a Current VFSmod Project File FIGURE 4 Opening a project file to run the
116. ed New water quality module Only the option for calculating pesticides based on Sabbagh s algorithms is included eA generic pollutant simulation engine based on the Transport and Reaction Simulation Engine TaRSE is left prepared on the windows interfase to be included in future releases 2 Some modifications to the Design option to include the new water quality capability 06 2011 Changes in vfsm v4 x x WT Added new subroutine to solve the soil infiltration problem for unsteady rain in the presence of a shallow water table using a modified Green Ampt infiltration model as proposed by Salvucci and Entekhabi 1995 Chu 1996 and work by the authors of this program The method was extended to include mass balance on the surface as proposed by Skaggs 1982 and Khaleel 1982 in Hydrologic modeling of small watersheds ASAE mon no 5 and Chu 1978 Water Resour Res An extended soil input file iso is required in this case If an additional numeric parameter WTD water table depth m is found in the second line of the standard iso file then soil characteristic curve inputs are read below the second line in iso and the new subroutine gasubwt f is called Notice that SAV and OI values are ignored Details of the structure of the extended iso input file are provided in the user manual and sample files are also the distribution package see sampleWT prj New surface ponding forcing scheme NPFORCE 1 when overland flo
117. ed by averaging the elementary effects and this eliminates the need to consider the specific points at which they are computed Saltelli et al 2005 Morris 1991 recommended applying u or u thereof to Part I VFSMOD W Model Documentation 34 rank parameters in order of importance and Saltelli et al 2004 suggested applying the original Morris measure o when examining the effects due to interactions To interpret the results in a manner that simultaneously informs about the parameter ranking and potential presence of interactions Morris 1991 suggested plotting the points on a n o Cartesian plane Figure 13 Because the Morris method is qualitative in nature it should only be used to assess the relative parameter ranking 10 CN DP K y R VKS EWIDTH 0 04 0 06 0 08 0 10 Figure 13 Morris sensitivity graph 4 2 2 Extended FAST A variance based method like the Fourier Amplitude Sensitivity Test FAST can be used to obtain a quantitative measure of sensitivity Cukier et al 1973 1978 Koda et al 1979 FAST decomposes the total variance V o y of the model output in terms of the individual factors X f X X gt X using spectral analysis so that V o y Vit Vat Vat t V R 44 where V is the part of the variance that can be attributed to the input factor X alone k is the number of uncertain factors and R is a residual corresponding to higher order terms The first o
118. er a local OAT method is that it results in the ranking of parameter importance and provides information not only about the direct first order effect of the individual factors over the output but also about their interaction higher order effects Different types of global sensitivity methods can be selected based of the objective of the analysis the number of uncertain input factors the degree of regularity of the model and the computing time for single model simulation Cukier et al 1973 1978 Koda et al 1979 Morris 1991 Saltelli et al 2000a 2004 2005 Sobol 1990 Wallach et al 2006 These two methods of uncertainty and sensitivty analyis are presentated in the following sections 4 1 Local OAT Sensitivity Analysis Haan et al 1995 outlined the statistical procedure for evaluating hydrology and water quality models Their procedure included conducting sensitivity analysis generating probability distributions for model inputs generating probability distributions for the model outputs and using the probability distributions of the model outputs to assess uncertainty Using an example model they conducted a sensitivity analysis to identify the input parameters that have the most impact on the outputs The absolute sensitivity S of a given output O relative to input parameter P is defined as _ 0O S OP 38 i The relative sensitivity S of the output parameter with respect to changes in the ri input p
119. er increasing the potential for the kinematic shock problem VFSMOD implements a Petrov Galerkin formulation non standard finite element to solve equations and 2 This solution procedure reduces the amplitude and frequency of oscillations with respect to the standard Bubnov Galerkin method Mufioz Carpena et al 1993a thus improving the model stability and the sediment transport predictions which depend on overland flow values 2 2 Sediment Transport The hydrology model is linked to a model for filtration of suspended solids by artificial grass media developed and later tested for field conditions Barfield et al 1978 1979 Hayes et al 1979 1984 Tollner et al 1976 1977 Wilson et al 1981 It is based on the hydraulics of flow transport and deposition profiles of sediment in laboratory conditions The model presents the advantage of being developed specifically for the filtration of suspended solids by grass ENTRY Field Wedge Zone I Suspended Load Zone Flow needed at points 1 see text L Lit Aout Figure 3 Filter description for the sediment transport algorithm The University of Kentucky algorithm considers that during a rainfall runoff event field runoff reaches the upstream edge of the filter with time dependent flow rate Ginlem s and sediment load g g cm s The vegetation produces a sudden increase in hydraulic resistance that slows the flow lowers its transport capacity g g cm s and pr
120. esign run Information for standard USDA soil types Green Ampt infiltration inputs and vegetation covers spacing height to be used in the analysis can be found in this document For each combination of inputs a new project must be created and the model executed If the problem is to be prepared manually UNIX and DOS versions it is usually more efficient to create a naming convention for each project that reflects the simulation run characteristics The proposed sequence is to prepare the UH inp input files combination of source area soils types and design storms first and then process them with UH to produce the corresponding VFSMOD inputs iso irn isd Afterwards the user creates the project files prj one per simulation as combination of the UH outputs and modification of the remaining input files igr ikw iso as needed Each file must be then processed with VFSMOD and the SDR and RDR results obtained from the osm files From these outputs SDR or RDR versus filter length the user can obtain the optimal filter characteristics for each return period and soil type when overlaying the Part I VFSMOD W Model Documentation 41 pre defined sediment TMDL expressed in terms of a desired filter effectiveness SDR or RDR In the MS Windows VFSMOD W modelling system versions 2 x and up this task is automated The projects for each combination of design inputs are automatically created within the program GUI af
121. essment with VFSMOD 1 VFS pesticide residue between runoff events Technical Report University of Florida Agricultural Engineering Department URL Accessed June 15 2014 http abe ufl edu carpena vfsmod FOCUS VFSMOD Continuous Sim_Report PesticideResidue_MunozCarpena pdf Nash J E and J V Sutcliffe 1970 River flow forecasting through conceptual models Part 1 A discussion of Principles J Hydrol 10 282 290 Nelder J A and R Mead 1965 A Simplex Method for Function Minimization Com puter Journal 7 308 313 Neitsch S L Arnold J G Kiniry J R Williams J R Soil and Water Assessment Tool Theoretical Documentation Version 2005 USDA Agricultural Research Service Temple TX 2005 www brc tamus edu swat downloads doc swat2005 SWAT 202005 20theory 20final pdf NRCS The National Conservation Buffer Initiative National Resource Conservation Ser vice USDA Washington DC 1999 p 53 Parsons J E R B Daniels J W Gilliam and T A Dillaha 1991 The effect of vegeta tion filter strips on sediment and nutrient removal from agricultural runoff In Proc of the Environmentally Sound Agriculture Conference ed A B Bottcher K L Campbell and W D Graham 324 332 Orlando FL April 16 18 1991 Part I VFSMOD W Model Documentation 51 Parsons J E and R Munoz Carpena 2001 Impact of Uncertainty on the Design of Veg etative Filter Strips Statistical Methods in Hydrology for the 2001 ASAE Annua
122. essteesesresessesreseent 31 4 2 Global Sensitivity and Uncertainty Analysis cccceccesseeseessesceeseceseeseceecesceeeeeeeseeseeenees 33 421 The Mortis Methods jesse erranti EE E E EE hater a 33 422 Extended FAS T raseria oreore inn Geass cag E E E etd ENA 34 5 Inverse Calibration aniis nann n E E a r ara cay 37 6 D sign Proc d Te ossidasi tariak EE aE E E E ES a Aa Eiaa 40 T Potential Users and Applications of the Modelling System s snsssnssseeseeseeseeseee 42 8 Known Limitations and Applicability of the Models 0 cc eseeseesceteeneeeneeeeeees 43 8 1 Known Limitations of the Model cc eceeessesseeceeceseneeeeesecnecseeeeaecaeeeseceecnaesessesaeeasenees 43 8 2 Future Research Model Releases ccscssesessssesesseeeceeceeeseeseesecseseceeceaeeeeeceecneeaeeeeeaeeaeenees 43 9 Distribution and Training ccc scecos Sear inesesctatscasecaseccisdasyeccisls Geteuceaseicdaae sons aveiet 44 10 Acknowledsements x cicdcsdncececish case acdsee dis ce Raed catia ues Sse 45 11 REIT ech ae Aah ata eae et ea ag Sia E E 46 Part Il VFSMOD and UH User s Manual 0 eccceccceceseceseceeeeeeseeeeeceeeeeeseecsuecneeneneenaees 53 1 VESMOD User sana cc iesiyce withers Sealer aa ma ee enous 53 1 1 Obtaining VESMOD AEAN ENEI E EA A A E E 53 1 2 Installing and running VFSMOD 20 00 ecceccecseeeceeseescecseceeeeseceeceseeseeseseceseeeeeeaeeeeeeaeeeeeaees 53 1 2 1 Installing for a DOS command prompt window under W
123. factor P values for MUSLE equation in UH Contour factors P Factor from Wischmeier and Smith 1978 Land Slope Contour Factor Maximum Length FPACT ft m 1 2 0 6 400 122 3 5 0 5 300 91 6 8 0 5 200 61 9 12 0 6 120 36 13 16 0 7 80 24 17 20 0 8 60 18 21 25 0 9 50 15 3 7 References for Soils and Vegetation data References for the above Tables are Knisel Walter G F M Davis R A Leonard 1992 GLEAMS Version 210 Users Man ual Pre Publication Copy US Department of Agriculture Agricultural Research Ser vice Available from University of Georgia Coastal Plain Experiment Station Bio and Ag Engineering Tifton GA UGA CPES BAED Publication No 5 259 pp McCuen R H W J Rawls and D L Brakensiek 1981 Statistical Analysis of the Brooks and Corey and the Green Ampt parameters across soil textures Water Resour Res 17 4 1005 1013 Rawls W J and D L Brakensiek 1983 A procedure to predict Green Apmt infiltration parameters Adv in Infiltration pp 102 112 ASAE Pub no 11 83 U S NRCS Formerly Soil Conservation Service National Engineering Handbook Hydrology Section 4 1972 and USDA ARS 41 172 1970 USDA NRCS 210 VI TR 55 2nd Edition June 1986 Wischmeirer W H and D D Smith 1978 Predicting rainfall erosion losses a guide to conservation planning Agriculture Handbook No 537 USDA Washington DC 58 pp Part IV VFSMOD
124. ficiencies are possible especially if you don t get much runoff at the end of the filter due to infiltration Does clay in your sediment samples aggregate into larger particles Remember also that clay even if in small amounts could also deposit depending on the velocity and flow regime x If the soil characteristics are known in other words the soil silt and clay balance of deposited sediments are known so that NPART 7 is it best to use a weighted average of the three to determine the values for sediment particle size DP and sediment particle density SG or do you have another suggestion In essence there are 3 characteristics that define the sediment transported into the filter These are DP median particle size or d50 SG sediment particle density and Vf fall velocity of the particle in water One other characteristic is important here COARSE or the of particles of diameter gt 0 0034 cm One important thing to remember is that the soil as a whole is not transported by runoff from the source area into the filter but usually only those particles that the energy velocity of the flow can carry at any given time This means that there is usually a selection of the finer soil by runoff during transport If you had actual sediment samples from runoff samples collected at the ed of the source area or field you could plot the cumulative frequency graph of particles less than a diameter and then choose d50 and COARSE from there If
125. following sections herein present the theory behind the methods implemented in UH 3 1 Generation of Synthetic Rainfall Hyetographs 3 1 1 Equations for storm types II amp III For storm types II and III the equations presented by Haan et al 1994 are used to generate the hyetographs The equation is p t _ en 24 04 M Pos 0 3 347 0 04 ae where T t 12 with t in hours p24 the 24 hour total rainfall in cm For storm durations less than 24 hours the ratio of p t p24 is used to derive the amount of rainfall at time from the total rainfall for the period The computation procedure follows that given by Haan et al 1994 3 1 2 Equations for storm types I amp IA Based on tabulated data Haan et al 1994 pg 48 the fitted equations using Mathematica Wolfram 1999 are e Storm type I P__ 0 4511 t 9 995 l 0 1617 Pog 3 0163 t 9 995 0 013 0 5129 3 0163 t 9 995 0 013 gt 0 0 5853 3 0163 t 9 995 0 013 lt 0 11 With an Root Mean Square Deviation RMSD 0 0088 and 1 3 363 Part I VFSMOD W Model Documentation 16 e Storm type IA 40 4228 p 0 0843 0 3919 t 7 190 20 _ 7 O60 0 2567 k P 0 3919 7 960 T0796 03367 a With an RMSD 0 0033 and y7 1 539 The comparison of fitted vs real values can be seen on Figure 6 fitted IA fitted P Po4 0 5 10 15 20 Time hours Figure 6 NRCS storm types fitted by propose
126. g hydrograph from the adjacent field This also represents a linkage with measured data or to other water quality models describing the incoming runoff and polutant from the field source area The kinematic wave represents an acceptable approximation to overland flow when the Froude number and kinematic wave number are within certain limits Woolhiser and Ligget 1967 LS g Fr KATS and k 5 gh v gt 10 4 Part I VFSMOD W Model Documentation 4 The rainfall excess i is calculated from the hyetograph and a modification to the Green Ampt infiltration method at every time step Mufioz Carpena et al 1993 The overland flow model was coupled for each time step with an infiltration submodel based on a modification of the Green Ampt equation for unsteady rainfall Chu 1978 Mein and Larson 1971 1973 Skaggs and Khaheel 1982 Mufoz Carpena et al 1993b KMS fy K 5 P K t t 1 F MS ln 1 o 6 a v where fp is the instantaneous infiltration rate or infiltration capacity for ponded conditions m s K is the saturated vertical hydraulic conductivity m s M 0 9 is the initial soil water deficit m m S is the average suction across the wetting front m F is the cumulative infiltration after ponding m F is the cumulative infiltration for the event m is the actual time s tp the time to ponding and is the shift of the time scale to correct for not having ponded condi
127. ge difference in loads between g and g 2 for those parts of the event when flow was low beginning Part II VFSMOD and UH User s Manual 68 and tail whereas most of the sediment in the suspended sediment zone was retained at high flow rates when the sediment by passes the wedge 0 2 o 0 15 1000 2000 3000 2 g 8 0 1 lt oO E O 3 0 05 0 0 500 1000 1500 2000 2500 3000 3500 Time s a 100 incoming g is lower area 80 E 2 2 60 D ze 5 40 E 3 om 20 0 1000 1500 2000 2500 Time s The sediment balance for the simulation was Total sediment inflow 116 40 g cm 45 030 Total sediment outflow 0 4195 g cm 162 3 Trapping efficiency T 99 6 VFSMOD finally predicts the final sediment wedge geometry and deposition over the filter as Sediment wedge depth Y t 0 85 cm Sediment tail at field X t 15 05 cm Sediment wedge length X t 4 55 cm Effective filter length L t 860 95 cm Part II VFSMOD and UH User s Manual 69 Sediment depth in low section DEP 0 145 cm Rough mass balance wedge depth error lt 1 1 6 2 4 Filter performance indicators file sample osp Parameter Value Units Source Area input 136 00 m2 Source Flow Length input 34 00 m Source Area Width input 4 00 m Filter Strip Length input
128. ghness at the sediment wedge ICO 0 for NO or 1 for YES Note 0 NO is recommended in most cases 0 see the Users Guide for more information Save Continue Editing Save and Close Close Help SS spacing of the filter media elements cm See Vegetation types for VFS s on page 163 VN filter media grass Manning s n See Vegetation types for VFS s on page 163 0 012 for cylindrical media scm 3 H filter media height cm See Vegetation types for VFS s on page 163 VN2 bare surface Manning s n See Manning s roughness coeficient n on page 162 Part III VFSMOD W WindowsTM User s Manual 106 ICO integer flag to feedback the change in slope and surface roughness at the sediment wedge for each time step O no feedback 1 feedback See Sample application on page 66 6 4 Incoming Sediment Characteristics isd a B vfsm Editing C vfsnod w inputs sample isd oloje Incoming Sediment Properties file isd _ inputs sample isd Sediment Properties Incoming flow sediment Porosity of deposited sediment concentration g cm 3 Cl 034 fraction POR 434 Incoming sediment particle class Portion of Particles from incoming NPART 7 sediment with diameter gt 0 0037 cm 5 l COARSE ranges from 0 0 1 0 Sediment particle size diameter d50 Sediment particle density g cm 3 cm DP read only if NPART 7 1 0013 SG read only if NPART 7 2 65 Sa
129. gram calculates those internally based on the shallow water table conditions _ eoma VFSmod Windows Editor v 5 1 9 File Source Area UH Filter_Strip_ VFS Design Sensitivity Uncertainty Calibration Options Window Help GCE Jc wtsmod w 7 m Help sampleliso prj B gt vfsm Project sampleliso prj Working Directory Filter Strip Project File Save T Include Water Quality 5 vfsm Editing C vfsmod w inputs sampleisomod iso keda inputs sampleisomod iso M Shallow Water Table Present Input Overland Flow Inputs Infiltration Soil Properties file iso Files Infiltration Soil Proper r Green Ampt Infiltration Parameters Buffer Vegetation Prop Vertical Saturated K VKS 00000000389 m s or 0014004 cmh Incoming Sediment CI 37304 Average Suction at the Wetting Front Sav m Storm Hyetograph Volumetric Water Contents cm 3 cm 3 Initial Water Content Ol 125 Maximum Surface Storage SM m fo Fraction of the filter where ponding is checked 0 lt SCHK lt 1 4 Saturated Water Content 0S 39 Source Area Storm Fi Sediment Transport Flow through VFS Detailed Hydrographs Water and Sediment E Overall Summary Ol Calculation Method Surface Soil Moisture Avg Soil Moisture Brooks and Corey Show parameters C Gardner Save Continue Editing Save and Close Close Help Water Table Depth WTD m 10 7 Soil Characteriste Curves r So
130. h Vegetative filter strip model vfsmod Sample project for uh Sample project for vfsmod Global Sensitivity model This information the Users Manual Directory containing the inputs Sample Overland flow inputs for VFSMOD Sample Buffer vegetation inputs for VFSMOD Sample Rainfall hyetograph for VFSMOD Sample Runoff hydrograph for VFSMOD Sample Incoming sediment characteristics for VFSMOD Sample Infiltration soil properties for VFSMOD Sample overland flow inputs created by uh Sample inputs for uh Sample infiltration soil properties created by uh Directory containing the outputs from uh and VFSMOD Sample output from uh used as a placeholder FORTRAN source code for UH FORTRAN source code for VFSMOD Sample Calibration Hydrograph Sample Calibration Sedigraph Inverse calibration program Part III VFSMOD W WindowsTM User s Manual 87 inverse inverse inverse inverse inverse Outputs inverse inverse inverse inverse inverse irn isd ikw iro iso ogl og2 ohy osm osp Patterns pattern pattern pattern pattern pattern pattern pattern igr ikw inp irn iro isd iso And after your first execution of vfsmod w then the Options file is written to this directory vfsmod w cfg The Directory for Saving Project Files should be C vfsmod w After this is done vfsmod w should be ready to analyze your vegetative filter strips Part III VFSMOD W Wind
131. he output subdirectory In general the project file contains all of the keywords which are Inputs Outputs igr buffer properties for the sediment ogl detailed time series describing the filtration submodel sediment transport and deposition within the buffer ikw parameters for the overland flow solution og2 detailed information on the singular points defined in the theory section of the manual irn storm hyetograph ohy detailed outputs on the inflow and outflow hydrographs iro storm hydrograph from the source area osm detailed summaries of the water and sediment balance final geometry of the filter isd sediment properties for the sediment osp overall summary of filter performance filtration submodel with comparisons between the source area and filter Part II VFSMOD and UH User s Manual 56 iso soil properties for the infiltration submodel iwq water quality transport submodel owq water quality transport balance and output details 1 4 VFSMOD input files All files are in FORTRAN77 free format The inputs are distributed among 6 files filename ikw parameters for the overland flow solution filename irn storm hyetograph filename iro runoff from the adjacent field into the VFS filename iso soil properties for the infiltration model filename igr buffer properties for sediment filtration model filename isd sediment properties for sediment filtration model Note that filename could
132. hift both time scales for the incoming hydrograph and hyetograph accordingly files i ro and irn p How do I handle multiple storms Part II VFSMOD and UH User s Manual 82 VFSMOD W is a single event simulation program Each storm should be handled in independent project files In your application case there are in fact two separate storms in your files first starting at 900 s and second at 44800 s Moreover only the second one seems to produce any runoff at all Although you can run them together see norman1 prj results you should separate them as individual storms or likely just run the second event only q How do I select N CR and MAXITER in the ikw file This is explained in the User s Manual 5 7MB Here are some tips N the number of finite elements does not have to be the same as the number of physical land segments you measured in the field i e 15 Instead you can numerically subdivide these into a sufficiently large number of elements to give better numerical stability to the solution Doing this will result into much faster runs The way this is done is that elements within each land segment characterized by a Manning s n and slope has the same values as the segment The program does this internally when you select N gt NPROP e Changing the CR to a smaller number will also slow down the simulation we recommend 0 8 e The MAXITER should probably be left at 350 If the program is not allowed to conve
133. ic pollutants now under development Reference G J Sabbagh G A Fox A Kamanzi B Roepke and J Z Tang 2009 Effectiveness of Vegetative Filter Strips in Reducing Pesticide Loading Quantifying Pesticide Trapping Efficiency J Environ Qual 2009 38 762 771 6 8 VES Description of the Output Files a filename ohy This file contains information related to the hydrology side of the problem overland flow and infiltration The content of this the file is controlled by the input parameter IELOUT The first part of the file summarizes information read from the ikw iso and irn input files along with some of the calculated parameters needed for the simulation The second part of the file contains the inflow hydrograph from iro rainfall excess ie calculated with the Green Ampt model and the output hydrograph from the filter Only 100 time steps are printed to this file each one is the average of the precedent NWRITE steps where NWRITE NDT 100 b filename og1 The file contains information related with the sediment filtration model The first part of the file summarizes information read from the igr and isd input files along with some of the calculated parameters needed for the simulation The second part of the file contains sediment transport and deposition time series for the simulation period As before only a 100 time steps are printed to this file In this case the sediment filtration step is calculated with th
134. ifficult to judge in which direction the parameters should be modified and quantification of the uncertainty on the obtained parameters cannot be performed in a rigorous way Therefore the manual calibration method cannot ensure that the best parameter set is found A more elaborated complex and increasingly attractive form of parameter estimation is inverse modeling This procedure provides effective parameters in the range of the particular model applications and overcomes the drawbacks of manual calibration Ritter et al 2003 Basically the process searches for the best set of parameters in an iterative way by varying the parameters and comparing the numerical solution given by the model with the observations of a certain state variable Lambot et al 2002 Ritter et al 2003 2004 By coupling the computer model with an optimization algorithm the parameter search consists of finding the global minimum of an objective function defined by the error between measured and simulated values Different techniques have been developed in the past to numerically solve inverse problems Among others we may consider methods such as the Steepest Descendent Newton s Gauss Levenberg Marquardt Simplex and Global Optimization Techniques Hopmans and Simunek 1999 Each of these have their own advantages and drawbacks and the success of finding the global minimum depends generally on the presence of multiple local minima in the objective function In ad
135. il Water Characteristic Curve r Hydraulic Conductivity Curve yan Genuchten van Genuchten C Brooks and Corey When selecting the Show parameters button the soil characteristic curves parameters are selected Part III VFSMOD W WindowsTM User s Manual 102 X veSmod Windows Editor v 5 1 9 i _ s Tec f _ File Source Area UH Filter_Strip_ VFS Design Sensitivity Uncertainty Calibration Options Window Help B gt vism Project sampleliso prj Working Directory C vismod w Filter Strip Project File T Include Water Quality arland Elou Inn nouts sampleisomod iso 6 Soil Characteristic Curves Col fos M Shallow Water Table Present Soil Chatacteristic Curves Parameters Vi z h THETA Type m s or 0014004 CM OR VGALPHA VGN VGM p 17 13 46 1 52 0 348 M KUN Type Save Continue Editing OR VGM Save and Close 0 17 0 348 Close Detailed Hydrographs Soil Characteristc Curves r Soil Water Characteristic Curve Hydraulic Conductivity Curve Water and Sediment yan Genuchten van Genuchten Brooks and Corey parameters C Gardner J T Save Continue Editing Save and Close Close Help ated Water Content 0S 39 Overall Summary C Brooks and Corey After the new parameters are provided the rest of VFSMOD W remains unchanged and simulations can be performed as before Modifications to original VFSMOD W input files to accommodate sha
136. ile is missing or invalid I found the file in the directory so I don t know Part II VFSMOD and UH User s Manual 81 what the problem is I ll try re downloading the program and see if that works I am trying to operate in Vista is there anything special I need to do Yes you need to be logged in as administrator or a user with poweruser role to be able to register the libraries during installation Please contact your system administrator if you cannot login with those roles l We are running VFSMOD W using runoff data with sediment concentration and are getting this warning message that the Froude number gt 2 We have tried changing a lot of things but can not seem to get this to go away Is this a serious warning or not and what is likely to be causing this problem The Froude number F represents the ratio of inertial to gravitational forces that act during the overland flow wave formation When this number is lt 1 5 or 2 following other texts the kinematic waves dominate against the dynamic waves and thus the kinematic wave approximation to the full Saint Venant equation is appropriate Greater values just mean that the conditions of the problem start deviating from these assumptions In this case more error between the mathematical representation and the physical reality should be expected Notice that based on equation 4 this might be caused by relatively high flow velocity with very shallow flow Is your Manning s n too
137. ilter Flow is described by the continuity equation and steady state infiltration i e flow decreases linearly from upstream to downstream in the filter Wilson et al 1981 modified and incorporated GRASSF into SEDIMOT II a hydrology and sedimentology watershed model A simple algorithm to calculate the outflow hydrograph was incorporated into the model and up to three different slope changes throughout the filter could be considered The model does not handle time Part I VFSMOD W Model Documentation 1 dependent infiltration an accurate description of flow through the filter and changes in flow derived from sediment deposition during the storm event This work presents a design oriented computer modeling system VFSMOD W The MS Windows3 2 graphical user interface GUI integrates the numerical model VFSMOD a utility to generate inputs for the model based on readily available NRCS site characteristics UH and uncertainty analysis of sensitivity and design menu driven components VFSMOD the core of the design system is a model to study hydrology and sediment transport through vegetative filter strips The model combines the strength of a a numerical submodel to describe overland flow and infiltration b the University of Kentucky s algorithm developed specifically for the filtration of suspended solids by grass This model formulation effectively handles complex sets of inputs similar to those found in natural events The im
138. imulation 0 indicates that simulation had no errors CN the curve number input UHk soil erodibility K input UHc crop factor input UHp practice factor input isoks the Green Ampt saturated K input aRoa isothetai Green Ampt initial soil water content and igrss stem spacing input Next the summary outputs for the storm are given These include FldROmm and FldROm3 the runoff from the source area field in mm and m3 VFSROmm and VFSROm3 the runoff from the vegetative filter strip area in mm and m3 VFSinfm3 the amount of infiltration in the vegetative filter strip in m3 FIdSEDkg and FldSEDconc the sediment from the source area field in kg and kg L VFSSEDkg and VFSSEDconc the sediment lost from the vegetative filter strip in kg and kg L and SDR the sediment delivery ratio Mass Sediment from VFS Mass Sediment from Field and RDR the runoff delivery ratio Runoff from VFS Runoff from Field Fie Edt Block Convert Optors View Heb oe SOD VAR LPL SB 2 27635603 ii Feat BaP sd ite 480 Copwidt 1936 E Sai Goyvaerts Fe wwe BE SEO In addition the Analysis option for the Uncertainty section includes some analysis options for the data These options include plots of the frequency distribution and cumulative probability distributions These can be done for each sampled input parameter and for any of the output parameters 83 Working with E vfsmod w combo2 testgu
139. in the form of sheet flow Hortonian and the 1 D path represents average effective 2 D conditions field effective values across the VFS Part I VFSMOD W Model Documentation 3 The VFSMOD model uses a variable time step chosen to limit mass balance errors induced by solving the overland water flow equation The time step for the simulation is selected by the kinematic wave model to satisfy convergence and computational criteria of the finite element method based on model inputs Mufioz Carpena et al 1993a b The model inputs are specified on a storm basis State variables are integrated after each event to yield storm outputs 2 1 Hydrology This program solves the kinetic wave approximation of the Saint Vennant s 1881 equations for overland flow KW for the 1 D case as presented by Lighthill and Whitham 1955 such as Oh Og _ a By belt Continuity equation e S Sp Momentum equation oO Then a uniform flow equation equation can be used as a link between the q and A such as Manning s amp 3 f Gy e 2 Where A is depth of overland flow L q is the flow per unit width of the plane L T So is the slope of the plane Sy is the hydraulic or friction slope and n is Manning s roughness coefficient LT 3 The initial and boundary conditions can be summarized as 0 0 lt x lt L t 0 3 h x 0 t gt 0 I x h h where h can be 0 a constant or a time dependent function such as the incomin
140. in the second line of the isd file since only DP will be perturbed during the calibration and some of the prescribed NPART have specific values associated see the User s Manual 5 7MB As an alternative the user can select to optimize DP and SG currently Also consider that when optimizing particle size the range has to fall within the value of COARSE required for fine DP 0 0037 cm COARSE 0 5 or coarse particles DP gt 0 0037 COARSE 0 5 j When optimizing particle size or specific density for predefined particle classes NPART lt 7 in isd file with the inverse calibration component the results don t seem to be right By default the automatic calibrator overrides the NPART setting and forces it to NPART 7 so changes in DP are considered by the model However the user must be careful to consider the specific density value given for SG in the second line of the isd file since only DP will be perturbed during the calibration and some of the prescribed NPART have specific values associated see manual As an alternative the user can select to optimize DP and SG currently Also consider that when optimizing particle size the range has to fall within the value of COARSE required for fine DP 0 0037 cm COARSE 0 5 or coarse particles DP gt 0 0037 COARSE 0 5 k After VFSMOD W installation when I try to open the program I receive an error message saying Component comdlg32 o0cx or one if its dependencies not correctly registered a f
141. indows 9x NT 2000 KP A E A R T E E EE Helen A 53 1 2 2 Installing together with the Windows Graphical Interface Windows 9x NT 2000 DED EE eset E N N cot engencvdgheveneds A A 54 1 2 3 Installing on a UNIX system s sessesseseseeeesreesssesssesresessenesstseeseesestesesesseeeseesese 54 1 3 Using the project file for input and output ssssseesseeeesessesesseseesteresersesersesessesessesesseesese 55 1 4 VESMOD input files oss ny nena a E a E T RE N 57 1 4 1 filename ikw parameters for the overland flow solution seeeeeeeeeeeeeee 57 1 4 2 filename irn storm hyetograph ccccecceescescescecseceseeeceseeeeeeecesceeeeeeeeseeeeeenees 58 1 4 3 filename iro runoff from the adjacent field into the VFS cc eeeeeseeeeetees 59 1 4 4 filename iso soil properties for the infiltration model 0 0 ee eeeeeeeeeeeeeeeees 60 1 4 5 filename igr buffer properties for sediment filtration model 0 00 eee eee 62 1 4 6 filename isd sediment properties for sediment filtration model ee 62 1 4 7 filename iwq water quality transport model eecceceeseeeeceseeeeeneeeeeseeesees 63 1 5 Model file outputs reo a hornets eeepc theo caen ee th eee O N 64 1 6 Samplesapplications soinn ccs a eee ees eee eee einen 65 ViGd Inputs eoa enei eiei revlon bes E date E E ERE eee eses 65 1 62 QUEUES ic oe ve ea eE erena a EE EOT EE EAA EOE AERA E ETAT EEOSE AE EOE TERRAE TAES 66 2 UH for Input Preparation User s
142. ion and the analysis method selected Morris or extended FAST in this case With this information the pre processor produces a matrix of sample inputs to run the model step 2 Figure14 An interface program was written in C C sharp language and added to the VFSMOD W v 3 x and above GUI to run the model Part I VFSMOD W Model Documentation 36 for each new set of sample inputs The program automatically substitutes the new parameter set into the input files runs the model and performs the necessary post processing tasks to obtain the selected model outputs for the analysis that are stored in a matrix step 3 Figure 14 The output file created by VFSMOD W is compatible with Simlab so that the analysis can be completed using the Statistical Post Processor module of SimLab For this the input Simlab sample file and output matrices from VFSMOD W are called into the program to calculate the sensitivity indexes of the Morris and the Extended FAST methods step 4 Figure 14 The Data Analysis Toolpack of the Excel spreadsheet software Microsoft Corp Redmond Washington USA can also be used to construct the output probability distributions and to quantify the uncertainty based on the set of Extended FAST results step 6 Figure 14 inputi Std gt Output n Global Uncertainty Analysis GUA Input 2 Model structure control o O O n H 5 2 i a z 4 Q r Q lt re hed g
143. ional element s J Water Resour Planning and Mgmt Div ASCE 116 6 803 819 Wallach D D Makowski and J W Jones 2006 Working with Dynamic Crop Models Evaluation Analysis Parameterization and Application Elsevier Williams J R 1975 Sediment yield prediction with the Universal equation using runoff energy factor In Present and prospective technology for predicting sediment yields and sources ARS S 40 USDA Agricultural Research Service ppp 244 252 Wilson B N B J Barfield and I D Moore 1981 A Hydrology and Sedimentology Watershed Model Part I Modeling Techniques Technical Report Department of Agricultural Engineering University of Kentucky Lexington Wilson L G 1967 Sediment removal from flood water by grass filtration Transactions of ASAE 10 1 35 37 Wischmeirer W H C B Johnson and B V Cross 1971 A soil erodibility nomograph for farmland and construction sites Journal of Soil and Water Conservation 26 5 189 193 Wischmeirer W H and D D Smith 1978 Predicting rainfall erosion losses a guide to conservation planning Agriculture Handbook No 537 USDA Washington DC 58 pp Wolfram S 1999 The Mathematica Book 4th edition Cambridge Univ Press Woolhiser D A 1975 Simulation of unsteady overland flow In Unsteady Flow in Open Channels Vol I Ed K Mahmood and V Yevjevich 485 508 Fort Collins Water Resources Woolhiser D A R E Smith and D C Goodrich 1990
144. ions for storm types ID amp TD cececceccecsecsseeseceeceseeeceeeeeeeeeecaeeeecaeeseenseens 15 3 1 2 Equations for storm types I amp IA woe ececceesecseceseeseceeceseceeceseeneeeaeeaeecaeeeenseentens 15 3 2 Generation of Runoff Hydrographs 0 cccccesseesececceseceeceeeecesceseeeeceseeeeecaeesaenaeeesenseeneens 17 3 2 1 Computation of Total Runoff using NRCS Curver Number method SI units 17 3 2 2 Peak flow calculation using NRCS method SI units cc eceeceseeseeeeeeeeeeeees 18 3 2 3 Time correction for hydrograph to match hyetograph ccecceceseeeseeeeeeeeeeeees 22 3 3 Incoming sediment load calculation cceccecesseesecssecseceseeseceseesecesceeeeeeeseeeeeeaeeeseeeeeaees 24 3 3 1 Universal Soil Loss Equation USLE cccccceccessesecesceeeceeceseeeeecaeesecseeneenseens 24 3 3 2 Modifications to USLE to handle storm event cccesseeceeseeeceeseeeseeteeseeneeees 26 3 4 Computational Structure Of UH isicing iini E EE EE REE 28 3 5 Sensitivity Analysis of VFSMOD s ssssssessesesserssssrsrsressesesstsrsstereseesestesestessesessesesseeeseesee 29 3 6 Previous Testing and Applications ccccsccescesseesceseeesceseeesecaeesseseceseeseceseeeeeeeteneenseenees 30 4 Sensitivity and Uncertainty Analysis Procedures for UH and VFSMOD Built In VESMODENW axtintsetand enti nities luewaiee ee oe mel ane eae Ae 31 4 1 Local OAT Sensitivity Analysis sesessesessesesseeeesersrsresrsrestrstsetsesesses
145. ions include Design Sensitivity and Uncertainty These are discussed in more details in other sections K 4 VFSmod Windows Editor v 4 1 0 lt a wm nina File Source Area_ UH Filter_Strip_ VFS Design Sensitivity Uncertainty Calibration Options Window Help 63 vfsmod w Initialization Information ToN ee Accept and Save Close User Name Paus t lt Cs C iOSSC S User Affiation Mses i SCS User Address fiquser myemailaddess User e mail Address J Bien ae ae Eismo SSS Browse JT Check here to turn on Debug Option usually not checked only on for this run Associate Project files prj and lis with vfsmod w option not yet available To Register Email the vfsmod w cfg File to carpena ufl_ edu Thank You The Options menu is used to identify the user and select a default directory for project files Although we request registration to download the model this menu also stores information about the user and their installation For bug reports we may request that you e mail us this file to assist in debugging The file is located in the installation directory and is called vfsmod cfg 4 1 vfsmod w Options File On the main window the Options Menu allows the user to review the program s options and user information This information is entered the first time vfsmod w is executed and can be checked and changed using the Options Menu Part III VFSMOD W WindowsTM User s Ma
146. iply the given densities by 1 3 2 3 1 4 3 and 5 3 for poor fair good very good and excellent covers b Values vary depending on mixture If a given grass type predominates values for that species should be used c Values of Ss above 2 5 cm can cause scour and are not recommended Part IV VFSMOD Appendices 163 3 4 NRCS SCS Curve Numbers Runoff curve numbers for urban areas From USDA NRCS 210 VI TR 55 2nd Edition June 1986 Table 2 2a Cover Description Curve Numbers for hydrologic soil group Average Cover type and hydrologic condition ae t A B C D impervious area Fully developed urban areas vegetation established Open space lawns parks golf courses cemeteries etc P Poor condition grass cover lt 50 68 79 86 89 Fair condition grass cover 50 to 75 49 69 79 84 Good condition grass cover gt 75 39 61 74 80 Impervious areas Paved parking lots roofs driveways etc 98 98 98 98 excluding right of way Streets and roads Paved curbs and storm sewers exclud 98 98 98 98 ing right of way Paved open ditches including 83 89 92 93 right of way Gravel including right of way 76 85 89 91 Dirt including right of way 72 82 87 89 Western desert urban areas Natural desert landscaping pervious 63 77 85 88 areas only Artificial desert landscaping impervious 96 96 96 96 weed barrier desert
147. ived from a randomized sampling procedure they can be used as the basis for the uncertainty evaluation by constructing cumulative probability functions CDFs for each of the selected outputs This leads to a very efficient Monte Carlo type of uncertainty analysis since only the sensitive parameters are considered as the source of uncertainty In general global sensitivity and uncertainty analysis follows six main steps Figure14 1 PDFs are constructed for uncertain input factors 2 input sets are generated by sampling the multivariate input distribution according to the selected global method i e Morris method for the initial screening and extended FAST for the quantitative refining phase 3 model simulations are executed for each input set 4 global sensitivity analysis is performed according to the selected method 5 if the Morris screening method is selected it results in a subset of important parameters and steps 2 4 are repeated only for those important parameters using the extended FAST method 6 uncertainty is assessed based on the outputs from the extended FAST results by constructing PDF CDF and statistics of error calculated A batch processor is available within VFSMOD W to perform the global sensitivity and uncertianty procedure outlined in Fig 14 SimLab v2 2 Saltelli et al 2004 statistical pre processor module executes step 1 Figure 14 based on the PDF types and statistics provided described in the next sect
148. ively Figure 3 are needed for these calculations Details of the implementation of the submodel are given in Mufioz Carpena 1993 Under extreme sediment inflow events the filter can be filled up with sediment to the top of the standing vegetation VFSMOD accounts for this in a realistic way by allowing normal filtration up to the time step when the sediment wedge reaches the end of the filter Part I VFSMOD W Model Documentation 7 X 2 L and bypassing filtration from then on g g The original University of Kentucky sediment model uses a simple approach to calculate flow conditions at specific points of the filter and does not consider the complex effects of rainfall infiltration and flow delay caused by the buffer VFSMOD provides a more accurate description of the flow conditions from the hydrology submodel whereas changes in surface conditions topography roughness due to sediment deposition during the event are obtained from the sediment filtration submodel 2 3 Chemical tranport trapping 2 3 1 Pesticides For aquatic organisms such as plants fish aquatic phase amphibians and invertebrates the U S EPA Environmental Exposure and Effect Division EFED uses computer simulation models to calculate estimated pesticide environmental exposure concentrations EECs in surface water The EECs are compared to critical toxicological values to determine the level of potential risks to aquatic species A tiered system of m
149. ka et al 2009 Consider for example that the presence of sheet versus concentrated flow will significantly impact the resulting sediment and or contaminant removal efficiencies Numerical process based models have been available for some time for predicting runoff and sediment reduction by VFS such as the Vegetative Filter Strip Modeling System VFSMOD W Mufioz Carpena et al 1999 and Mufioz Carpena and Parsons 2004 The VFSMOD W is a finite element field scale storm based model developed to route the incoming surface flow hydrograph and sedigraph from an adjacent source area field road urban area etc through a VFS and to calculate the resulting outflow infiltration based on the extended Green Ampt equation for unsteady rainfall and sediment trapping based on GRASSF Mujfioz Carpena et al 1999 and Mufioz Carpena and Parsons Part I VFSMOD W Model Documentation 9 2004 Researchers have demonstrated the model s ability to predict reductions in runoff volume and sediment concentration moving through VFS Such numerical models can account for site specific conditions not able to be captured by the empirical models VFSMOD W has been used by state regulators and city engineers for the design and evaluation of VFS to control surface runoff pollution Recent research has developed and evaluated an empirical model for pesticide trapping by VFS with a foundation of hydrological sedimentological and chemical specific paramete
150. ks the Green Ampt saturated K input aRoa isothetai Green Ampt initial soil water content and igrss stem spacing input Next the summary outputs for the storm are given These include FldROmm and FldROm3 the runoff from the source area field in mm and m3 VFSROmm and VFSROm3 the runoff from the vegetative filter strip area in mm and m3 VFSinfm3 the amount of infiltration in the vegetative filter strip in m3 FldSEDkg and FldSEDconc the sediment from the source area field in kg and kg L VFSSEDkg and VFSSEDconc the sediment lost from the vegetative filter strip in kg and kg L and SDR the sediment delivery ratio Mass Sediment from VFS Mass Sediment from Field and RDR the runoff delivery ratio Runoff from VFS Runoff from Field A separate file is written for each parameter The user selects the Analysis option from the Sensitivity menu and selects the file to analyze a B Working with CAvfsmod woutput igrSSsens sen o a Plot Selected Output Sensitivity Analysis far Stem Spacing SS Plot Selected Output Select Output Parameter Absolute Sensitivity i Source Sediment kg Plot Selected Output Relative Base Sensitivity Plot Selected Output Oi Source Sediment Concentration g L Source Runoff m3 Filter Runotf mm Filter Sediment kg Filter Runoff m3 Filter Infiltration m3 Filter Sediment Concentration g L Relative Sensitivity e 0 o io Se
151. l International Meeting Sacramento California ASAE Paper of ASAE no 01 ASAE St Joseph Parsons J E and R Mufioz Carpena 2002 VFSMOD W a graphical Windows system for the evaluation and design of vegetative filter strips for sediment trapping In Watershed Management to Meet Emerging TMDL Environmental Regulations Proc 11 13 March Fort Worth Texas USA eds A Saleh B Wilson pp 532 535 St Joseph Michigan ASAE Poletika N N Coody P N Fox GA Sabbagh G J Dolder S C White J Chlorpyri fos and atrazine removal from runoff by vegetated filter strips Experiments and pre dictive modeling J Environ Qual 2009 38 3 1042 1052 Rawls W J and D L Brakensiek 1983 A procedure to predict Green Apmt infiltration parameters Adv in Infiltration pp 102 112 ASAE Pub no 11 83 Reichenberger S Bach M Skitschak A Frede H G Mitigation strategies to reduce pesticide inputs into ground and surface water and their effectiveness a review Sci Total Environ 2007 384 1 35 Ritter A F Hupet R Mufoz Carpena M Vanclooster and S Lambot 2003 Using inverse methods for estimating soil hydraulic properties from field data as an alterna tive to direct methods Agric Water Manage 59 2 77 96 Ritter A R Mufioz Carpena C M Regalado M Vanclooster and S Lambot 2004 Analysis of alternative measurement strategies for the inverse optimization of the hydraulic properties of a volc
152. l 85 event One is rainfall excess derived from the infiltration capacity of the soil but the other is a flood wave from the field moving into the filter The model checks to see if such a wave is in the filter and the automatically switches infiltration to ponding regardless of the infiltration capacity for that specific time The question is where to check for i e at the beginning of the filter or at the end since you will stop your regular Green Ampt infiltration calculation at that time You can check for your particular application with 0 0 5 and 1 values to see if you get any changes z In the incoming sediment characteristics file there is the parameter Porosity of deposited sediments How do you measure this parameter in the field Or can it be estimated from other parameters Yes it could be measured at the field in an undisturbed column separating the upper layer of sedimentation from the actual soil measuring the volume and getting the dry weight You could possibly calculate it from the mean particle size of the sediment assuming some form of packing scheme spheres etc but we have never done so We normally use use a value of 0 437 since does not seem to be a very sensitive parameter Let us know if you find otherwise aa For the roughness in the file buffer vegetation characteristics a difference is made between the Manning s coefficient of the bare surface and that of the vegetation the grass Does this mean then th
153. le names Part HI VFSMOD W WindowsTM User s Manual 96 Inputs Outputs buffer properties for the sediment filtration jogl detailed time series describing the sediment igr submodel transport and deposition within the buffer og2 detailed information on the singular points ikw parameters for the overland flow solution defined in the theory section of the manual detailed outputs on the inflow and outflow irn storm hyetograph ohy hydrographs detailed summaries of the water and iro storm hydrograph from the source area osm sediment balance final geometry of the filter overall summary of filter performance with sediment properties for the sediment osp comparisons between the source area and isd filtration submodel filter iso soil properties for the infiltration submodel If the Include Water Quality Option is selected then two more files will be shown Inputs Outputs Pollutant properties for the Water Quality Transport submodel only needed if IWQ 1 owq details describing the water quality transport iwq in ikw and removal efficiency This version of VFSMOD incorporates the algorithms derived by Sabbagh et al 2009 to predict pesticide trapping by vegetative filter strips by selecting the Option 1 when the Include Water Quality option is selected The program has been modified to achieve 3 main objectives a incorporate the new water quality WQ functionality b ensure backward comp
154. llow water table option The input file iso has been modified to ensure backwards compatibility For this if a number appears in a new line below the standard file the program considers that the shallow water table option has been selected and reads the value in that line as the water table depth WTD m 0 lt WTD lt 10 0 and continues reading on the next lines for additional parameters The following structure will then be followed VKS SAV OS Ol SM SCHK WTD ITHETATYPE PAR I IKUNSTYPE PARK J Vhe Note that SAV and OI are now calculated internally but two dummy numbers must be included to ensure backwards compatibility of the file even although these numbers are Part III VFSMOD W WindowsTM User s Manual 103 ignored PAR I and PARK J are parameters of the soil water retention and unsaturated hydraulic conductivity curves respectively ITHETATYPE is an integer to select the soil water characteristic curve type with values 1 van Genuchten 2 Brooks and Corey The parameters for these curves will be Equation ITHETATYPE PAR1 PAR2 PAR3 PAR4 ITHETA van Genuchten 1 OR VGALPHA VGN VGM 1O0OR2 Brooks and Corey 2 OR BCALPHA BCLAMBDA 10R2 If the effective saturation S is defined as 0 0 S e _ 0 90 where is the moisture content 0 is saturated moisture content the van characteristic curve is 0 90 0S 0 S 1 laygh O h 0 S wher
155. lly we end up by plugging in the BC for the problem subroutine BCA 1 8 ELEM EK PGPAR Form the element arrays EK a first initialize the element arrays b begin integration point loop for the Gauss quadrature rule b 1 obtain shape function values b 2 get the value for each element of the array 1 9 SHAPE XIS PSI DPSI WF PGPAR Calculate the values of the weighting and basis functions PSI and their derivatives DPSI with respect to the master element coordinates at a specified value of XIS A typical ele ment x X consisting of k 1 nodes x X 4 is always normalized into the master element 1 1 by the transformation over a typical element x x and there exist k 1 element shape functions PSI each is a polynomial of degree k The type of shape func tion used linear quadratic modified quadratic and cubic is selected according to NPOL 1 10 ASSM A EK NBAND NEL This subroutine adds the EK s to the global matrix A Part IV VFSMOD Appendices 152 1 11 BCA A NBAND Plug in first kind of BC Dirichlet in the system matrix A 1 12 FACTOR A N NBAND Perform the lower and upper decomposition LUD over the system matrix A and store the lower and upper triangular matrices on the old A matrix 1 13 GASUB TIME DT L R RAIN NEND TRAD This subroutine solves the infiltration problem for unsteady rainfall case using the Green Ampt infiltration model After ponding at the surface is detected the infil
156. loutput UncRestilts 5y Lomaxt E 4 10 x Select Input Parameter Select Output Parameter eea T it Change UncResuts Fie e LH Curve Number C Source Puno ren Source Sediment ka 5 F UH Soll Erodibaly C Source Punat m3 Ee eeel NEE SU Ht F Filter Sediment kq POH ar Filter Runo mm Plot Frequency Distributors c Patan Darna TUE Pred Fitter Runot m3 Filter Sediment Concentration g L e C VFSM Ksat Groen Ampi F Fitar ini rotion m3 Sediment Delivery Rato Fiter Source C Runot Delivery Ra o Fiter Source arent Results File C VFSM Particle Class Diameter E vismod w combo testgui output UncResutts 5y 10m ta F VFSM Thetal Green Ampt Results Based on 1502 Simulations Examples of the frequency and cumulative probability distributions for sampling the curve number are shown below Part III VFSMOD W WindowsTM User s Manual 141 Examples of the graphs for the outputs are given for the sediment delivery ratio SDR 2 Part III VFSMOD W WindowsTM User s Manual 142 12 Design Menu The design section of VFSMOD can be used to examine a range of storms and filter strip design parameters to assist in finding the optimum length for a given situation A base UH and VFSMOD project is selected The user can specify a range of storms for generating varying input runoff hydrographs and sediment loads A range of filter strip lengths along with varying grass media spacing can be spe
157. m g s including Manning s n The follow ing steps are followed a flow depth and velocity set to zero for no flow at any given point b otherwise calculate R V and dy for the given point c the resulting equation is solved by the Newton Raphson iterative method 1 22 EINSTEIN GS2 NTRCAP COARSE This program solves Einstein s bed load transport equation to find the sediment transport capacity g at the end of C t by following the method proposed by Barfield et al 1979 where known values are d particle size diameter cm S filter main slope R hydraulic radius of the filter at Bit cm g g water and sediment weight density g cm3 g acceleration due to gravity 980 cm s7 COARSE of particles from incoming sediment with diameter gt 0 0037 cm i e coarse fraction that will be routed through Part IV VFSMOD Appendices 155 wedge and the unknown is g gt g 7 sediment transport capacity or sediment load enter ing downstream section g s cm Notice that all units in CGS system cm g s The fol lowing steps are implemented a check if the transport capacity is lower than concentration a 1 if lower deposition at the wedge occurs first part of subroutine STEP3 a 2 if higher there is enough energy to transport sediment through the wedge and no deposition occurs all sediment is transported to the suspended sediment zone zones C t and D t 2nd part of subroutine STEP3 1 23 STEP3 GS2 TIME
158. n preparation next release 2 3 3 Multi reactive transport In preparation next release 2 4 Linkage between submodels Flow conditions at the entry exit and three inner points 1 2 and 3 of the filter are needed for the sediment transport calculations qim 9 72 73 and qout in Figure 3 The GRASSF and SEDIMOT II models use a simple approach to calculating those values and do not consider the complex effects of rainfall infiltration and flow delay caused by the filter A more accurate description of the flow conditions are obtained from the hydrology submodel presented above In turn the sediment transport model supplies information on changes in surface conditions topography roughness due to sediment deposition during the event that affect overland flow This interaction between submodels is depicted in the Part I VFSMOD W Model Documentation 11 flowchart in Figure 4 Rainfall lo rainfall infiltration a gt Field dt Sediment Transport Field water He Model sediment inflow zk lt inflow So n water outflow Figure 4 Flowchart showing linking between hydrology and sediment submodels During the simulation feedback between the hydrology and sediment models is produced The hydrology model supplies the flow conditions at the five locations entry 1 2 3 and exit set in the last time step Figure 3 The other parameters that interact through the linkage are the length slope and
159. nced to this standard plot For example a C 0 5 indicates that one would expect about one half the erosion with this cover management than from the standard plot Since the USLE is applicable to areas dominated by overland flow with little or no concentrated flow pathways it lumps rill and interrill erosion The R factor combines rainfall and runoff erosivity In the annual version of the equation the units are usually expressed as EI units per unit time The original units used by Wischmeier are 100 ft tons acre in h which are often referred to as the Wishmeier English EI units R ranges from 50 550 for eastern US In North Carolina R ranges 330 in the Southeastern portion of the state to 175 in the Appalachians In the Piedmont area the annual R is approximately 250 Foster 1982 indicates that no single metric unit has been accepted although for modeling convenience he suggests Newtons h So to convert the Wischmeier English units to N h multiply R by 1 702 The soil erodibility factor K is generally selected based on the top soil The english units for K are tons acre EI with typical values ranging from 0 05 0 60 The SI metric units for K are usually expressed as kg N h m The factor to convert english units to SI metric is to multiply by 0 1317 So for soil losses A RK the two quantities in the USLE with dimensions expressed as kg m then the SI units for R is N h K can be approximated based on data from Wischmeier et
160. ncentrated flow occurs within the filter However concentrated flow can be effectivelly simuated in VFSMOD by setting diferent dimensions between the field or source area edge SWIDTH and the filter entry side FWIDTH See Mufioz Carpena and Parsons 2004 and Fox et al 2010 Notice that there are also critical filter lengths beyond which the sheet kinematic flow assumption is violated In addition to the kinematic wave number criteria eq 4 McCuen and Spiess 1995 presented a maximum length criteria as nL S 2 lt 100 A nomograph of this relationship can be seen in Figure 17 200 Maximum filter length criteria m Manning s n Figure 17 Nomograph of critical filter length McCuen and Spiess 1995 Since parameters to describe hydrology and sediment transport in VFS are highly variable field variability is an inherent source of error A range of variation in the saturated conductivity parameters is usually needed to fit the model to observed data Although this variation can be explained by changes in surface conditions due to seasonal and biological factors these changes are difficult to quantify in field situations 8 2 Changes in Model Releases Current release v6 x Option to simulate flow and infiltration affected by shallow water table conditions Infiltration option to consider heterogeneous soil profile i e soil horizons Part I VFSMOD W Model Documentation 44 Next releases Opti
161. ng Flow sediment Constrai concentration Cl g em3 A Porosity of deposited sediment POR 100 Sediment particle class diameter Save Close Close d50 DP cm Sediment particle density SG g cm3 Advanced Calibration Run Results Cancel Help Cyn FIGURE 11 Example of use of the calibration option In order to run the calibration option let us select the hydraulic conductivity parameter with min 0 0000001 m s and max 0 0005 m s When the Run button is clicked two windows are displayed The first one shows the current iteration that the program is running and the other window if selected shows the graph for that iteration Part III VFSMOD W WindowsTM User s Manual 128 EJ FSmod Windows Editor v 3 0 22 x JH Filter S Inverse Calibration Hydrograph Project Name oe O Browse Paota TT verse mens fyel bt Browse Edit Select parameter s to be changed or calibrated No Change Calibrate New value Min Max C C ettical Saturated K VKS m s 0 0000001 0 0005 WINDOWS system32 cmd exe INISHED UFSMOD v2 4 3 ations of 180 31 FINISHED SMOD v2 4 3 100 UFSMOD v2 4 3 of 108 108 FINISHED UFSMOD v2 41 it of 188 FIGURE 12 Execution of the calibration Once the calibration finishes the program requests to click any key to continue At this point the output file has been written and it can be viewed by clicking the C
162. ng the model we compared several of these for different field experiments and model runs and found the simplification to be acceptable This simplification had the benefit that the user did not have to come up with samples through the event but just the average Could you check if this assumption holds for your particular case If enough users think that using a changing sediment concentration with time is critical for their application let us know and we will add this feature to the program Part II VFSMOD and UH User s Manual 83 t You mention sediment transport capacity gsd at the end of the filter how do you calculate sediment transport capacity In fact gsd is also called sediment load in the manual This is calculated depending of the region of the filter The basic idea is to compare the calculated value of gsd with the incoming sediment concentration If gsd gt sediment concentration the difference is allowed to deposit in whatever section of the filter is being calculated coarse sediment at wedge or fine sediment at the lower section of the filter u The equations have a dp or median particle size but it seems that some particle size distribution must be implicit in the equations I am a bit confused how dp or d50 can simulate the deposition realistically if it is a single particle size Clearly different sized particles settle at different rates and a single particle size specification must be implying some distribution
163. ng the rainfall period In this situation the product of E and the maximum 30 min rainfall intensity 13 divided by 100 is used as the erosivity factor R in the USLE for the particular storm Multiplication of this by 1 702 yields consistent SI metric units of N h In GLEAMS and the daily rainfall version of CREAMS the EI3 for a 24 hour rainfall Vp in inches is computed as Els 1 51 sag ee 34 The units for the daily EI3 are ft ton acre in For daily rainfall amounts another approximation for EI for a storm is EI 8 0V 35 where Vp volume of rainfall in inches The default units for EI are ft tons in h and if we multiply by 1 702 then we obtain N h Foster et al 1977 suggested an improved erosivity factor for a single storm over that of substituting storm EI for R This approach combines the effect due to runoff and rainfall into the erosivity factor So for a single storm Foster et al 1977 defines R in N h Wile Ry 0 5Rs 0 35V 05 36 Rg E Izo where E storm s total energy and I37 maximum 30 min rainfall intensity in N h V volume of runoff mm peak rate of runoff mm h Williams 1975 suggested a further modification to R to handle areas larger than field scale This modification makes an attempt to account for deposition within the area which would reduce the sediment losses from the area Foster et al 1982 reported the modification of R as R 9 05070 37 Part I VFSMOD
164. nit hydrograph by definition the time to peak in the unit hydrograph is defined as see Figure 10 tp t De 2 0 6 t De 2 25 Part I VFSMOD W Model Documentation 23 O gt Time t eo eon excess gt lt De duration of rainfall excess ke Bx lt gt Runoff volume p Peak runoff lt P gt Time t t b lt gt Figure 10 Hydrograph quantities used in calculation of time shifting We can now calculate the duration of the rainfall excess De as De D ti 26 In this option 2 the corrected time to peak of the hydrograph can be obtained from the ordinate of the unit hydrograph as tp 0 127481 O A qp 27 A time shifting is needed in the hydrograph to match the rainfall as toff tp tp 28 and all the hydrograph times will be corrected as t t toff 29 An example showing the calculations for both options 1 and 2 is shown below Versions of UH after 0 7 implement Option 1 since it produces runoff after the peak of the hyetograph as in observed natural events see Figure 11 Part I VFSMOD W Model Documentation 24 0 01 Option 1 gt toff ti 0 008 UH based Option 2 gt toff tp tp Rainfall 5x10 0 006 D oO gt 0 004 1 10 0 002 0 1 52105 0 5000 1 104 1 5 104 2104 Time s Figure 11 Time shifting of hydrographs to match the storm 3 3 Incoming sediment load calculation 3 3 1 Universal Soil Loss
165. nnual International Meeting Chi cago Illinois Paper 02 2133 ASAE St Joseph MI Knisel Walter G F M Davis R A Leonard 1992 GLEAMS Version 210 Users Man ual Pre Publication Copy US Department of Agriculture Agricultural Research Ser vice Available from University of Georgia Coastal Plain Experiment Station Bio and Ag Engineering Tifton GA UGA CPES BAED Publication No 5 259 pp Koda M G J McRae J H Seinfeld 1979 Automatic Sensitivity Analysis of Kinetic Mechanisms Int J Chem Kin 11 427 444 Lighthill M J and C B Whitham 1955 On kinematic waves flood movement in long rivers Proc R Soc London Ser A 22 281 316 Lambot S M Javaux F Hupet and M Vanclooster 2002 A global multilevel coordi nate search procedure for estimating the unsaturated soil hydraulic properties Water Resour Res 38 11 1224 Legates D R and G J McCabe 1999 Evaluating the use of goodness of fit measures in hydrologic and hydroclimatic model validation Water Resour Res 35 233 241 Lin J A progress report for aquatic exposure assessment in the U S EPA Office of Pesti cide Programs U S EPA Office of Pesticide Programs Washington D C 2009 http www epa gov oppefed1 ecorisk presentations setac_eu htm Some Lin J Young D Kennedy I The Tier II Modeling approach for aquatic exposure assessment in the U S EPA Office of Pesticide Programs U S EPA Office of Pesti cide Programs Wa
166. nt the field and to interpret the results We will make every effort to provide assistance and encouragement as our other commitments allow We do ask that you reference our work if you find it helpful in your pursuits Ag amp Bio Eng IFAS U of Florida by R Mufioz Carpena carpena ufl edu Table of Contents Teds Sted Aiea E E A A a E ead ak aie atta E A oa ale ald di es alaepnel coals al l Part I VFSMOD W Model Documentation c c cccsscscccesecesensessecsscsesacsesssessaceseeesacens 1 l TOU COI asa foto ene a a E E E T e E E EA l 2 VFSMOD Model Components Processes and Solution Techniques c00 3 2 1 Hydrol py re e aed E TE E E A EE EA E Red hae 4 22 Sediment Transport srr serasi n a a a A a a E 6 2 3 Chemical tranport trapping i scie eei EE EEEa EE EE EEE Eein ie Ei 8 231 Pesticidesmsnenane aaa a a a Aan eat a a A 8 2 3 2 Solute Transport esenee iE R EEEE KEE E EE E R E 11 2 3 3 Moulti reactive transport s esesessesesseseeseeeesstsseesresessestssteresttsesersrsessesesseseseesee 11 2 4 Linkage betwecnmsubmode lS iinne n e g EE EEE OEGE E 11 2 5 Soltition procedure sa ceceok Sesh RG eG eS BATA Res A O E 12 2 6 Modeliinputts cecestc ee et SAE rie IS St AA NIN a Os NNN oe dt ol 14 3 UH utility preparation of model inputs for design purposes ccceeeeeseeetees 15 3 1 Generation of Synthetic Rainfall Hyetographs cccesesccescesecesceeceeeeeseeseecseceenaeeneeaees 15 3 1 1 Equat
167. nual 91 VFSmod Windows Editor v 4 1 0 a Lol File Source Area UH Filter_Strip_ VFS Design Sensitivity Uncertainty Calibration Options Window Help B vfsmod w Initialization Information fr ea net Lin vi ty Cl Accept and Save Close User Name Pauses ss S User Affliation Mise U User Address iquser myemail address User e mail Address Directory For Saving C vismodw tufana Browse Project Files J7 Check here to turn on Debug Option usually not checked only on for this run I Associate Project files prj and Jis with vfsmod w option not yet available To Register Email the vfsmod w cfg File to carpena ufl edu Thank You bme C Documents and Settings ABE User My Documents desktop backup oct 2007 VFSMOD CODE HELP with HELLLP help5 vfsmod w vfs options bmp Fill in the registration information Be sure to select the directory where you installed vfsmod w exe for the Directory for Saving Project Files On the Options screen we have included an option to associate files with extensions prj and lis with vfsmod w exe This is currently not implemented Once available this option will allow the user to click on a file with prj or lis extensions from the Explorer window and automatically load the project You can manually accomplish this by associating these extensions with vfsmod w exe using the file Properties menu from Explorer Currently we do not have an automatic registration for
168. ode 2 Part III VFSMOD W WindowsTM User s Manual 124 J FSmod Windows Editor v 4 0 7 File Source_Area_ UH Filter_Strip_ VF5 Design Sensitivity Uncertainty Calibration Options Window Help s Inverse Calibration Hydrograph Project Name sample prj Browse Hydrograph Measure Data File Soleat parameters Look in i inverse gt amp EE inputs B output My Recent 3 start_inv_mer Documents meas_gso Cj Desktop a a a a My Documents a My Computer a 9 File name meas_hyd x My Network Files of type File Type txt X PETI I Open as read only U D O S a ee ee en 2 e 2 2 fe 2 a FIGURE 7 Selecting a field data file within the inverse forlder to run the hydrograph calibration Two additional buttons are provided associated to these files in each of these interfaces Browse Edit and Advanced settings The Browse buttons are used to change the selected files The Edit option is used to modify the field data see section 11 2 a 4 VFSmod Windows Editor v 4 0 7 DER File Source_Area_ UH Filter_Strip_ VF5 Design Sensitivity Uncertainty Calibration Options Window Help 1 Inverse Calibration Hydrograph Project Name sample pri Browse Hydkogranh ovr inverse meas_hyd txt im Edit Measured Data Measured Data File inverse meas_hyd txt Time s Flow m3 s weigth gt Editing a 3
169. odeling is considered with the Tier I GENEEC model representing a highly conservative screening tool USEPA 2001 For compounds with uses resulting in unacceptable TIER I EECs EFED implements a Tier II modeling system that reflects labeled uses for the compounds Lin et al 2007 The Tier II assessment procedure is based on simulation modeling with PRZM EXAMS using the linking program PES additional details of the process provided in the supporting information see Supplemental Material S1 PRZM simulates pesticide fate and transport from an agricultural field to an adjacent water body Carsel et al 1985 and Fox et al 2006 while EXAMS models pesticide fate in the water body Burns 1990 and Jackson et al 2005 The U S EPA has created various benchmark scenarios by crop Lin 2009 These scenarios are static in terms of the field and pond geometry but include variations in soil weather and management practices PRZM EXAMS simulations are typically conducted for a 30 year period 1961 1990 using daily weather data and assuming the maximum use rates and patterns as specified on the pesticide label Risks are determined based on the upper 90th percentile annual peak 4 d 21 d 60 d or 90 d mean concentrations depending on the target critical toxicological endpoint For acute risk assessments peak and 4 d EECs are used while the chronic risk assessments are based on longer mean averages For pesticides with uses that do not pass the Tie
170. oduces deposition of the coarse material particle diameter d gt 0 0037 cm carried mostly as bed Part I VFSMOD W Model Documentation 6 load transport The sediment trapped in this first part of the filter forms a geometrical shape that varies depending on the thickness of the deposited sediment layer at the entry of the filter Y t m and the effective top of vegetation H cm A triangular shape at the adjacent field area and the beginning of the filter is formed when Y t lt H After Y t H a trapezoidal wedge is formed Figure 3 with three well defined zones the upslope face of the wedge with zero slope O t cm the upper face of the wedge parallel to the soil surface A t and the downslope face B t with an equilibrium deposition slope S for each time step Figure 3 Together these first filter zones are termed wedge zone and its length changes with time as sediment is deposited Zone O t external to the filter is important in explaining field observations where a portion of the sediment is deposited in the field area adjacent to the filter After the wedge has formed no sediment is deposited in zone A t and the initial load g moves through to the next zone B t In this zone deposition occurs uniformly with distance to the deposition edge with transport mostly as bed load The model assumes that the sediment inflow load g is greater than the downstream sediment transport capacity g at point 2
171. of sediment deposited for that Dt and CDEP as a multiplier to reduce the actual sediment outflow g Wilson et al 1981 g write outputs h update values for next time step 1 24 POINTS N XPOINTS NODEX VBT This program finds a in uniform flow equation X mid point of downface of sediment wedge cm X bottom point of downface of sediment wedge cm X mid point of effective filter length L t cm and their associated NODEX i nodes for the X points to feed back to the overland flow submodel The procedure is as follows a find points for each of the areas in filter Part IV VFSMOD Appendices 156 b if required reshape the surface topography and roughness of the filter Notice that for entry a 0 slope value is not possible thus a minimum SCENTRY 0 005 is chosen The new values assigned are Section A slope Sc n VN bare Section B t slope Set n VN gt length VBT Sections C t amp D t slope unchanged n unchanged length VLT 1 25 KWWRITE N L M QTEMP X BCRO FWIDTH Write hydrology outputs to the ohy file as a hydrograph i e flow rate at the downstream end of the plane 1 26 OUTMASS VL FWIDTH TRAI LISFIL This subroutine processes the output hydrograph and calculates the components of the water and sediment balance The results are written to the summary file osm and osp Part IV VFSMOD Appendices 157 2 APPENDIX 2 Model parameters and variables 2 1 Overland flow A L
172. of the filter s hydraulic roughness and subsequent augmentation of infiltration Decreasing flow volume and velocity translates into sediment deposition in the filter due to a decrease in transport capacity Wilson 1967 Barfield et al 1979 and Dillaha et al 1986 reported that grass filter strips have high sediment trapping efficiencies as long as the flow is shallow and uniform and the filter is not submerged Researchers Dillaha et al 1989 Parsons et al 1991 have found that the filter length L controls sediment trapping up to an effective maximum length value thereafter additional length does not improve filter performance This maximum effective length depends on the source area topography and the hydraulic characteristics of the strip Several modeling efforts have been undertaken to simulate VFS efficiency in removing pollutants from surface waters Researchers at the University of Kentucky Barfield et al 1978 1979 Hayes 1979 Hayes et al 1982 1984 Tollner et al 1976 1977 developed and tested a model GRASSF for filtration of suspended solids by artificial grass media The model is based on the hydraulics of flow and transport and deposition profiles of sediment in laboratory conditions This physically based model takes into account a number of important field parameters that affect sediment transport and deposition through the filter sediment type and concentration vegetation type slope and length of the f
173. olson solution 0 5 recommended Courant number for the calculation of time step from 0 5 0 8 recom mended See Sample application on page 66 integer maximum number of iterations alowed in the Picard loop integer number of nodal points over each element polynomial degree 1 Recommended value 3 See Sample application on page 66 integer flag to output elemental information 1 or not 0 integer flag to choose the Petrov Galerkin solution 1 or regular finite element 0 Recommended value 1 See Sample application on page 66 integer number of segments with different surface properties slope or roughness real X distance from the beginning on the filter in which the segment of uniform surface properties ends m Manning s roughness for each segment s m slope at each segment unit fraction i e no units Part III VFSMOD W WindowsTM User s Manual 99 Fy Slope and Roughness Factors Fai om Ex segment Properties Add segments Change the slope and roughness of each segment Enter Distance m Roughness Slope on each line ibe 4 062776 gt 1 2364 1 5546 24729 3 0411 3 7093 43275 4 9457 5 5639 6 1821 032639 071528 075 031944 019444 029886 Done 2094 Ea m1 041667 134028 Cancel I fe ee ee l l Selecting the View Segments button displays a graph of the elevation change across the filter strip The elevation change is relativ
174. on to simulate numerical transport of solutes and multireactive transport beta Part I VFSMOD W Model Documentation 45 9 Distribution and Training The modelling system is provided free of charge to qualified users as an educational and research tool The model and documentation can be downloaded from the internet at http abe ufl edu carpena vfsmod or obtained from the authors Limited support is available from the authors Through the web site the user can send feedback and questions to the authors No formal training is available but can be arranged with the authors Part I VFSMOD W Model Documentation 46 10 Acknowledgements This work has been supported by the following programs and institutions a 1990 1993 Fellowship from INIA Agricultural and Food Research Institute of Spain Ministry of Agriculture in cooperation with USDA OICD and the NC State University b 1997 Study Leave for Researchers Program of INIA Agricultural and Food Research Institute of Spain Ministry of Agriculture in cooperation with USDA OICD and the NC State University c 1997 ICIA Agricultural Research Institute of the Canary Islands d North Carolina Agricultural Research Service e USDA CSREES and Southern Region Research Project S 249 and S 273 f Univ of North Carolina Water Resources Research Institute g USDA CSREES and Southern Region Research Project S 1004 h Florida Agricultural Experiment Station FAES Par
175. ound cover or heavily grazed with no mulch Fair 50 to 75 ground cover and not heavily grazed Good 75 ground cover and lightly or only occasionally grazed b Poor lt 50 ground cover Fair 50 to 75 ground cover Good gt 75 ground cover c Actual curve number is less than 30 use CN 30 for runoff computations d CN s shown were computed for areas with 50 woods and 50 grass pasture cover Other combinations of conditions may be computed from the CN s for woods and pasture e Poor Forest litter small trees and brush are destroyed by heavy grazing or regular burning Fair Woods are grazed but not burned and some forest litter covers the soil Good Woods are protected from grazing and litter and brush adequately cover the soil 1 Average runoff condition and Ia 0 25 Part IV VFSMOD Appendices 168 Runoff curve numbers for arid and semiarid rangelands Table 2 2d From USDA NRCS 210 VI TR 55 2nd Edition June 1986 Cover Description Curve numbers for hydrologic soil group c Hydrologic b B c 5 over type oe Condition Herbaceous mixture of grass Poor 80 87 93 weeds and low growing brush with Fair 71 81 89 brush the minor element Good 62 74 85 Poor 66 74 79 aspen mountain mahogany bitter brush maple and other brush me 48 2 8 Good 30 41 48 Poor 75 85 89 Pinyon juniper pinyon juniper or Fair 58 73 80
176. ownface of sediment wedge C t amp D t effective filter The calculation procedure is as follows a select flow and sediment load at filter entry If strip was filled up in a previous step NFUP 1 bypass sediment deposition calculations add incoming and outgoing mass to totals and RETURN to main program b calculate the hydraulic properties at points 1 2 3 the filter to be used later on subroutine OCF c solve Einstein s bed load transport equation to find the transport capacity g gt at the end of B t subroutine EINSTEIN d calculate shape of sediment wedge sediment outflow and trapping efficiency for the filter subroutine STEP3 e position points 1 2 3 at system nodes so that flow rates can be read at those points at next time step subroutine POINTS f write outputs of sediment transport calculations 1 21 OCF NPLACE This subroutine solves the hydraulic properties for each of the filter s singular points by using Manning s equation and open channel flow theory It utilizes the method proposed by Barfield et al 1979 where the known values are S spacing of the filter media ele ments cm Se filter main slope n Manning s for cylindrical media s em q unit overland flow rate at the given point k cm7 s and the unknowns are dp depth of flow at D t cm V depth averaged velocity at D t cm s R hydraulic radius of the filter cm Notice that all units are in CGS system c
177. owsTM User s Manual 88 3 Using VFSMOD How Can the Model be Used This package can be used to comprehensively evaluate and develop designs for vegetative filter strips to trap sediment and other contaminants and enhance infiltration A typical application of the package would follow the outline below 1 2 3 4 5 6 Develop input datasets for UH to generate storm data for a typical upslope source area Run UH to develop input hydrograph and hyetograph data for VFSMOD Develop input datasets for VFSMOD for describing the filter strip Run VFSMOD to simulate the performance Modify any of the inputs for UH and or VFSMOD to better reflect target source area filter strip Use the Design Option to examine a range of storm events filter strip combinations to evaluate alternate possible designs After Step 5 or 6 an alternate path could examine the uncertainty associated with the proposed design Following this path the user can use the Sensitivity and Uncertainty Options to investigate The steps would be 1 2 3 4 Use the Sensitivity options to identify the most sensitive parameters for the design centered on the base input values for the target source area and filter strip Select the most sensitive parameters and assign these probability distributions Use the Uncertainty section to perform Monte Carlo Simulations Using the Analysis portion of the Uncertainty Section examine the probability
178. ple would be to create a project sample file that includes the newly created sample2 files and specifies sample igr and sample ikw as igr and ikw files see Section 1 3 on page 55 An example of a project file sample lis for UH is given in the following Table Unix Windows 9x NT 2000 XP inp inputs sample inp inp input sample inp iro inputs sample iro iro inputs sample iro imn inputs sample irn imn inputs sample im isd inputs sample isd isd inputs sample isd out output sample out out output sample out hyt output sample hyt hyt output sample hyt The project file in this example sample lis would be saved in the VFSMOD directory where the executable UH or UH EXE is located To execute the model with the project file the following would be entered uh sample lis In this example the input files would be read from the inputs subdirectory and the output files would be created in the output subdirectory In general the project file contains all of the keywords which are Inputs Outputs inp inputs for the source area for UH irn rainfall hyetograph input for vfsmod iro runoff hydrograph from the source area input for vfsmod isd sediment properties for the sediment filtration submodel out summary of the inputs and outputs from UH hyt detailed summary of of MUSLE calculations and the runoff hydrograph Part II VFSMOD and UH User s Manual 72 All inputs for UH are in FORTRAN77 free format The inp
179. portance of those specific data with respect to the rest typically with a range of 0 1 It is important to make sure that these data are distributed in three columns not in a single line Part III VFSMOD W WindowsTM User s Manual 120 EditPad Lite C wtsmod tinve E OE Pde Ect Search Glock Convert Options Wew Help a F iD J 1 m PBReGeeeteeeeeeeaeaaaag ll ll FIGURE 1 Fig Input text file with data of time hydrograph and the weight factor Calibration model parameters A total of ten parameters can be selected to calibrate the hydrograph A range including the minimum and maximum values for each parameter search must be input A description of those parameters is shown in the next table Table Parameters that can be used for the hydrograph calibration Parameter Units Name Vertical saturated conductivity m s VKS Average suction wetting front m SAV Saturated water content m3 m3 OS Initial water content m3 m3 Ol Maximum surface storage m SM Filter fraction where ponding is checked SCHK Filter width m FWIDTH Filter length m VL Filter Manning s roughness coefficient s m1 3 RNA Average filter slope m m SOA On the other hand nine parameter in total can be used to calibrate the sediments Similarly to flow the parameters for sediment calibration are shown below Table Parameters that can be used for the sedigraph calibration Part
180. posure concentrations with monitoring results from surface drinking water supplies J Agr Food Chem 2005 53 22 8840 8847 Laboratory and field conditions Transactions of ASAE 27 5 1321 1331 Hickey A U Doran B A review of the efficiency of buffer strips for the maintenance and enhancement of riparian ecosystems Water Qual Res J Canada 2004 39 311 317 Hollenbeck K J J Simunek and M Th van Genuchten 2000 RETMCL Incorporating maximum likelihood estimation principles in the RETC soil hydraulic parameter esti mation code Computers amp Geosciences 26 319 327 Part I VFSMOD W Model Documentation 49 Hopmans J W and J Simunek 1999 Review of inverse estimation of soil hydraulic properties In M Th van Genuchten F J Leij and L Wu eds Proc Int Workshop Characterization and Measurement of the Hydraulic Properties of Unsaturated Porous Media 643 659 University of California Riverside C A Huyer W and A Neumaier 1999 Global optimization by multilevel coordinate search J Global Optimization 14 331 355 James L D and S J Burgues 1982 Selection calibration and testing of hydrologic mod els Ch 11 In C T Haan H P Johnson and D L Brakensiek eds Hydrologic Mod eling of Small Watersheds 437 472 ASAE Monograph no 5 St Joseph ASAE Kizil U and L A Disrud 2002 Vegetative Filter Strips Modeling of a Small Watershed 2002 ASAE Annual International Meeting CIGR A
181. problem overland flow and infiltration The content of this the file is controlled by the input parameter IELOUT The first part of the file summarizes information read from the ikw iso and irn input files along with some of the calculated parameters needed for the simulation The second part of the file contains the inflow hydrograph from iro rainfall excess ie calculated with the Green Ampt model and the output hydrograph from the filter Only 100 time steps are printed to this file each one is the average of the precedent NWRITE steps where NWRITE NDT 100 b filename og1 The file contains information related with the sediment filtration model The first part of the file summarizes information read from the igr and isd input files along with some of the calculated parameters needed for the simulation The second part of the file contains sediment transport and deposition time series for the simulation period As before only a 100 time steps are printed to this file In this case the sediment filtration step is calculated with the average flow conditions calculated as described above c filename og2 This file contains the flow characteristics at the singular points 1 3 in and out as defined in Part I of this manual of the filter for the simulation period for the same 100 steps described above d filename osm This file contains a summary of the most relevant input parameters and output results including a sediment an
182. provements of this combined model over the GRASSF or SEDIMOT II models are the inclusion of a state of the art description of flow through the filter b changes in flow derived from sediment deposition c physically based time dependent soil water infiltration d handling of complex storm pattern and intensity and e varying surface conditions slope and vegetation along the filter VFSMOD UH and additional components are described in this Part I from a theoretical and modelling structure perspective The user manual for the command line versions of VFSMOD and UH is given in Part II along with annotated applications detailed description of input and output files and recommended input values Part II describes the integrated package VFSMOD W as a whole under the MS Windows environment Part IV contains appendixes with detailed description on model variables and a collection of tables with recommended inputs for a variety of soil climate and plant conditions Each Part builds on the previous ones Although the reader is encouraged to read through the sections in sequence to gain in depth knowledge of the system section II contains the essentials to run the MS Windows design oriented application Part I VFSMOD W Model Documentation 2 2 VFSMOD Model Components Processes and Solution Techniques VFSMOD is a field scale mechanistic storm based model designed to route the incoming hydrograph and sedimentograph from an adjacent field
183. put and the effect of media height was only visible for large events when the filter began to be inundated with sediment 3 6 Previous Testing and Applications VFSMOD was tested with natural events data at a North Carolina Piedmont Mu oz Carpena et al 1999 and a Coastal Plain Mufioz Carpena 1993 experimental sites Both sites had grass filter strips mixture of fescue bluegrass and bermuda grass with ratios of field to filter lengths from 4 5 1 to 9 1 The field area had varying slope from 5 10 and the filter strip somewhat less The soil types were Cecil sandy clay loam at the Piedmont site and Rains loamy sand at the Coastal Plain site Parsons et al 1991 In general good agreement was obtained between observed and predicted hydrology and sediment outflow values Some sources of variability were discussed to explain some anomalous events Researchers at the University of Guelph Canada tested the model against field experimental data Abu Zreig et al 1999 2001 They reported good agreement R 0 9 between model predictions infiltration volume and sediment trapping efficiency and measured values if actual filter flow widths discounting concentrated flow segments are used rather than total filter length Factors affecting sediment trapping in VFS were also studied using VFSMOD in a follow up study Abu Zreig 2001 Recently the program has been used to model the effect of VFS in a small watershed 72 Ha Kizil and Lowell
184. r table value Practice Factor P Save and Close C CREAMS GLEAMS Cc Close this Window Help The inputs for the UH program are entered in the text boxes Rainfall This is the total rainfall for the storm in mm Curve Number This is the NRCS curve number for the source area The range is from 0 to 100 There are tables in the appendices to select See NRCS SCS Curve Numbers on page 164 for Agricultural and Urban source areas The curve numbers Part III VFSMOD W WindowsTM User s Manual 94 given in these tables represent Antecedent Moisture Condition AMC II which is average moisture conditions Length This is the length of the source area in m or the distance from the edge of the area bordering the filter strip along the upslope line to the farthest upslope point contributing runoff to the filter strip Area This is the source area in hectares Slope This is the slope of the source area as a fraction or the slope divided by 100 Storm Duration The time of the storm in hours used to compute the hyetograph and hydrograph Storm Type This is the type of rainfall event 1 IA II or III Type I is typically associated with Hawaii coastal side of Sierra Nevada in southern California and the interior regions of Alaska Type IA is used to represent storms for the coastal side of the Sierra Nevada and the Cascade Mountains of Oregon Washington and northern California and th
185. r II risk assessments vegetation filter strips VFS are required on the label as a mitigation practice For example a typical label might read the following construct and maintain a minimum 3 0 m wide vegetative filter strip of grass or other permanent vegetation between field edge and down gradient aquatic habitat The VFS can reduce pesticide movement to streams by reducing runoff volumes Part I VFSMOD W Model Documentation 8 through infiltration in the filter strip s soil profile through contact between dissolved phase pesticide with soil and vegetation in the filter strip and or by reducing flow velocities to the point where eroded sediment particles with sorbed pesticide can settle out of the water NRCS 1999 Mufioz Carpena et al 1999 Abu Zreig et al 2001 Hickey and Doran 2004 Reichenberger et al 2007 and Stutter et al 2009 Therefore VFS can provide both retention and detention mechanisms through infiltration and hydraulic resistance Other potential mechanisms of pesticide removal include sorption of pesticides to vegetation and enhanced or phytomediated degradation of pesticides within the VFS However specification of the required VFS characteristics is largely subjective due to the lack of a predictive tool that can explain the wide range of field reported efficacies Two VFS with equivalent lengths slopes and vegetation characteristics may yield drastically different pesticide reductions dependent on the hy
186. r of bugs that may appear The graphical front end GUI for VFSMOD was developed using Visual Basic Professional Edition version 6 0 vb6 The Visual Basic source code is available upon request The Part III VFSMOD W WindowsTM User s Manual 86 programs UH and VFSMOD were developed in FORTRAN and the source code is supplied with the installation package This program is supplied to be installed using the Visual Studio Installer As such a number of controls dll s and other files are included and usually installed We have made no attempt to eliminate or reduce the package although this is a future desire If you attempt to bypass setup we have no idea what must be installed The install package includes the complete Win32 distribution for vfsmod The default installation directory is C vfsmod which can be changed to any location For example if the installation was done for D vfsmod then this directory would contain C vfsmod w Vfismod w exe Vfismod w hlp Uh exe Vism exe Sample2 lis Sample prj globalSensitivity exe globalSensitivity exe xml Readme txt Documentation Vism pdf Inputs Sample igr Sample ikw Sample irn Sample iro Sample isd Sample iso Sample2 igr Sample2 inp Sample2 iso Output Sample2 out SourceCode Uh Vism Inverse inverse cfg meas_hyd txt meas_gso txt start_inv exe star_inv ctf Inputs inverse igr Graphical user interface Windows Help file Utility program u
187. ratio RDR computed as SDR Mass of Sediment Exiting the Filter Mass of Sediment Entering the Filter RDR Runoff Exiting the Filter Runoff Entering the Filter From a design perspective we require the VFS to accommodate storms with return periods of at least 1 and 2 years and probably 5 and 10 years The first step in the analysis is to generate inputs into the VFS from the soils and crops present in the source study area for each of the design storms and soils selected for the analysis To do this the precipitation depths of selected return periods for the area along with the area s NRCS runoff and MUSLE erosion inputs are processed through the input preparation utility UH to create formatted inputs for VFSMOD hyetograph sample irn incoming sedimentograph sample isd and hydrograph sample iro With these inputs the VFSMOD model routes the incoming runoff and sediment and calculates water and sediment retained at the filter outflow and filter performance For this we must describe the actual vegetative filter strip characteristics to analyze for each design runoff event Usually the most relevant VFS characteristics to consider from a design prespective are soil type sample iso filter length uniformity and slope sample ikw and vegetation characteristics sample igr The VFSMOD sample project sample prj provided with the package installation for all platforms that can be used as a pattern and changed for each d
188. ration of the parameter vector VFSMOD is coupled with the Global Multilevel Coordinate Search GMCS algorithm Huyer and Neumaier 1999 This algorithm combines global and local search capabilities with a multilevel approach The GMCS is a good alternative to other existing optimization algorithms because it can deal with objective functions with complex topography it does not require powerful computing resources and initial values of the parameters to be optimized are not needed To refine the minimization of the objective function the GMCS is combined sequentially with the Nelder Mead Simplex NMS algorithm Nelder and Mead 1965 Figure 15 Further details about application of GMCS NMS to inverse modeling of soil hydraulic properties are given in Lambot et al 2002 and Ritter et al 2003 Furthermore model adequacy uncertainty and correlation associated with the estimated parameters is determined according to Hollenbeck et al 2000 and Ritter et al 2004 Part I VFSMOD W Model Documentation 39 VFSMOD parameters input files VFSMOD 15t Process VFSMOD parameters input files VFSMOD Figure 15 Scheme for the inverse modeling procedure to calibrate the VSFMOD parameters When calibrating parameters some criteria must be defined to evaluate the goodness of fit of the model simulation using the optimized parameters Several authors point out that to assess the performance of the model calibration the
189. rder sensitivity index S defined as a fraction of the total output variance attributed to a single factor can then be taken as a measure of global sensitivity of Y with respect to Xj Le Se 45 To calculate S FAST technique randomly samples the k dimensional space of the input parameters using a series of sinusoidal trayectory of changing phase The number of evaluations required in the analysis can be expressed as Part I VFSMOD W Model Documentation 35 N M k 2 46 where M is a number between 500 and 1000 For a perfectly additive model S 1 ie no interactions are present and total output variance is explained as a summation of the individual variances introduced by varying each parameter alone In general models are not perfectly additive and S lt 1 FAST was extended by Saltelli et al 1999 to incorporate the calculation of the total order effects through the total sensitivity index S7 calculated as the sum of the first and all higher order indices for a given parameter Xi For example for parameter number 1 Sry Sy FSu t Sykt FS yn ae and then Sri Si Sut Sikte S 48 For a given parameter X interactions can be isolated by calculating Sq S which makes the extended FAST a powerful method for quantifying the individual effect of each parameter alone S or through interaction with others S7 S An additional benefit of the Extended FAST analysis is that since the results are der
190. rectory This folder should not be deleted or modified by the user except when new files with the series of measured data for hydrograph or sedigraph need to be added The Matlab runtime component The inverse calibration procedure uses a Matlab runtime library to do the calculations During the installation of VFSMOD W the Windows Installer Package automatically installs this library in the computer For more information about this component please visit the MathWorks website www mathworks com Administrative privileges might be required to install correctly the Matlab component in the PC Please contact your computer administrator if you need help Input files When the calibration option is performed the user selects one o various model parameters in order to fit the simulated results using VFSMOD with a series of measured field data of water hydrograph or sediment sedigraph A range for each parameter selected must be specified to run the simulation The description for both the measured data file and the model calibration parameters to select are described in more detail below Measured data The measured data text file contains three columns The first column represents the time in seconds the second column contains the data of the outflow hydrograph m3 s or sedigraph g s that the program will seek to match during the calibration procedure and the third column gives a specific weight to each pair of data depending on the im
191. ree options Number of iterations Show graph Exit file name The number of iterations defines the value for np in the formula iter 100 np2 0 is set for automatic where Iter Number of iterations to be done np base value The graph option let the user to watch the graph that is generated during the calibration process This graph is a representation of the current field data hydrograph sedigraph and the results of the simulation with the data that have been selected for the calibration Finally the exit file name defines the name of the output file where the program will write the results of the calibration this will be written in the inverse directory Part III VFSMOD W WindowsTM User s Manual 127 Advanced setting options for calibration aJ VFSmod Windows Editor v 4 0 7 Fie Source_Area_ UH Filter_Strip_ F5 Design Sensitivity Uncertainty Calibration Options Window Help S Inverse Calibration Sedigraph Project Name sample pri pramen aeh Fie inverse meas_gso txt Browse Edit Select parameter s to be changed or calibrated No Change Calibrate ic Spacing for grass steam SS cm I Advanced Settings DER Roughness grass Manning s N sZem 173 ree lumber of iterations 0 eA Height of grass H cm 0 for automatic Exit file name 7 Roughness bare surface fittedparam Manning s n VN2 s m 1 3 Coarse sediment fraction Show graph y n yCn gt 0 0037 cm COARSE g 9 Incomi
192. responds with the desired simulation time chosen by the user typically coupled with a rainfall intensity of 0 Note also that each time corresponds to the beginning of the rainfall period i e storm such as Period Time interval s Rainfal m s 1 0 0 to 299 9 1693E 05 2 299 9 to 599 8 6773E 05 3 599 8 to 900 0 1101E 04 Would be input as 0000E 00 1693E 05 2999E 03 6773E 05 5998E 03 1101E 04 9000E 03 1947E 04 1 4 2 3 File example 12 1947E 04 0000E 00 1693E 05 2999E 03 6773E 05 5998E 03 1101E 04 9000E 03 1947E 04 1200E 04 1947E 04 1500E 04 1524E 04 1800E 04 5080E 05 2100E 04 1693E 05 2400E 04 2540E 05 2700E 04 8467E 06 3001E 04 0000E 00 3603E 04 0000E 00 Note the last pair of numbers is used to set the time when the simulation ends 1 4 3 filename iro runoff from the adjacent field into the VFS Part II VFSMOD and UH User s Manual 1 4 3 1 Structure of the file SWIDTH SLENGTH NBCROFF BCROPEAK BCROFF I J I 1 NBCROFF J 1 2 1 4 3 2 Definition SWIDTH Source area width m SLENGTH Source area flow path length m NBCROFF integer number of time steps of the incoming field hydrograph BCROPEAK peak flow of the incoming field hydrograph m s BCROFF I J incoming field hydrograph flow rate time s vs qin m s 1 4 3 3 File example 4 0 34 0 68 2192E 02 8417E 03 0000E 00 8716E 03 5724E 07 9018E 03 5724E 07 9317E 03 5724E 07
193. rge at each time step errors can accumulate and the ensuing numerical instabilities lengthen the simulation time r What is the H value vegetation height represent in the igr file This is explained in the User s Manual Remember H is not only the length height of the stem that does not topple in other words remains erect under flow conditions but also tells the model how high it can build the sediment wedge In your particular application case H 110 cm this is not a realistic value for grass Are you modeling grass or other species s Is there a way to vary the sediment concentration during the runoff event or are we forced to keep the concentration constant during a run Notice that the fact that an average inflow sediment concentration is considered does not mean that the sediment inflow is not dynamic Let me explain The important thing to consider is the sediment load gs M T When you multiply the average sediment concentration for the event i e total inflow sediment total inflow runoff volume by the hydrograph q t i e gs M T Ci M L3 x q L3 T you obtain a dynamic sedimentograph into the filter The basic assumption here is that these two sedimentographs i e one calculated from several samples through the event vs one using the average concentration for the event are not too different Furthermore that the difference between the resulting sediment deposition and outflow might be also small When developi
194. rmal triangular and uniform are available After selecting the distribution the Set Parameters button opens the window to enter the parameters defining the distribution For the normal and log normal distribution the mean and standard deviation are entered The peak and maximum and minimum values specify the triangular distribution The minimum and maximum values determine the range for sampling the uniform distribution If the user would like to also do the uncertainty analysis for the VFSMOD parameters they can switch to the VFSMOD screen gt S UH Uncertainty Parameter Selection Uncertainty Selections for Source Area Base Project Files sell lis After Selecting the Distribution Parameters Base Values Distribution Click on the Button to set the parameters V Curve Number CN 85 Triangular Soil Erodibiliy 9 33 Select Distribution EE J Crop Factor C fi Select Distribution EEEN loxi Triangular Distribution Parameters for Practice Factor P 1 Select Distribution w Samples 10 7 Set VFS Peak i A Load A Different Minimum Do Simulations Base Project Maximum Done Cancel Curve Number Similar to the UH uncertainty selection screen only selected parameters are available for uncertainty analysis for VFSMOD Currently the parameters are the saturated vertical conductivity and initial water content for the Green Ampt infiltration submodel for the filter strip and the
195. rs Mufioz Carpena et al 1999 and Poletika et al 2009 AP a b AQ c AE din F 1 e C 7 where AP is the pesticide removal efficiency AQ is the infiltration defined as the difference between total water input to the VFS i e rainfall plus inflow runon minus the runoff from the VFS AE is the sediment reduction C is the clay content of the sediment entering the VFS Fph is a phase distribution factor 1 e ratio between the mass of pesticide in the dissolved phase relative to the mass of the pesticide sorbed to sediment and a b c d and e are regression parameters i e 24 8 0 54 0 53 2 42 and 0 89 respectively Mathematically Fph was written as the following Q E KE 8 where Q and E are the volume of water L and mass of sediment kg entering the VFS and K_ is the distribution coefficient defined as the product of the organic carbon sorption coefficient K and the percent organic carbon in the soil divided by 100 Sabbagh et al 2009 Additional details of the derivation of this equation are provided in the supporting information see Supplemental Material Section S2 Parameters within this equation were used to represent some of the processes within the filter strip including infiltration AQ sedimentation A E and sorption F pr Degradation processes were not simulated in the VFS due to the assumption of a small residence time during typical rainfall runoff events Th
196. s Somplesi 10 Set UH Parameters g x Load A Different Do Simulations Base Project Cancel Help Once the simulations are complete the user can do some analysis using VFSMOD Selected storm outputs are saved in the output file The format of this file is space separated and is easily imported into another analysis package such as spreadsheet The first thirteen lines contain header and general information on the parameter and the base project file This information includes the parameters and their probability distributions and the base rainfall and filter strip length along with the base project filenames Lines 3 10 include the information on each input parameter 0 7 indicating the parameter selection of the probability distribution for sampling the inputs the options are 1 no uncertainty 0 normal 1 lognormal 2 triangular or 3 uniform and the parameters to define the probability distribution as shown below Distribution Distribution Parameters Number 1 Not Sampled 0 Normal Mean Standard Deviation 1 Lognormal Log of Mean and Log of the Standard Deviation 2 Triangular Peak Minimum and Maximum 3 Uniform Minimum and Maximum Part III VFSMOD W WindowsTM User s Manual 140 The tabular information presents the event level outputs and starts on the 14 line Each line contains the results for one of the simulations The columns are retv the return value for that s
197. s 101 6 3 VFS Buffer Vegetation Characteristics 121 cscceccsssssceseeescesceeeeeeeeeeeeseeseecaeenseeeenaees 106 6 4 Incoming Sediment Characteristics isd csccesesccesceeseescesceeeceseeeeecaeeaecaeeaeceeeseearens 107 6 5 Storm Hyetograph mn vec tek ck btewcs a r Maca Haass Rae 108 6 6 VFS Source Area Storm Runoff iro ccccceccesceescecceeseeseesceceecseecaecaeaececseeeeseeeesneeeneeaes 109 6 7 VFS Water Quality Input File iwq ccccceseeseceeeseeeecesceeeeeeceseeseecaeeseecaecaeeesrenseeseenes 111 6 8 VFS Description of the Output Files cccceesceseeseceeceeceeeeeeceseeseecaeeseecaeesseeeensenseerenes 113 7 Processing and Analysis of VFSMOD Resullts cccccssesssecsceeeeetseeeteeeteeeees 115 Using the Plot Wind OWS 206i iee 8rd sa eieiretesteln etn a eia iaa 117 9 Calibration Mode cca astnnecsutig ca an Sie a a a a a a eases 120 10 Sensitivity ADalysis SCTEEN Sas 1s 23schs sacar ahead aa a a i A Bers 133 11 Uncertainty Anal ysis SCreens 1 201 224 ear ie asire abs RREA oh TERRE 139 TD Desea MOU E E A E tues RAS 143 13 Troubleshooting vfsmod W ssssssesssessessessresseesresrosseessesresseessesersstessessrssressesse 145 14 VFSMOD W Change History ccecssccssccessccssscseccssecesssessccseccssscseescssansses 146 Part IV VESMOD Appendices ss ccdsesaccsicies aseeis edna Aces ovidns Gniane Geen ns As 150 1 APPENDIX 1 Description of the model subroutines 0 ce
198. s is not a good estimate and the simulation will become unstable or even blow up This can be avoided by lowering the CR at the expense of more simulation run time Instabilities can also be avoided by reducing increasing the number of nodes in the domain N b With large sediment input into the filter strip the program blows up Set ICO 0 In this case the sediment deposition is so large that the change in the nodal slope in the downstream face of the wedge creates problems to the finite element flow solution Petrov Galerkin does well but it does not perform miracles on a drastically changed domain Setting ICO 0 ignores the changes in slope and allows the simulation to be completed Previous comparison by the authors between runs with ICO 1 or 0 show that difference in results sre typically in the rage of 5 10 Part II VFSMOD and UH User s Manual 79 c Assigning values to KPG NPOL The order of the shape function used in the numerical solution finite elements can be set to linear NPOL 2 quadratic NPOL 3 or cubic NPOL 4 Please note that if the order of the function is changed to any other type than quadratic recommended the regular finite element formulation will be run instead of the improved Petrov Galerkin method One could also select a regular quadratic finite element by setting KPG 1 Tests made during program development show the increase in execution time induced by selecting the PG method are small as compared wi
199. selects Foster s Method 2 selects Williams method and 3 selects the CREAMS GLEAMS method The program produces two output files that summarize the program execution In this case these are sample2 out and sample2 hyt The sample2 out file contains a printout of the input data along with the runoff hydrograph and a summary The sample2 hyt file contains the information about the rainfall hyetograph along with the outputs related to the erosion from the storm From these results the input files for VF SMOD sample 2 iro sample2 irn and sample2 isd are also automatically created in the output directory file sample2 out File output sample2 o ut UH v1 06 3 2002 HYDROGRAPH CALCULATION FOR WATERSHED SCS METHOD Inputs Storm Rainfall 80 00 mm SCS storm type II Storm duration 6 0 h SCS Curve number 72 0 Watershed area 5 00 ha Maximum flow path length 100 00 m Average slope of flow path 2 00 MUSLE type 2 where 1 Foster 2 Williams 7 3 GLEAMS See Manual Outputs Runoff volume 22 82 mm 1141 16 m3 Initial Abstraction 19 76 mm Concentration time 0 19 h 11 64 min Peak flow 0 3753 m3 s 27 0228 mm h Time to peak 0 65 h 38 76 min Hydrograph based on SCS unit hydrograph time h q m3 s 0 00 0 0000 0 06 0 0019 0 13 0 0178 0 19 0 0562 0 26 0 1139 0 32 0 1813 0 39 0 2472 0 45 0 3032 0 52 0 3440 0 58 0 3678 0 65 0 3753 0 71 0 3687 0 78 0 3511 0 84 0 3256 0 90 0 2953 wR OO 21
200. shington D C 2007 www epa gov oppefed1 ecorisk presenta tions setac_slc htm Liu X Zhang X Zhang M Major factors influencing the efficacy of vegetated buffers on sediment trapping A review and analysis J Environ Qual 2008 37 1667 1674 McKay M D W J Conover R J Beckman 1979 A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code Technometrics 21 239 245 McKay M D 1995 Evaluating prediction uncertainty NUREG CR 6311 U S Nuclear Regulatory Commission and Los Alamos National Laboratory McCuen R H W J Rawls and D L Brakensiek 1981 Statistical Analysis of the Brooks and Corey and the Green Ampt parameters across soil textures Water Resour Res 17 4 1005 1013 Mein R G and C L Larson 1971 Modelling the infiltration component of the rainfall runoff process Bulletin 43 University of Minnesota MN Water Resources Research Center Part I VFSMOD W Model Documentation 50 Mein R G and C L Larson 1973 Modeling infiltration during a steady rain Water Resourc Res 9 2 384 394 Morgan M G and M Henrion 1990 Uncertainty Cambridge University Press Cam bridge MA Morris M D 1991 Factorial sampling plans for preliminary computational experiments Technometric 33 161 174 Mufioz Carpena R 1993 Modeling hydrology and sediment transport on vegetative filter strips Ph D dissertation North Carolina
201. smod These files include data for the hyetograph the runoff hydrograph and sediment loss Part I VFSMOD W Model Documentation 29 Read Input Data for Source Area Calculate Runoff by SCS Method Calculate time of concen tration Peak Flow Time to Peak by SCS TR55 Create Runoff Hydrograph from SCS unit hydrograph Develop i hyeto graph from SCS storm type Calculate Storm erosion using MUSLE amp average Sediment Concentration in Runoff Write Input Files for VFSMOD Figure 12 Computations in UH 3 5 Sensitivity Analysis of VE SMOD A sensitivity analysis was performed to gain some insight in the dependence of model outputs on certain model parameters and to assist in the model calibration Mu oz Carpena et al 1999 The study showed that the main parameters controlling the hydrology outputs were soil hydraulic conductivity and initial water content whereas the model was fairly insensitive to changes in saturated water content and suction at the wetting front values Previous research Mufioz Carpena et al 1993a showed that Manning s surface roughness controls mainly the time to peak of the outgoing hydrograph Testing on the sediment component of the model showed that the main parameters controlling sediment outflow are media spacing and particle diameter Variations in the Part I VFSMOD W Model Documentation 30 modified Manning s roughness had relatively little effect on the out
202. smodPesticides inputs sampleP iwq Buffer Vegetation Properties Water Quality Properties finputs sampleP iwq Incoming Sediment Character file iwq Storm Hyetograph Pesticides Distribution Coefficient Source Area Storm Runoff 5 Direct Kd KOC 0C Water Quality Properties Input z bl iz Sediment Clay content 40 Sediment Transport Flow through VFS Detailed Hydrographs Water and Sediment Balance Save Continue Editing Save and Close Close Window Help Overall Summary Water Quality Summary This file is only required and read when CWQ 1 in the input file IKW In this case the file must be present in the inputs directory of the application The water quality component is specified in the first line of this input file The parameters required in the Part III VFSMOD W WindowsTM User s Manual 111 rest of the file depend on the type of water quality component selected Currently only a pesticide component is available in this version Option 1 Pesticides using the windows interface Contents of the new IWQ file with description of parameters required only the no degradation case is handled in the GUI other cases can be handled through the IWQ text file see Users Manual l Type of wq problem0 1 pest cide Bayer l 147 30 Kd proc sO KALKA 1 Koe Koc LKA 000 40 Clay contentin incoming sediment Option Selection using the Windows Interfa
203. step values of n and S Part I VFSMOD W Model Documentation 14 are selected as nodal values for the finite element grid The parameters are fed back into the hydrology model for the next time step thus surface changes due to sediment deposition within the filter sediment wedge area are accounted for in the next time step of the hydrology simulation as described in the previous section 2 6 Model inputs The program reads inputs model parameters and model input variables from external ASCII files which can be prepared from given examples using a conventional text editor A summary of the model inputs is given in the following Table Class Inputs Green Ampt infiltration Rainfall hyetograph soil saturated hydraulic conductivity soil saturated water content soil initial water content soil suction at the wetting front and surface storage Overland flow Field inflow hydrograph filter length filter width nodal slopes and Man ning s roughness across the filter Sediment filtration Modified grass Manning s roughness Manning s roughness for bare soil incoming sediment characteristics median particle size weight density fall velocity effective filter media spacing and height porosity of deposited sediment incoming sediment inflow concentration sedimentograph and proportion of fine sediment Water quality transport Transport parameters for pesticide simple first order decay sol
204. strips can lead to increased release of phosphorus to waters A biogeochemical assessment of the mechanisms Env Sci Tech 2009 43 1858 1863 Part I VFSMOD W Model Documentation 52 Suwandono L J E Parsons and R Mufioz Carpena 1999 A design guide for vegetative filter strips using VFSMOD Presented at the 1999 ASAE CSAE Ann Intl Meet ing 19 20 July Paper 99 2147 ASAE St Joseph MI Tingle C H Shaw D R Boyette M Murphy G P Metolachlor and metribuzin losses in runoff as affected by width of vegetative filter strips Weed Sci 1998 46 4 475 479 Tollner E W B J Barfield C T Haan and T Y Kao 1976 Suspended sediment filtra tion capacity of simulated vegetation Transactions of ASAE 19 4 678 682 Tollner E W B J Barfield C Vachirakornwatana and C T Haan 1977 Sediment deposi tion patterns in simulated grass filters Transactions of ASAE 20 5 940 944 USDA NRCS 210 VI TR 55 2nd Edition June 1986 U S Environmental Protection Agency GENEEC User s Manual Gen eric E stimated E nvironmental C oncentration Model U S EPA Office of Pesticide Programs Washington D C 2001 www epa gov oppefed1 models water geneec2_users_manual htm U S NRCS Formerly Soil Conservation Service National Engineering Handbook Hydrology Section 4 1972 and USDA ARS 41 172 1970 Vieux B E V F Bralts L J Segerlind and R B Wallace 1990 Finite element watershed modeling one dimens
205. t eo 3 z a e l 0 Input k Model input factor PDFs Global Sensitivity f Analysis GSA 5 Figure 14 General schematic diagram of the global sensitivity and uncertainty analysis Numbers in circles represent the steps in the global evaluation procedure explained in the text Part I VFSMOD W Model Documentation 37 5 Inverse Calibration Modeling water solute and or sediment transport is nowadays widely used for assessing the impact of human activities on water resources and for designing best management practices to reduce these impacts Particularly the vegetative filter strip model system VFSMOD W allows for predicting water and contaminant transport through vegetated filters VFSMOD W developed by Mufioz Carpena and Parsons 1999 2005 simulates water and sediment transport in vegetated filters based on overland flow hydraulics and infiltration into the soil matrix The success in modeling such processes heavily depends on the quality of the model parameters i e they are representative of the hydraulic properties of the soil and the vegetated filter A popular method for parameter estimation is manual calibration by a trial and error procedure comparing simulated values of runoff sediment outflow from the vegetative filter with those experimentally measured However this method is time consuming subjective since the modeler does not know when to stop the calibration process it is d
206. t I VFSMOD W Model Documentation 47 11 References Abu Zreig M 2001 Factors affecting sediment trapping in vegetated filter strips simula tion study using VFSMOD Hydrological Processes 15 8 1477 1488 Abu Zreig M Rudra R P and H Whitley 2001 Validation of a vegetated filter strip model VFSMOD Hydrological Processes 15 5 729 742 Abu Zreig M R P Rudra and H Whitley 1999 Sediment trapping in vegetative filter strips Presented at the 1999 ASAE CSAE Ann Intl Meeting 19 20 July Paper 99 2078 ASAE St Joseph MI Arcement G J and V R Schneider 1989 Guide for selecting Manning s roughness coef ficients for natural channels and flood plains U S Geological Survey Water Supply Paper No 2339 Barfield B J E W Tollner and J C Hayes 1978 The use of grass filters for sediment control in strip mining drainage Vol I Theoretical studies on artificial media Pub no 35 RRR2 78 Institute for Mining and Minerals Research University of Kentucky Lexington Barfield B J E W Tollner and J C Hayes 1979 Filtration of sediment by simulated vegetation I Steady state flow with homogeneous sediment Transactions of ASAE 22 5 540 545 Barfield B J L G Wells and C T Haan 1981 Applied Hydrology and Sedimentology for Disturbed Areas Oklahoma Technical Press Stillwater Berthouex P M and L C Brown 2002 Statistics for Environmental Engineers Boca Raton Lewis Pub Burns L A
207. t be left and incorporated i Soil surface and chopped residues for matured preceding crop undisturbed except in narrow slots in which seeds are planted j Top of old row ridge sliced off throwing residues and some soil into furrow areas Reridging assumed to occur near end of cropstage 1 k Where lower soil loss ratios are listed for rows on the contour this reduction is in addition to the standard field contouring credit The P value for contouring is used with these reduced loss ratios 1 Field average percent cover probably about three fourths of percent cover on undisturbed strips m If again seeded to WC crop in corn stubble evaluate winter period as a winter grain seeding lines 132 148 Otherwise see table E 9 n Select the appropriate line for the crop tillage and productivity level and multiply the listed soil loss ratios by sod residual factors from table E 10 o Spring residue may include carryover from prior corn crop p See table E 9 q Use values from lines 33 62 with appropriate dates and lengths of cropstage periods for beans in the locality r Values in lines 109 122 are best available estimates but planting dates and lengths of cropstages may differ s When meadow is seeded with the grain its effect will be reflected through higher percentages of cover in cropstages 3 and 4 t Ratio depends on percent cover See table E 9 Part IV VFSMOD Appendices 175 u See item 12 table E 8 3 6 Contour
208. t parameters are independent and uncorrelated an assumption that is often made then the second term is 0 ie Cov P P 0 The slope of the sensitivity relationship between O and P is S With these assumptions the variance equation becomes n Var 0 S VarP 42 i 1 This type of analysis produces good estimates of the mean and variance of the output parameter O when the coefficient of variation Mean Standard Deviation of the input parameter is small and the relationship between O and P over the range of potential inputs is linear An alternative more general approach is the technique of Monte Carlo Simulations MCS An outline of this procedure is Part I VFSMOD W Model Documentation 33 1 select the most sensitive input parameters 2 develop probability distribution functions for each input parameter 3 randomly generate input parameter datasets based on the probability distributions 4 perform the model simulation with the randomly generated input dataset 5 repeat steps 3 and 4 for a large number of trials 6 generate probability distribution functions for the model outputs of interest and 7 use the output probability distribution functions to evaluate uncertainty in the model by placing confidence levels on the outputs Additional details on the application of this procedure can be found in Parsons and Mufioz Carpena 2001 4 2 Global Sensitivity and Uncertainty Analysis 4 2 1 The Morris Me
209. ter the user selects a range on the desired parameters This new version also automatically produces combined analysis output tables see Part III Section 12 on page 143 Additionally the program provides two powerful tools Once the optimal design parameters are selected an uncertainty analysis can be conducted using the graphical tools provided The objective of this analysis is to identify the level of confidence that the adopted design has against the uncertainties present when selecting the model inputs Parsons and Mufioz Carpena 2001 2002 Finally a sensitivity analysis procedure is included in the GUI to identify the parameters to which the model is more sensitive for a given scenario thus allowing the user to economize effort by focusing on better identifying just the sensitive parameters Parsons and Mufioz Carpena 2001 An example of design results see Mufioz Carpena and Parsons 2002 is included below The graph depics the optimal filter lengths to achieve a 75 sediment reduction SDR 0 25 in a North Carolina Piedmont site clay soil 0 5 Ha source area 2 slope 6 hr storm duration with a grass mixture vegetation on the filter Filter lengths from 14 57 m are needed to accomodate storm events associated to 1 10 year return periods The design assumes homogeneous sheet flow across the filter in all cases Scheme for the inverse 0 9 0 8 0 6 0 5 T 10 yr T 2 yr 0 3 T 1 yr Sediment Delivery Ratio SDR
210. th extension PRJ This is a text file that specifies the names of the input and output files of dissimilar names and allows for the user to quickly combine soil types and hydrological events represented by different input files created by the user for advanced analysis When the CWQ 1 the same rule applies but the new input file IWQ needs to be present or specified in the project file along with the new water quality output file OWQ In the previous example executing vfsm sample with the CWQ 1 in the IKW file will require the set of files sample ikw sample im sampe iro sample iso sample isd and sample iwq in the inputs directory An example of project file for this case is specified in File 2 Sample project file sampleP prj showing the specification of the new water quality input and output files ikv inputsisample ikw isg inpwtasarmple iso igr inputsisample igr isd inputsisample isd im inputsisample irr irosinputsisample io og outputsample ag og d outputisample ag2 ohy outputisample ohy osm output sample osm ea i When executing the program either way the new input files are shown during the beginning of the run 6 1 Overland Flow Inputs ikw Part III VFSMOD W WindowsTM User s Manual 98 R Vrsmod Windows Eor 021 ae E iE File Source Area UH Filter_Strip_ VFS Design Sensitivity Uncertainty Calibration Options Window Help B gt vfsm Project sample prj Working Direc
211. th the gains in stability and accuracy obtained Thus the setting NPOL 3 and KPG 1 are recommended d Assigning values to N The number of nodes of the system must be an odd number for a finite element quadratic solution an even number if the solution is cubic and any of them if linear The program adjusts the number of nodes automatically if the requirement is not made e If no incoming sediment characteristics are known d Zs In the absence of measured inflow sediment characteristics an estimate of the particle size could be made by knowing the soil texture of the contributing field Woolhiser et al 1990 d in x10 cm Soil texture USDA Expected d Soil texture Expected d Clay 0 45 Clay loam 5 30 Silty clay 2 45 Sandy loam 35 160 Silty clay loam 3 46 Loamy sand 90 180 Silt loam 3 50 Sandy clay 2 130 Silt 8 30 Sandy clay loam 21 160 Loam 9 60 Sand 140 200 f Setting the total simulation time DR The last time interval of the rainfall series file irn is used to set the desired simulation typically with the rainfall intensity set to 0 g Reducing execution time by stopping surface changes during the simulation ICO 0 Setting the flag ICO 0 in the igr file will stop the model from reshaping the entrance of the filter during sediment deposition This in turn will result in a reduction of the total execution required since the problem will become less non line
212. than that one used in the current project The option Calibrate activates two dialog boxes Min and Max Those are used to define the range minimum and maximum for the parameter to be calibrated If the user changes his her mind about a selection the option No can be chosen in order to run the calibration mode with the current values set in the original project without a modification The same steps depicted above are used for the Sedigraph option Part III VFSMOD W WindowsTM User s Manual 126 24 FSmod Windows Editor v 4 0 7 File Source_Area_ UH Filter_Strip_ VF5 Design Sensitivity Uncertainty Calibration Options Window Help S Inverse Calibration Sedigraph Project Name sample pri Browse ens inverse meas_gso txt Browse Edit Select parameter s to be changed or calibrated No Change Calibrate New value Min ee Spacing for grass steam SS cm Roughness grass Manning s WN sma J C Height of grass H em Roughness bare surface La s Manning s n VN2 s m 173 C Coarse sediment fraction d gt 0 0037 cm COARSE g g C Cc Incoming Flow sediment concentration Cl g cm3 Porosity of deposited sediment kn POR 4 100 Oo el Sediment particle class diameter 450 DP cm ec C Sediment particle density SG g em3 e me ear ee te FIGURE 10 Interface for Sedigraph calibration NNN The Advanced settings options let the user to select th
213. thod The screening method proposed by Morris 1991 herein Morris method or Morris and later modified by Campolongo et al 2005 was used in this study because it is relatively easy to apply requires very few simulations and its results are easily interpreted Saltelli et al 2005 Morris 1991 proposed conducting individually randomized experiments that evaluate the elementary effects relative output differences of changing one parameter at a time Each input may assume a discrete number of values called levels that are selected within an allocated range of variation for the parameter For each parameter two sensitivity measures are proposed 1 the mean of the elementary effects u which estimates the overall effect of the parameter on a given output and 2 the standard deviation of the effects o which estimates the higher order characteristics of the parameter such as curvatures and interactions Since sometimes the model output is non monotonic Campolongo et al 2005 suggested considering the distribution of absolute values of the elementary effects u to avoid the canceling of effects of opposing signs The number of simulations N to perform in the Morris analysis results as N r k 1 43 where r sampling size for search trajectory r 10 produces satisfactory results k number of factors Although elementary effects are local measures the method is considered global because the final measure u is obtain
214. through a vegetative filter strip VFS and to calculate the outflow infiltration and sediment trapping efficiency The model handles time dependent hyetographs space distributed filter parameters vegetation roughness or density slope infiltration characteristics and different particle size of the incoming sediment Any combination of unsteady storm and incoming hydrograph types can be used VFSMOD consists of a series of modules simulating the behavior of water sediment and pollutants in the VFS The current modules available are Figure 1 FIELD SOURCE VEGETATIVE FILTER STRIP AREA i RAINFALL Infiltration module Overland flow module INFLOW l PESTICIDES Sediment filtration SOLUTES gt module COMPLEX MULTIREACTIVE POLLUTANTS OUTFLOW Figure 1 Schematic representation of VFSMOD modules i Green Ampt infiltration module a module for calculating the water balance in the soil surface ii kinematic wave overland flow module a 1 D module for calculating flow depth and rates on the infiltrating soil surface iii sediment filtration module a module for simulating transport and deposition of the incoming sediment along the VFS VFSMOD is essentially a 1 D model for the description of water transport and sediment deposition and pollutant trapping along the VFS The model can also be used to describe transport at the field scale or field edge if flow and transport is mainly
215. time s rainfall rate m s 0 000001693 a 299 9 000006773 Maximum rainfall intensity for 599 8 00001101 900 0000194 the storm RPEAK m s 1200 00001947 1500 00001524 1947E 04 1800 00000508 2100 000001693 2400 00000254 2700 0000008467 Plot Hyetograph 3001 0 3603 0 Save Continue Editing Save and Close Close Help m NRAIN integer number of rainfall periods including period to end simulation RPEAK maximum rainfall intensity for the storm m s time s and rainfall rate o intensity m s over the VFS for each period RAIN I J The last time step corresponds with the desired simulation time chosen by the user typically coupled with a rainfall intensity of 0 Note also that each time corresponds to the beginning of the rainfall period The hyetograph can be viewed by selecting the Plot Hyetograph button Part III VFSMOD W WindowsTM User s Manual 108 24 02 36 03 Time min Copy Plot to Clipboard Print Plot Edit Plot 6 6 VFS Source Area Storm Runoff iro The runoff hydrographs from the source area can be manually entered or generated using the UH program See the UH program documentation for further information Part III VFSMOD W WindowsTM User s Manual 109 Storm Runoff inputs sample iro Hydrograph iro Enter Time s Runoff rate m 3 s l 0000005 Source Area Width 00000005724 SWIDTH fm 4 0 00000005724 Source
216. tion is assumed to be Ja 0 2 S Tables for selecting the curve number CN are given in Appendix 3 of this manual see also NRCS 1984 In the case of multiple land uses a composite CN can be derived using a weighted average of the respective CN based on the land use areas As in the original derivation of the method Q is set to 0 if the precipitation is less than 0 2 of S This Part I VFSMOD W Model Documentation 18 assumes that the precipitation does not replenish the available storage ie 0 2 S 3 2 2 Peak flow calculation using NRCS method SI units Based on the triangular hydrograph assumption the time to peak can be stimated as eb pai pe Gh O 16 Where the concentration time can be estimated by the following equations SL te Fy that thy pe 17 4 where ty is the transit time for each of the segments of the path between furthest point to the watershed outlet from a hydraulics point of view L and v are lengths and flow velocities for each segment The velocity can be estimated from Haan et al 1994 Table 3 20 pg 76 When there is little information on flow paths an alternative equation is used in UH 1000 pos 1000 9 t ail 18 e 4407 JY where t is in hours CN is the NRCS curve number L in m is the maximum linear distance to the watershed outlet Y m m is the average slope altitude difference between furthest point and outlet divided by L As a third option there
217. tions at the start of the event dthivvadded out s Urfacg Outflow dout Sediment deposition 7 4 i r 1 op k Figure 2 Domain discretization for the finite element overland flow submodel Rainfall excess i in equation 1 is calculated for a given rainfall distribution for each node and time step by the infiltration model The hydrograph representing runoff from the adjacent field is input as a time dependent boundary condition at the first node of the finite element grid The program allows for spatial variation of the parameters n and S over the nodes of the system Figure 2 This feature of the program ensures a good representation of the field conditions for different rainfall events The model can be operated to provide Part I VFSMOD W Model Documentation 5 information on the effect of soil type infiltration slope surface roughness filter length storm pattern and field inflow on VFS performance i e reduction of the runoff peak volume and velocity Mufioz Carpena et al 1993b It also describes the flow rate q velocity V and depth A components throughout the filter for each time step The numerical solution is subject to kinematic shocks or oscillations in the solution that develop when a sudden change in conditions slope roughness or inflow occurs When linking the kinematic wave and the sediment transport models the soil surface conditions are also changed for each time step furth
218. tory Filter Strip Project File ECE B vfsm Editing C vfsmod w inputs sarnple ikw tol o JEa Overland Flow Properties file ikw inputs sample ikwy Simulation Title Unit9 g8 u1 83 91 Overland Flow Input Buffer Dimensions Infiltration Soil Prop Buffer Length m VL 8555 source Area aw Vegetative Filter Stip Width of the Strip m FwIDTH N Flow Direction View Edit Buffer Segment Properties slope roughness gat iagh rence Sediment Transport Kinematic Wave Numerical Solution Paramteters Incoming Sediment Width sicng Sou ee Ava Edge FWID H Storm Hyetograph Source Area Storm A Filter Length Y Flow through VFS Number of Nodes in Solution Domain N 57 Courant Number CR 0 5 0 8 EE Time Weight Factor THETAW 0 5 5 Nunbe of Element Nodal Points NPoL 3 3 Mavinuneaions MAXITER 350 Petrov Galerkin Solution Regular Finite Mo t ipu Elenen H nfomation 1 Element KPG 1 recommended or 01 Save Continue Editing Save and Close Close Window Help Overall Summary LABEL FWIDTH VL THETAW CR MAXITER NPOL IELOUT KPG NPROP SX I RNA I SOA I a label max 50 characters to identify the program run width of the strip m length of the filter strip m number of nodes in the domain integer must be an odd number for a quadratic finite element solution but the program checks and corrects if needed time weight factor for the Crank Nich
219. tration is allowed to reach its maximum potential for the rest of the run The assumption here is that the incoming field runoff moving at the surface will supply enough water to sustain the maxi mum infiltration for that time step This means that the effective rainfall fed into the kine matic wave equation ie will be in most cases a negative value The procedure is as follows a check if the end of runoff has been reached b check for surface ponding at beginning yes NPOND 1 no NPOND 0 b 1 without surface ponding at the beginning of the period NPOND 0 b 1 1 with ponding at the end of the period Cu gt 0 b 1 2 no ponding at the end of the period Cu lt 0 b 1 3 Find values at the limit of this rainfall period regardless of time step b 2 with surface ponding at the beginning of the period NPOND 1 c return ie value to be used in that time step 1 14 FORMB B0 X0 Q0 N BCRO PGPAR In this subroutine the right hand side part of the matrix equation vector b is assembled in the following steps a find dx1 for integration rule b initialize vector b c begin vector formation element by element c 1 Initialize temporary vectors c 2 do integration point loop c 3 plug the element vector into the b0 vector d Plug in the boundary condition b 1 BCRO Part IV VFSMOD Appendices 153 1 15 MODIFY QM B BCRO PGPAR In this subroutine the right hand side part of the equation vector b is assembled follow ing th
220. trov Galerkin parameters i 1 4 i th shape function at XI nodal a in Manning s uniform flow equation flow vector at iteration m cm2 s flow vector at previous time step m2 s maximum flow rate at steady state condition cm2 s input coming and out going flow runoff from filter cm2 s rainfall excess at the node lateral inflow m s RAIN L 2 FPI Manning s roughness coefficient s m 1 3 slope of the element duration of the simulation s Source area width m Part IV VFSMOD Appendices 158 SLENGTH SX I TE THETAW VL VNI W 5 5 WF I X 5 5 X1 5 5 XM N XON 2 2 Infiltration AGA BGA CU DM F FPI L LO NCHK NEND NPOND NPFORCE OI OS PS PSOLD PST RAIN 200 2 RO SCHK RPEAK SAV SM STO TP TPP TRAI VKS Source area flow path length m Distance from origin to the start of the i th surface segment m Henderson s time to equilibrium s time weight factor Filter length m Mean filter Manning s roughness coefficient s m 1 3 Gauss quadrature weights modified weighting functions solution vector dimension 1xn at time step 1 1 Gauss quadrature point solution vector xn at iteration m time step I 1 solution vector xn surface water depth h m at time step Green Ampt s A saturated hydraulic conductivity Ks m s Green Ampt s B Ks Sav DM m s Chu s surface ponding indicator at end rain period lt 0 ponded initial soil water deficit DM OS
221. use of a single statistic might be misleading and more should be used along with graphical representations Berthouex and Brown 2002 James and Burgues 1982 Tufte 1983 Legates and McCabe 1999 The goodness of fit of the simulations with the optimized parameters was evaluated in terms of the coefficient of efficiency Nash and Sutcliffe 1970 and the root mean square error The coefficient of efficiency Ceff has been widely used to evaluate the performance of hydrologic models It compares the variance about the 1 1 line perfect agreement to the variance of the observed data and it ranges from 8 to 1 Thereby Ceff 1 implies that the plot of predicted vs observed values matches the 1 1 line Legates and McCabe 1999 The root mean square error also called residual variation or standard error of estimate RMSE is a useful single measure of the prediction capability of a model since it indicates the precision with which the model estimates the value of the dependent variable Berthouex and Brown 2002 Part I VFSMOD W Model Documentation 40 6 Design Procedure The design objective is to find optimal constructive characteristics length slope vegetation of a VFS to reduce the outflow of sediment from a given disturbed area soil crop area management practices to achieve a certain reduction in sediment i e that for TMDLs Proposed target outputs for analysis will be the sediment delivery ratio SDR and runoff delivery
222. utes and mul tireactive transport General Number of nodes for the domain Courant number for numerical solution total time for the simulation Part II of this manual gives suggested literature values for some of these parameters when no field measurements are available In the case of the soil hydraulic and sediment parameters these can be chosen from soil texture using tables from the manual The structure of these files is discussed in detail in Part II Section 1 4 on page 57 Part I VFSMOD W Model Documentation 15 3 UH utility preparation of model inputs for design purposes As an aid to set up the model inputs the distribution package includes an utility UH that creates synthetic model inputs for the upslope source area based on the NRCS SCS design storm for a given location and soil type The utility implements the NRCS SCS curve number unit hydrograph and Modified Universal Soil Loss Equation MUSLE concepts to produce ready to use input files for VFSMOD These inputs are rainfall hyetograph field inflow hydrograph and field sediment inflow and characteristics UH and VFSMOD are intended to be run in sequence for a design case After running UH the remaining VFSMOD inputs needed pertain only to the vegetative filter strip characteristics dimension soil vegetation and numerical solution parameters The structure of UH input and output files is discussed in Part II Section 2 3 on page 73 The
223. uts are contained in filename inp Note that filename could and should be replaced by any other name you would like to identify the case study with max 8 characters as in the example above A description of this file follows 2 3 UH input files 2 3 1 filename inp parameters for generating inputs for VFSMOD 2 3 1 1 Structure of the file P CN A storm type D L Y soiltype K CFACT PFACT dp IEROTY OM 2 3 1 2 Definition P CN A storm type D L Y soiltype K CFACT PFACT dp IEROTY OM amount of storm precipatation in mm NRCS SCS Curve Number for the source area see Appendix 3 Area of the upstream portion in ha storm type 1 I 2 II 3 HI 4 Ia storm duration h Length of the source area along the slope m Slope of the source area expressed as a fraction See Table for Acceptable Soil Types Soil Erodibility If K lt 0 then K is computed based on texture and organic matter See eq 31 C factor See Table in Appendix 3 P factor See Table in Appendix 3 particle size 1 selection basis of texture otherwise user given NOTE Because dp is one of the most semsitive parameter for sediment transport Mufioz Carpena and Parsons 1998 whenever possible the user should provide this from measurements Select the method to compute the storm R factor in MUSLE not present or 1 selects Foster s Method 2 selects Williams method and 3 selects the CREAMS GL
224. ve Continue Editing Save and Close Close Help integer incoming sediment particle class according to the USDA NPART 1975 particle classes NPART Particle class Diam range cm d cm MV cm s cm s 1 Clay lt 0 0002 0 0002 0 0004 2 60 2 Silt 7 0 0002 0 005 0 0010 0 0094 2 65 3 Small aggregate 0 0030 0 0408 T 80 4 Large aggregate 0 0300 3 0625 T 60 5 Sand 0 0050 0 2 0 0200 3 7431 2 65 6 Silt 2 0 0002 0 005 0 0029 0 0076 2 65 T User selected DP model SG of particles from incoming sediment with diameter gt 0 0037 cm COARSE coarse fraction that will be routed through wedge unit fraction i e 100 1 0 Cl incoming flow sediment concentration g cm POR porosity of deposited sediment unit fraction i e 43 4 0 434 Part III VFSMOD W WindowsTM User s Manual 107 DP sediment particle size diameter d5 cm read only if NPART 7 SG sediment particle density s g cm read only if NPART 7 Note COARSE and DP are related so that their values need to follow the this rules COARSE DP gt 0 5 gt 0 0037 0 5 0 0037 lt 0 5 lt 0 0037 6 5 Storm Hyetograph irn The hyetograph input files can be manually entered or generated using the UH program See the UH program documentation for further information fe B vfsm Editing CAvfsmod w inputs sample irn fo S jes Storm Hyetograph irn inputssample irn Enter
225. vfsmod package The registration information is included on the Options screen This information is stored in the file vfsmod w cfg in the installation directory We suggest that you e mail this file to us as an attachment Part III VFSMOD W WindowsTM User s Manual 92 5 UH Project Window As an aid to set up the model inputs the distribution package includes a utility uh that creates synthetic model inputs based on the NRCS SCS design storm for a given location and soil type The utility implements the NRCS SCS curve number unit hydrograph and Modified Universal Soil Loss Equation MUSLE concepts to produce ready to use input files for VFSmod These inputs are rainfall hyetograph field inflow hydrograph and field sediment inflow and characteristics The files used by uh are identified in the project window There are also options to Save the project Edit an input file and Browse Select a different input file In addition any of the input or output filenames can be changed from this window Sethe aes aie Save Heb Source Area UH Input Fie Edit Output Files Runoff Hydragraph for VFSmod Run This Project Rainfall Hyetograph for FSmod eee Yetograph ISD Input file for VFSmod inputs sample2 isd User Output Information Part 1 Hydrograph User Output Information Part 2 output sample2 hyt View Output Files Graph a Runoff Option buttons provide shortcuts to the UH menu entry these include th
226. w reached the check node NCHK and zw gt 0 t lt tw for the WT case the infiltration is at capacity regardless of ponding or not from point excess calculation Part III VFSMOD W WindowsTM User s Manual 147 05 2012 Changes in vfsm v4 2 0 Pdegr New in filter mass pesticide mass balance calculated at the end of the event Prepared for integration in EU SWAN registration tool Degradation subroutine to calculate degradation of sediment bonded and mixing layer residue at the end of the event towards the beginning of the next event It uses FOCUS equations that need daily surface soil temperature and moisture for every day betwwen two consecutive runoff events Daily eair temperature i e PRZM files or other source can be used as mixing and surface soil moisture Moisture content can be estimated based on running mass balance For additional details see EU AIM and SWAN reports http abe ufl edu carpena vfsmod FOCUSreports shtml 08 2012 Changes in vfsm v4 2 1 Fixed minor bug for case when only lateral inflow is provided no rain like in some model testing laboratory scenarios UCL Belgium Outflow hydrograph now ends in zero if sufficient time is provided 08 2013 Changes in vfsm v4 2 3 Fixed bug in duration of the simulation handling IRN file If the user did not add a double end line in the IRN i e end of simulation to the IRN file the program now sets the end of the simulation to the longest of the
227. y soils 1to4 maximum runoff potential vertisols clyas limiting layers Otol Part IV VFSMOD Appendices 165 Runoff curve numbers for cultivated agricultural lands From USDA NRCS 210 VI TR 55 2nd Edition June 1986 Table 2 2b Cover Description Curve numbers for hydrologic soil group Cover Type Treatment Ai A E g 2 condition Bare soil Ut 86 91 94 Fallow Crop residue cover CR Poor 76 85 90 93 Good 74 83 88 90 Straight row SR Poor 72 81 88 91 Good 67 78 85 89 SR CR Poor 71 80 87 90 Good 64 75 82 85 Contoured C Poor 70 79 84 88 Good 65 75 82 86 Row crops C CR Poor 69 78 83 87 Good 64 74 81 85 Contoured amp terraced Poor 66 74 80 82 C amp T Good 62 71 78 81 C amp T CR Poor 65 713 79 81 Good 61 70 77 80 SR Poor 65 76 84 88 Good 63 75 83 87 SR CR Poor 64 75 83 86 Good 60 72 80 84 Cc Poor 63 74 82 85 ell Se Good 61 73 81 84 C CR Poor 62 73 81 84 Good 60 72 80 83 Poor 61 72 79 82 Good 59 70 78 81 C amp T CR Poor 60 71 78 81 Good 58 69 77 80 SR Poor 66 77 85 89 Close seeded or oe 2 i broadcast C Poor 64 75 83 85 legumes or rota Good 55 69 78 83 tion meadow C amp T Poor 63 73 80 83 Good 51 67 76 80 1 Average runoff condition Ia 0 2S Part IV VFSMOD Appendices 166 a Crop residue cover applies only if residu
228. ydrology problem e get the Gauss quadrature parameters f assemble the system matrix A g perform LU decomposition over A h Start the time loop to solve the problem for each time step h 1 select the rainfall intensity and BC transform into depth m at the first node of system incoming hydrograph for each time step h 2 get effective rainfall and control execution of overland flow for an infiltrating surface by calling Green Ampt model The assumption is that when a certain node NCHK is flooded i e X NCHK gt 0 all the surface will be flooded and thus the maximum infiltration capacity for the rest of the event is selected as given by the Green Ampt model NCHK is selected by the user h 3 form of r h s vector for that time step h 4 start Picard iteration h 4 1 update b b h 4 2 feed the vector to the solver h 4 3 check for convergence h 4 4 update X X h 4 5 find flow component at iteration step h 4 6 Picard iteration converges proceed with time step otherwise repeat h 6 update h and q for next time level h 7 do the following only 100 times each time using the average flow of the last NWRITE values in between h 7 1 call sediment transport subroutine if there is inflow change units from Q m7 s to QSED cm s h 7 1 write outputs to files h 8 repeat time loop for next time step until the end of the run i write a summary of results at the end of the run j close files and end

User`s Manual - Agricultural and Biological Engineering

Contents

Download Pdf Manuals

Related Search

Related Contents