Home

user manual - Cawthron Institute

1. A CAWTHRON reverse the concentrations read to match the reversed grid file Rhyhabsim we re reading in outline rev unlist strsplit xSindata drconclines dl cat outline file openfile sep t append TRUE cat n file openfile append TRUE cat n file openfile append TRUE close openfile rm openfile store the concentrations openfile file initialconc txt open r concstowrite readLines openfile close openfile rm openfile for nnn in l length concstomrite cat concstowrite nnn file icstore append TRUE cat n file icstore append TRUE store the velocities for nnn in 1 Ncolrow 2 cat xsindata 7 Ncolrow 2 1l nnn file velstore append TRUE eg for five layers this gives rows 14 to 18 incl cat n file velstore append TRUE store the NREI for this cross section in an archive for all sections openfile file resultsOutNREI txt newNREI readLines openfile close openfile rm openfile newNREI as numeric unlist strsplit newNREI 12 length newNREI dim newNRET c 3 length newNRET 3 newNREI t nevNREI NREIstore rbind NREIstore newNREL crossSec crossSectl close icstore close velstore write out the archived NREI for all cross sections in this section at this flow write csv NREIstore file NREIstore txt row names FALSE col names NA
2. 23 CAWTHRON P resultsOutMaxNREI Notepad lol x Fie Edit Format View Help Best point 1524 11 1488 43 Best cross section index 18 Best NREI 4 46009 4 Deleatidium 5mm 10 0897664 112103 118084 13013 7 128271 139525 135856 145892 141545 150891 146314 154958 150372 15833 0 161552 153828 161132 156783 163515 158965 165361 eleatidiumy 5mm 10 0902707 112733 118748 130868 128991 140309 13662 0 146712 0 150801 14234 0 151738 0 15552 14 7136 0 155828 54217 0 159219 o 118239 135067 143946 149959 154651 158433 0 162933 164133 166312 168011 120658 136804 145596 151547 156168 159879 0 115414 165458 167579 169233 119724 136188 145194 151297 148576 130879 0 114892 0 133197 0 142758 0 149275 0 111141 0 0888365 0 0779102 Q 103942 0 0740857 0938035 0 0751464 0 0 0 0 Q Q Q 0895488 075974 o 118903 135826 144755 0 152398 157045 159323 162459 121336 137573 146414 0 147413 129017 160777 163849 120397 136953 415537 133946 14356 Q 0 ooo OOODODODOOCO 151576 0738813 0732343 0 0 Q 0 0 0 0 0 0 0 D 0 0 0 0 0 0 0 COO ODOOTFOOCO COODDDOOF Oooo OOO OOO OOQOOCOCO Coo OOO OOO OOOQODCO 0998801 a Figure 22 Example of resultsOutMaxNREL txt output from the foraging model e epsOut eps is an EPS format file Figure 23 that shows a cros
3. 16 Figure 15 An output file generated by CrossSections gt Dump for Foraging Model from the drift model 17 This file format has sufficient data to reconstruct the volumetric invertebrate density distribution for graphical display further investigation evaluation validation etc Note that the coordinates provided in the file are for the corners of cells while the invertebrate densities specified are for the centres of the cells thus defined Some external manipulation Will be required cccssssssssssssssssssssssssssssesscssscessessesseseeseee 17 Figure 16 A cross section dump from the drift model 18 Figure 17 The Set H Params dialog DOX ssssssssssscssscsssssessesssssessssssscsssssssessessssessessesessess ee be Figure 18 Plan view of the foraging model showing the geometry of prey interception sssesssersessseees 21 Figure 19 Cross sectional view of the geometry used by the foraging model showing fish position foraging radials and predicted prey capture area interpolated from predicted foraging radius along each foraging radial based on prey reaction distance and water velocity Hughes ef al 2003 sesssesesesesessseseseses 21 Figure 20 A paramsIn txt file specifying fish and habitat parameter sccscsssssssesseeseees Figure 21 A resultsOutNREI txt output from the foraging model En Figure 22 Example of resultsOutMaxNREL txt output from the foraging model sssssssesssssssseeseeseress 24 Ju
4. 1507 55 1481 1506 84 1482 1506 13 1483 1505 42 1484 1512 73 1475 1512 1476 25 1511 26 1477 1510 52 1478 1509 79 1479 1509 05 1480 11 25 31 15 2 87456 3 00863 2 73659 2 28285 18 1 50517 0 66158 0 611676 0 810771 0 862871 REF 45 85 64 2 14704 3 10628 3 17056 2 97592 Figure 21 A resultsOutNREI txt output from the foraging model 2 resultsOutEFF txt contains the same X Y coordinate data in the same format as resultsOutNREI txt and a measure of foraging efficiency in place of NREI Foraging efficiency is calculated as GREI swimming costs resultsOutCosts txt contains the same X Y coordinate data in the same format as resultsOutNREI txt and the swimming costs calculated for each of these points resultsOutMaxNREI txt presents data on the best foraging spot found in the modelled reach Figure 22 It provides the X and Y coordinates of the best location the cross section in which it is located and the NREI at this point It also describes the drift density that remains in each of the streamtubes at this cross section after a fish located at the best point has fed i e it reflects drift depleted by fish foraging at that optimum point It provides this information in essentially the same format as the Initial drift densities file input to the drift model see section 4 2 Drift model data requirements only the initialisation cross section is not given June 2012
5. Hughes and Kelly 1996 e Swimming cost equations 1 Stewart et al 1983 and e 2 Ritchell et al 1977 parameterized for steelhead rainbow trout from Fish Bioenergetics 3 0 Hanson et al 1997 e Swimming cost equation for brown trout in Hayes et al 2000 based on an equation and parameters for rainbow trout in Rand et al 1993 and parameters for brown trout from Elliott 1976 A2 1 Polynomial swimming cost model coho salmon This model is a polynomial fit to Brett and Glass s 1973 swimming cost data for coho salmon fry developed by Hughes and Kelly 1996 It takes the form In SMR u e L AMB SMR n u e t W u e e O e ms 3600 1000 72 where SC is the swimming cost Joules t is time s W is fish mass g OQ isthe oxycaloric equivalent Joules mg of On taken as 14 1 after Videler 1993 SMR is the standard metabolic rate mg of Oz Xg hr oxygen consumption at rest AMR is the active metabolic rate mg of O2 kg hr oxygen consumption at maximum swimming speed ums And u is the swimming speed in this case taken to equal ums since the fish is assumed to swim at its maximum sustainable swimming speed throughout its foraging forays A2 2 Rainbow trout swimming cost models These models were sourced from Fish Bioenergetics 3 0 Hanson et al 1997 and are reproduced below The parameters for the equations are listed in Table 1 The basic form of the
6. Sciences 44 832 845 Elliott J M 1976 Energetics of feeding metabolism and growth of brown trout Salmo trutta L in relation to body weight water temperature and ration size Journal of Animal Ecology 45 923 948 Elliott J M Davidson W 1975 Energy equivalents of oxygen consumption in animal energetics Oecologia Berlin 19 195 201 Elliott J M 1982 The effects of temperature and ration size on the growth and energetics of salmonids in captivity Comparative Biochemistry and Physiology 73 81 91 Gordon N D McMahon T A Findlayson B L 1992 Stream Hydrology An introduction for ecologists John Wiley and Sons Chichester U K 526p June 2012 29 Guensch G R Hardy T B Addley R C 2001 Examining feeding strategies and position choice of drift feeding salmonids using an individual based mechanistic foraging model Canadian Journal of Fisheries and Aquatic Sciences 58 446 457 Hanson P C Johnson T B Schindler D E Kitchell J F 1997 Fish Bioenergtics 3 0 for Windows University of Wisconsin Madison Center for Limnology and University of Wisconsin Sea Grant Institute Hayes J W Hughes N F Kelly L H 2003 Overview of a process based model relating stream discharge to the quantity and quality of brown trout habitat Proceedings of the International IFIM Workshop Fort Collins Colorado June 2003 Hayes J W Hughes N F Kelly L H 2007 Process based modelling of invertebrate drift transport
7. Cross section locations can be specified explicitly by selecting X Sec gt Draw and then double clicking on the desired location in the display window This constructs a cross section perpendicular to the maximum discharge at that location black line Figure 3 As cross sections are created they are oriented with the first point origin on the true right bank Care should be taken to ensure that this initial cross 3 Note that in RhyHabSim the origin of each cross section is on the true left bank June 2012 6 x CAWTHRON section is at the upstream end of the reach and spans from one no flow boundary a bank to the other X sec gt Clear all will clear all cross sections from the modelled area for instance if the first manually positioned one does not span from one bank to the other Cross sections can also be constructed by specifying the coordinates of their end points by selecting Specify in the X Sec menu and entering the endpoint coordinates of the desired cross section into the dialog box The X Sec gt Fill command automatically constructs a series of cross sections at a user specified spacing downstream from the initial cross section Figure 4 The program estimates the point of maximum discharge on a cross section moves downstream in the direction of this maximum flow by the distance the user has specified for the spacing and then orients the new cross section at right angles to the average flow repeating
8. Figure 12 This file can be created by deleting the first three lines from a copy of the resultsOutMaxNREL txt file except for the cross section number The edited file will contain e the number of the cross section at which the new run is to be initialized e the number of drift classes e matrices of drift concentrations in each streamtube at that cross section for each drift class preceded by the drift class name and the number of horizontal and vertical streamtube divisions Save this new text file Then reinitialize the drift model with this new file using DriftModel gt Initialize Leave the other parameters the same as in the initial run and rerun the model by selecting DriftModel gt Disperse The foraging model is run again using the output from this truncated run of the drift model to find the location of the next best fish location in the modelled reach in terms of NREI Net Rate of Energy Intake is still computed along all the cross sections in the reach but only points on cross sections downstream of the most recent initialisation cross section are ranked as potential fish locations So in the first instance the first cross section checked for the best location has index 0 1 1 which is the second cross section from the top of the modelled reach If the drift model had most recently been initialized at cross section 10 then the foraging model would predict NREI for all cross sections but only search for the best positions
9. and the abundance and behaviour of stream invertebrates will influence the quality and quantity of habitat for drift feeding fish The outputs from one component become the inputs for the next Figure 1 The administrative aspects of managing the input and output files and assembling the results can be handled by a script in a language such as R Matlab etc The process based approach is an advance over existing instream habitat procedures such as habitat modelling within the Instream Flow Incremental Methodology IFIM because its predictions of growth rate and fish abundance are more readily understood and biologically realistic than abstract indices of habitat suitability e g Weighted Usable Area Habitat modelling has been used for many years in attempts to predict the impact of changing flow regimes on stream life particularly salmonids The Instream Flow Incremental Methodology is the most common framework within which habitat models have been applied since the 1970s Bovee 1982 Stalnaker et al 1995 Annear et al 2002 A recurring criticism of conventional habitat modelling has been its reliance on empirically derived habitat suitability curves relating observed habitat use to physical habitat characteristics Empirical habitat suitability curves and the linkage between fish and their food supply are two main areas in which conventional IFIM is weak in biological realism Orth 1987 The impacts of aspects other than physical habitat on
10. as a new file It takes the form of a tab delimited text file where the first row gives column names x y depth drift class name and the following rows are the data i e drift concentrations No m This query works by finding the cross sections upstream and downstream of each point the column containing the specified X and Y and the row containing the specified depth The concentrations are bi linearly interpolated between the upstream and downstream cross sections The final method of data extraction is June 2012 17 Xs CAWTHRON 4 CrossSections gt Dump will create a text file describing the physical characteristics of the streamtubes broken down by column horizontal divisions and row vertical divisions for each cross section Figure 16 This dump exists for troubleshooting internal calculations in extensive detail and will be of little or no interest to the day to day user e column width the average width of the column m e column roughness the effective roughness height of the bed beneath that column e right bottom and left bottom the XYZ coordinates of the bottom corner nodes for each column e areas the surface areas of the polygons defining each streamtube in the given row across that cross section m i e viewed looking through the cross section from either upstream or downstream e velocities the average velocities m s in each streamtube in the given row for each column across the cro
11. drift Our model overcomes this obstacle It predicts how variation in bed topography discharge invertebrate abundance and behaviour will affect drift density in a spatially explicit manner The model s predictions of drift density have been shown to match well with measured invertebrate drift distribution Hayes et al 2003 2007 June 2012 11 CAWTHRON Figure 9 The lateral exchange of drift between streamtube cells in a cross section based on a series of Gaussian distributions Each different colour shaded area denotes the proportion of drift from a lateral position in source streamtube A exchanged into destination streamtubes B and C The total proportion of drift exchanged from streamtube A with streamtubes to the left and right is based on the average overlap of the destination streamtube with the Gaussian distributions centred on all possible lateral positions within column A see Appendix 1 June 2012 12 CA SH a HRON 07 640 OB 530 F 0 5 520 H lt 0 4 f 510 H 0 3 500 0 2 1490 a4 1480 H L L L L L f L f L 1490 1500 1510 1520 1530 1540 1550 1560 1570 1580 X Figure 10 A plan view of modelled invertebrate drift dispersed down through a river reach For this illustration the centre four streamtubes were given an initial concentration at the top of the reach of zero so the invertebrate density in these streamtubes lower in the reach has dispersed in from
12. employed depending on the situation For example a minimum NREI value that would provide for maintenance of condition as well as accumulation of reserves for reproduction might be specified as the cut off value Note If the drift model must be shut down or crashes during the iterative process of modelling multiple fish locations it is possible to recover the position you were in without rerunning the entire process Simply reopen the drift model and initialize the model at the top cross section as usual and disperse Then proceed to re initialize the model at successive cross sections down the reach using the drift initialization files that you had already created based on successive resultsOutMaxNREL txt files rerunning the drift model again between each initialization When you reach the position you had obtained before closing the drift model program you can pick CrossSections gt Dump for Foraging Model and carry on as usual 6 2 Automated approach The command line interface to the drift model allows it to be controlled by scripts in external languages In this way many repeated modelling runs can be incorporated into investigations by repeatedly modelling the same reach with different configurations such as flow initial drift density etc Script based control also allows the fish placement to proceed slightly differently from that described under the default graphical interface approach When run through the graphical interface the fi
13. fish is placed then the depleted densities output from the forage model are passed to the drift model If no fish is placed then the densities output from the drift model that had already been passed to the forage model are passed back in to the drift model for the next cross section Passing of input by the script to the model programs is of course implemented by saving files of the data to be passed under filenames that match those specified in a batch command that the script then calls An example of an R script implementing this procedure is included as Appendix 4 6 3 Defence radius While the drift depletion halo effect should preclude the close packing of fish a defence radius mechanism was introduced to ensure this This requires the radius to be specified via an extra parameter in the paramsIn txt file Number Parameter Example 1 Fork length of the fish m 0 5 2 Weight of the fish g 2500 3 Minimum prey size m 0 00575 4 Maximum prey size m 0 226 5 Probability of prey detection 0 55 6 Height of focal point above bottom m 0 1 7 Number of interrogations per cross section 36 8 Water temperature 16 9 Defence radius m 1 The input data in paramsIn txt are stored in a single space separated line e g 0 5 2500 0 00575 0 226 0 55 0 1 36 16 1 0 Maintaining the defence radius spacing is handled in the model code rather than in the script code If the location being evaluat
14. is required to produce reliable energetics equations for rainbow trout A2 3 Brown trout swimming cost model This model is sourced from Hayes et al 2000 It is based on an equation and parameters for rainbow trout in Rand et al 1993 and parameters for brown trout from Elliott 1976 R aew ee ee 4 1868 86400 R is energy cost Joules s W is fish weight g T is temperature C V is velocity swimming speed ms a bi bo b3 are parameters listed in Table 2 Table 2 Parameters for the brown trout swimming cost equation from Hayes et al 2000 a b bo were sourced from Elliott 1976 and bz was sourced from Rand et al 1993 and Stewart 1980 cited from Rao 1968 Temperature C a b bo b3 3 8 7 1 4 126 0 734 0 192 2 34 7 1 19 5 8 277 0 731 0 094 2 34 June 2012 Appendix 2 39 Appendix 3 Displaying model outputs Appendix 3 40 The XYZ data contained in the text file outputs of the foraging model and the drift model can be visually displayed as 2D colour contour plots This involves interpolating between data points from the model output to produce an image that gives the impression of a smooth continuous distribution which allows for more effective visualization of the position of favourable drift feeding habitat in the modelled reach There are several options for contour plotting the NREI output including e contouring functions in a GIS package e purpose written Matlab
15. number of vertical streamtubes and tick box to specify constant velocity rather than default logarithmic velocity profile with depth Saving the StreamTubes output creates a lightly annotated ASCII text file Figure 8 The filename used will need to be specified in the configuration of the drift and foraging components described below The format is as follows number of cross sections horizontal number of streamtubes vertical number of streamtubes Then the following repeated for each cross section cross section number and discharge at that cross section this discharge will be constant bed roughness height at each horizontal division across the section a numbered set of many nodes X Y Z depth below surface defining the corner coordinates of all streamtube polygons at that cross section the node indices describing all the streamtube polygons of that cross section P Untitled Notepad 3 File Edit Format View Help number of x secs 86 number of flow tubes 10 aixi un number of VvertTubes al 3 78891 0 117276 0 16 0 16 0 16 0 16 0 16 0 16 0 16 0 14648 0 036 0 036 1 1512 24 1473 41 0 2 1511 33 1474 78 0 464509 3 1511 33 1474 78 0 320028 1 2 3 di 1 3 4 1 ak 4 5 ne 2 3 79778 0 0693797 0 16 0 16 0 16 0 16 0 16 0 16 0 16 0 16 0 036 0 036 1 1512 15 1472 RS N Figure 8 Example of a StreamTubes output file See text for explanation of values There are memory leaks in the Stream
16. provided by the drift model to predict the spatially explicit net energy gain of fish drift feeding in that study reach The foraging model describes the fish s prey capture rate as a function of water depth and velocity invertebrate drift density and the fish s visual and swimming abilities The foraging model estimates the energy gained due to prey capture and the energy expended due to foraging which is also influenced by water temperature to predict Net Rate of Energy Intake NREI at all cells in the modelled reach Positive NREI values indicate that fish should be able to maintain condition or grow at least in the short term The foraging model will determine the single optimum foraging location for a trout of a given size based on maximum NREI of all cells in the streamtubes representation in its input file If multiple fish locations are needed the user needs to implement special program management by script as discussed below 3 STREAMTUBES MODEL 3 1 Purpose of the model This process converts a River2D Steffler et al 2003 description of the flow through a stream reach into a StreamTubes representation with a number of equal discharge streamtubes The process is embedded in a custom variant of Steffler s River2D program called R2DStreamTubes with the interface that will be familiar to users of that software A flow solution in River 2D CDG file format is the sole input required for this program Figure 1 The rea
17. the maintenance of fish populations have generally not been taken into account in these models This can potentially lead to a situation where two locations are judged to be of equal habitat value based on their physical characteristics when in fact they might differ markedly in their ability to support fish because of spatial variation in food availability In contrast our process based modelling approach focuses on the energy balance of fish as a basis for assessing habitat quality and quantity It recognizes that fish cannot survive grow or reproduce if the energy available from their food intake does not exceed the energy expended in obtaining that food The modelling process focuses on how discharge affects the availability of food to drift feeding fishes as well as their energy expenditure while foraging Modelling the transport and dispersion of invertebrate drift in relation to flow alleviates the need to assume uniform drift densities in the foraging model which has been a limitation of some earlier models e g Hughes 1992 Addley 1993 Guensch et al 2001 By modelling these dynamics in a spatially explicit way comparisons can be made between the energetic profitability of different areas for drift June 2012 1 CAWTHRON ra CA feeding fishes Energetic profitability net rate of energy intake can be further interpreted to predict spatially explicit growth potential as well as numbers and the distribution of fish in a rive
18. the presence of upstream fish in positions that are profitable but less so than the empty river optimum 6 1 Manual approach The optimal locations and ultimately the carrying capacity of fish within the reach can be modelled by an iterative process using the maximum NREI output from the foraging June 2012 25 CAWTHRON model resultsOutMaxNREIL txt This process can also provide insight into the effect of depletion by feeding fish on the density of invertebrate drift downstream The foraging model output resultsOutMaxNREL txt contains the best fish location in the modelled reach as defined by maximum NREI It also provides a matrix of the drift concentrations remaining in each streamtube at that cross section after a fish at that location has fed To model the impact of this depletion on the drift dispersion it is possible to rerun the drift model using the same StreamTubes model and invertebrate parameters as the initial run but using an adapted version of the resultsOutMaxNREL txt to initialize the model run at a new cross section To do this the drift model program DriftModel exe must be kept open following its initial run This is so that it can keep track of the drift dispersion upstream of the point at which it is initialized in subsequent runs A new initial drift concentration file can be created for the truncated model run by copying and pasting the pertinent parts of resultsOutMaxNREL txt into a new text file
19. the River 2D Bed window Figure 3 Use the File gt Open menu selection to open a flow model solution output in CDG format The file will open with the nodes and boundary displayed June 2012 5 CAWTHRON CAWTHRON CCF2P1_BC01 cdg R2DStreamTubes Oj x File Edit View Display Bed X Sec StreamTubes Prey Query Help ajsa vel Sle Bl elol Ja al elel x 476061 475541 y 5125859 550894 Figure 3 R2DStreamTubes exe interface window with boundary displayed and first cross section drawn The first step is to triangulate a mesh from the topography data and interpolate flow values using the Bed gt Triangulate menu item None of the program s other functions will work until you instruct the program to triangulate the CDG mesh The triangulation process may take a moment The resulting triangulated mesh can be displayed by toggling on Display gt Triangulation You are now able to display a number of properties e g bed elevation water depth or velocity as contour lines or colour shading in the same way as in River 2D programs As the mouse is moved around in the display window the Status Bar displays the X and Y coordinates the bed elevation water depth and flow in the X and Y directions 3 3 Creating cross sections The first step in the conversion to StreamTubes representation is to divide the modelled reach by a series of cross sections This process is performed using the functions in the X Sec menu
20. the subsequent streamtubes their number determines the number of cells per streamtube length Each virtual cross section is subdivided by a common user defined number generating a series of streamtubes with each conveying an equal fraction of the total discharge through the cross section In the vertical dimension water velocities for each streamtube are calculated assuming either a logarithmic velocity profile with depth Gordon et al 1992 or constant velocity with depth Although this manual describes a model for constructing a StreamTubes representation from a hydrological model from River 2D the output from 1D hydraulic surveys as undertaken in traditional IFIM habitat modelling can also be converted to a StreamTubes representation and used as input to the drift and foraging models For example Version 3 30 of Rhyhabsim Jowett 1999 incorporates a StreamTubes conversion process It can output simulated flow information for a series of cross sections in a representative reach as a StreamTubes representation of that river reach In this document StreamTubes is capitalised when it refers to the software tool or the method of representing a river reach and not capitalised when referring to an individual streamtube or the hydrological concept of a streamtube If this seems arbitrary think of the capitalised version as a brand Batteries not included 2 Note though that a difference between R2DStreamTubes and RhyHabSim in the
21. undepleted This implies that once the first fish has been placed at the optimum position in the whole reach subsequent fish will only be placed downstream of this This assumption may be appropriate for modelling positions of large fish in a short pool but clearly it is not appropriate for a long modelling reach The process involves identifying the cell in the modelled reach when empty of fish with the highest NREI value A fish is placed there depleted drift is propagated downstream NREI recalculated downstream and the next highest NREI cell is identified in that downstream section and so on The automated approach described below allows fish to be placed upstream of the overall optimum position in the reach This is suitable for larger reaches where positions exist sufficiently far upstream from the optimal location that a dominant fish would not defend those positions if other fish were to occupy them Such a scenario will occur with long modelled reaches and where the distribution of multiple size classes of fish is modelled In the latter case large fish will not compete with small fish because they feed on different sized prey The automated approach is also suited to dealing with the complexities that arise when upstream fish beyond the range of perception of the dominant fish would deplete its drift supply sufficiently to displace it That is the optimal foraging position in a river empty of fish may become non profitable due to
22. 57 0 160721 0 157377 0 158684 0 145969 0 11163 0 11026 0 0869275 0 0 0 0 0 144559 186199 229452 274883 323539 0 0 0 0 0 185717 238835 293933 351701 413418 0 Q 0 0 Q 222158 285401 350941 419571 492766 0 217573 0 279544 0 343772 0 346575 0 411038 0 414375 0 482792 0 486692 0 219365 0 281834 0 Q 0 0 0 201925 0 154788 25955 0 199285 319297 0 245488 381902 0 293993 448733 0 345906 0 15284 0 119453 0 196734 0 152822 0 242303 0 18725 0 29013 0 223124 0 341301 0 2610 0 0 377958 469368 0 482172 593429 0 0 57408 0 562531 0 567047 0 523086 0 403905 0 39844 0 302725 702547 0 688849 0 694206 0 642034 0 500295 0 492913 0 3627 75316 20159 17511 14505 11023 1 51783 1 35918 1 35119 1 30183 1 20719 1 03816 1 12969 0 929884 0 655026 1 05611 0 965683 0 966518 0 930671 0 860165 0 728174 0 781669 0 6729 0 499903 1 0333 0 945442 0 946468 0 911357 0 842222 0 712629 0 764662 0 659272 0 490724 1 0074 0 922456 0 923687 0 889393 0 821823 0 694962 0 745332 0 643762 0 480263 0 977379 0 895787 0 89727 0 863927 0 798166 0 674492 0 722962 0 625785 0 468123 06883 0 941673 0 864029 0 865782 0 833589 0 769994 0 650123 0 696327 0 60433 0 453605 01759 0 897436 0 824655 0 826741 0 79596 0 735056 0 619931 0 663365 0 577714 0 435545 950143 0 839105 0 772613 0 775106 0 746209 0 688
23. 871 0 580076 0 619915 0 542467 0 411518 849522 0 751906 0 694583 0 697602 0 671525 0 619586 0 520434 0 555015 0 489467 0 375122 505745 0 456979 0 434811 0 440963 0 42413 0 389485 0 319871 0 334812 0 312929 0 260195 0 0 0 0 0 0 0 0 0 0 0 L 1 1 1 1 1 1 0 0 0 4 Deleatidium 5mm Deleatidium 7_5Smm Deleatidium8_5mm Deleatidium9 5mm 0 0065 0 0075 0 0085 0 0095 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0014 00217 00319 00448 178 0 178 Q 178 Q 178 0 178 0 178 0 178 0 178 0 178 178 0 178 0 178 0 178 0 178 178 0 178 0 178 0 178 0 178 178 0 178 0 178 0 178 0 178 coco OOOO Figure 15 An output file generated by CrossSections gt Dump for Foraging Model from the drift model This file format has sufficient data to reconstruct the volumetric invertebrate density distribution for graphical display further investigation evaluation validation etc Note that the coordinates provided in the file are for the corners of cells while the invertebrate densities specified are for the centres of the cells thus defined Some external manipulation will be required A third option for data extraction is DriftModel gt Query This allows the output to be queried at user defined points It requires an input text file with columns defining the X and Y coordinates and the depth negative Z values of interest The output must be saved
24. Format View Help 1 49573 0 876479 0 56 6 1 59569 0 497439 1 18005 33 5749 22 5417 la 1 49573 0 876479 28 3 0 458772 0 143017 0 708028 27 9791 22 5417 1 1 49573 0 876479 2 14 15 0 0413564 0 0128924 0 236009 41 9687 22 5417 0 2 99715 0 876488 0 56 1 84671 0 532912 1 18005 33 5749 23 4241 1 2 99715 0 876488 1 28 3 0 404096 0 116612 0 708028 27 9791 23 4241 1 Figure 24 A resultsOutForaging txt output of the foraging model In this case Figure 24 prey type 2 has been dropped from the diet due to morphological constraints You can t tell from the information provided here but this prey is only 0 0015 m long and a 0 003 m minimum prey size was used in the simulation This prey type would not have been dropped from the diet for energetic reasons based on Charnov s 1976 prey model because the profitability of the drift class is greater than the fish s gross rate of energy intake at this position i e 41 9687 gt 22 5417 and so including it in the diet would increase the fish s net rate of energy intake 6 MODELLING CARRYING CAPACITY The drift and foraging tools can be used to investigate the number of fish supported at a certain NREI in the study reach Two approaches are described here one manual and the other more automated The manual description describes an approach more suitable to shorter reaches or reaches where the fish in the dominant position would defend all positions upstream to ensure its drift supply was
25. G MODEL 5 1 Purpose of the model The foraging model estimates the net energy intake of a drift feeding fish from information on water depth water velocity the body size and density of drifting invertebrates the body size of the fish and the temperature of the water It predicts the net rate of energy intake NREI as the gross rate of energy intake GRED less the energy costs of steady swimming to maintain station at the focal point The foraging model is based on the model described by Hughes et al 2003 It takes into account the size and velocity of the prey and the reaction distance of the fish to prey to calculate the capture rate Figures 18 amp 19 Hughes et al estimated the foraging area by predicting water velocity dependent foraging radii over a uniform square computational grid Figure 19 We modified this model substituting the uniform square grid with the StreamTubes representation of cross sections composed of triangles and quadrilaterals c f Figure 2 Water temperature swimming speed i e water velocity at the focal point and fish size are taken into account in the calculation of swimming costs Several options are provided for estimating swimming costs see Appendix 2 Hughes et al 2003 used 3D underwater videography to test the drift foraging model on large brown trout in the Travers River New Zealand They found that the model made good predictions of the foraging area but that it overestimated prey cap
26. June 2012 Appendix 3 46
27. LinesStart 4 crossSec 1l 3 Ncolrow 1 Ncolrow 2 1 Ncolrow 2 Ncolrow 1 Ncolrow 2 1 Ncolrow 1 targetLinesEnd 3 crossSec l 3 Ncolrow 1 Ncolrow 2 1 Ncolrow 2 Ncolrow 1 Ncolrow 2 1 Ncolrow 1 newFileDatal flowDat1l c 1 3 targetLinesStart targetLinesEnd newFileDatal 1 number of x secs 2 MODIFY THIS FILENAME TO MATCH THAT IN YOUR DRIFT BAT openfile file paste streamtubes txt sep w writeLines con openfile text newFileDatal close openfile rm openfile rm xsindata run drift model print Running Drift THE STREAMTUBES FILE IN THIS BATCH FILE SHOULD MATCH THAT SPECIFIED IN THIS SCRIPT For some reason show output has to be true for this to work system drift bat ignore stdout TRUE show output on console TRUE creates xsIn txt for the forage model to work on openfile file xsin txt xsindata readLines openfile close openfile rm openfile Run forage model The forage model used has to be able to read in existing fish positions print Running Forage sforageRun stores the name of the foraging model to be used system command sforageRun ignore stdout TRUE show output on console FALSE creates xsOut txt et al The NREImax file produced by forage exe contains the depleted drift for the next cross section openfile file resultsOutMAXNREI txt r NREIFil
28. Tubes component of River2D so do not just leave it running if you need to convert many study reaches to StreamTubes It is best to close the program and reopen it again after each file is processed 3 6 Querying data at points or cross sections June 2012 10 CAWTHRON The flow data can be queried at cross sections or points using the Query menu Use Query gt Read Query XY File to read in a text file containing the X and Y coordinates of interest in two tab separated columns Then select Query gt Process to display the results of a query of these points The query results window will show the X Y and Z coordinates of your query points the depth d and the flow in the X and Y directions Qx and Qy respectively You can save the results of the query by clicking the Save Results As button in the query results window This function can be helpful for comparing the flow solution s values with observed velocity values 4 INVERTEBRATE DRIFT MODEL 4 1 Purpose of the model The invertebrate drift model uses the spatial flow description provided by the StreamTubes model and data on initial drift concentrations and behavioural characteristics for each size class of each taxon provided by the user to estimate spatially explicit drift density within the reach Figure 10 4 2 Mechanics The model calculates the dispersion of drift between streamtubes in both the vertical and horizontal directions working downstream with f
29. User Guide Version 1 1 h CAWTHRON Flow Related Models for Simulating River Hydraulics Ym g p Foraging Energetics of Drift Feeding Salmonids In association with UF FAIRBANKS University of Alaska US Bureau of Land Management Flow Related Models for Simulating River Hydraulics Invertebrate Drift Transport and Foraging Energetics of Drift Feeding Salmonids User Guide Version 1 1 Lon Kelly Joe Hay Nicholas Hughes Eric Goodwin and John Hayes Cawthron Institute 98 Halifax Street East Private Bag 2 NELSON NEW ZEALAND Phone 64 3 548 2319 Fax 64 3 546 9464 Email info cawthron org nz Information contained in this report may not be used without the prior consent of the client EXECUTIVE SUMMARY This is a guide to a modelling suite that takes a river hydraulic survey determines the invertebrate drift transport in the study reach and predicts the foraging energetics of drift feeding salmonids The R2DStreamTubes tool converts a two dimensional hydraulic model River2D into a StreamTubes representation The river is represented as a bundle of longitudinal streamtubes each of which conveys an equal proportion of the river s total flow The area of each streamtube may change along its length to reflect changes in flow velocity The invertebrate drift transport model calculates the processes of entry bulk transport dispersal and settling of stream invertebra
30. bed following a near bed release s The time near bed parameter in the invertebrate parameters file is included to control the area of the upstream footprint from which invertebrates entering the drift are considered in the calculations at a given cross section This helps alleviate the potential for inflated estimates of drift density which would result if the entry rate No m s were applied to the entire area between cross sections in slow moving water P two_deleatidium_parameters_test Notepad loj x Fie Edit Format View Help farge de tac cfum entryRate 0 0000 settlingRate 0 004 size 0 0075 dryweight 0 00217 timeNearged 9 small_deliatidium entryRate 0 0000 settlingRate 0 001 size 0 0045 dryweight 0 0015 timeNearBed 3 a Figure 11 An invertebrate parameters text file 3 Initial drift densities a text file providing initial drift densities No m for each drift class at the top cross section to be modelled for each streamtube Figure 12 The file contains e the cross section that the drift model is to be initialized at O being the first upstream cross section in the reach e number of drift classes Then the following repeated for each drift class e name of drift class and horizontal and vertical number of streamtubes e a matrix of the initial drift densities Note that this file must include an individual matrix of initial drift concentrations for each class of drift These initial concentrations s
31. box Figure 17 where you can specify multipliers for the degree of dispersion in the horizontal and vertical directions as well as for the settling velocity and entry rate of invertebrates June 2012 18 Set H Params xi Cancel di Standard Dev H Dispersion Multiplier Standard Dev Y Dispersion Multiplier Settling Rate Multiplier Yd Pd z Entry Rate Multiplier Figure 17 The Set H Params dialog box 4 7 Command line control An alternative to the graphical interface is to control the drift model from the command line or batch script This enables the automated repetition of model runs altering input files or control parameters to build a set for comparisons search for optima investigate trends etc Help can be invoked for the drift model from the command line Noting that customised versions of the drift model may have names differing from the default drift exe help is invoked by passing the help flag to the drift exe being used For example drift exe help gives the output Usage drift exe d baseDir o lt output gt hmult lt double gt vmult lt double gt smult lt double gt emult lt double gt help version lt streamFile gt lt paramsFile gt lt concFile gt q lt queryInput gt This program estimates drift concentrations based on three input files specifying flow characteristics of drifting insects and initial concentrations
32. ch Perca flavescens and walleye Stizostedion vitreum vitreum Journal of the Fisheries Research Board of Canada 34 1922 1935 Orth D J 1987 Ecological considerations in the development and application of instream flow habitat models Regulated Rivers Research amp Management 1 171 181 Railsback S F Rose K A 1999 Bioenergetics modelling of stream growth temperature and food consumption effects Transactions of the American Fisheries Society 128 241 256 Rand P S Stewart D J Seelhach P W Jones M L Wedge L R 1993 Modelling steelhead population energetics in lakes Michigan and Ontario Transactions of the American Fisheries Society 122 977 1001 Rao G M 1968 Oxygen consumption of rainbow trout Salmo gairdneri in relation to activity and salinity Canadian journal of zoology 46 781 786 Rutherford J C 1994 River mixing John Wiley Chichester Steffler P Ghanem A Blackburn J 2003 River2D Version 0 90 computer program University of Alberta Canada http www River2D ualberta ca index htm Stalnaker C B Lamb L Henriksen J Bovee K Bartholow J 1995 The instream flow incremental methodology a primer for IFIM National Biological Service Fort Collins Biological Report 29 45 p June 2012 30 CAWTHRON Stark J D Shearer K A Hayes J VV 2002 Are aquatic invertebrate drift densities uniform Implications for salmonid foraging models Verhandlungen Internationale Ver
33. ch description in River2D format is first divided by a series of cross sections which are then subdivided horizontally and vertically into an array of polygons each of June 2012 4 CAWTHRON which is connected with its corresponding polygon in adjacent cross sections to form tubes each assigned an equal proportion of the total flow Figure 2 This StreamTubes representation of stream flow is required for the drift and foraging models Figure 2 Velocity vectors describing a 2D flow solution produced by River2D for a surveyed pool top converted to cross sections centre black and streamtubes centre blue using the StreamTubes model One cross sectional view through the StreamTubes grid is shown at the bottom not to scale The conversion of two dimensional flow data from a River 2D flow solution into a StreamTubes representation is performed interactively within River2D Since the flow information in a River 2D flow solution is depth averaged the StreamTubes model uses a theoretical velocity depth profile to estimate velocities at all depths from the depth averaged input The default profile is the text book logarithmic velocity depth profile with slower velocity near the bed Gordon et al 1992 However a uniform velocity depth profile can also be specified 3 2 Opening and displaying flow data Open R2DStreamTubes exe by double clicking on its icon You will be presented with a graphical display window based on
34. convention of orienting cross sections means a separate version of the Drift model needs to be used for these two sources of StreamTubes grid files June 2012 3 CAWTHRON The StreamTubes representation of the study reach provides flovv information in the format required for input to the Drift and Foraging models 2 2 Invertebrate drift model component The invertebrate drift model uses the flow description in the StreamTubes representation of the study reach to predict three dimensional spatial variation in invertebrate drift density The model is initialized by specifying invertebrate drift densities at the most upstream cell of each streamtube along with relevant physical and behavioural characteristics of the drifting invertebrates such as their settling velocities and rates of entry into the water column The model then simulates the processes of entry into the flow downstream transport turbulent mixing and settling to estimate the concentration of invertebrate drift at each subsequent downstream cell of each streamtube The invertebrate drift model can be used to investigate how changes in stream discharge will influence the spatial distribution of invertebrate drift 2 3 Foraging model component The foraging model a modification of that described in Hughes et al 2003 uses the spatially explicit flow description in the streamtubes representation of the study reach and the spatially explicit invertebrate drift concentrations
35. data 4 0 write NREIFiledata 4 5 openfile for driftwrite in 1 as numeric NREIFiledata 5 cat NREIFiledata 6 driftwrite 1 Ncolrow 2 1 append TRUE file openfile cat n append TRUE file openfile for driftRowwrite in 1 Ncolrow 2 cat rev unlist strsplit NREIFiledata 6 driftwrite 1 Ncolrow 2 1 driftRowwrite t append TRUE file openfile cat n append TRUE file openfile cat n append TRUE file openfile close openfile rm openfile else otherwise use the drift modified concentrations extract the concentrations from xsin txt and write them out in the right format for the drift model line4 unlist strsplit xsindata 4 nvert as numeric line4 2 nhori as numeric line4 1 ndriftline 6 1l nvertt l nvert 2 ndrift as numeric xsindata ndriftline driftnamelines s ndriftline tl ndriftlinetndrift rm ndriftline driftnames xsindata driftnamelines writeLines con openfile 0 writeLines con openfile as character ndrift for dr in l ndrift drift name output outline paste unlist strsplit driftnames dr 1 line4 1 line4 2 n sep t cat outline file openfile append TRUE drift concentrations output drconclines driftnamelines ndrift 2 dr 1 nvert 4 1 driftnamelines ndrift dr 1 nvert 1 nvert 1 for dl in 1l length drconclines June 2012 Appendix 3 45 cross
36. depletion by drift feeding fish However it is possible to model this in combination with the foraging model See section 6 Modelling carrying capacity 4 4 Modelling invertebrate drift Start by opening the drift modelling program DriftModel exe by double clicking on its icon The initial graphic in the display just lets you know the program in functioning Once you open a streamtubes reach representation the display will show the cross sections in plan view and coordinates will display in the status line as the mouse is moved over the display area The input files must be loaded in the correct order moving from left to right across the menu bar Figure 13 Start by reading in the streamtubes file that you created using StreamTubes using the CrossSections gt Read in menu item Next use Drift gt Read in to load an invertebrate parameters text file Finally load the initial drift densities using DriftModel gt Initialize command Untitled XSec Yieve Of XI File Edit view CrossSections Drift DriftModel Test Help Dad e 6 Figure 13 The DriftModel exe menu bar The drift transport model can now be run by selecting DriftModel gt Disperse Note You can load new invertebrate drift data using the CrossSections gt Read in menu item and Drift gt Read in and re run the model using DriftModel gt Disperse without starting over June 2012 15 CAWTHRON CAWTHRON 4 5 Saving the drift
37. downstream of this cross section i e in cross sections 11 and above In order to avoid the outputs of the original foraging model run being overwritten by the outputs of the new model run set up a new folder in which to run the truncated foraging model run This folder has to contain a copy of the foraging model ForagingModel exe a paramsIn txt file and an output from the latest truncated run of the drift model named xsIn txt Alternatively simply rename the output files from the original model run to avoid them being overwritten June 2012 26 x CAWTHRON Note that it is possible to alter the fish and habitat characteristics parameters specified in the paramsIn txt file in subsequent foraging model runs This makes it possible to model the distribution and abundance of a size structured group of fish In this situation the largest fish should be placed first on the assumption that they will be able to defend the most profitable areas By iterating this process it is possible to predict the location of several or all the fish in the reach Remember that a fish will not be able to maintain condition in a position where the predicted NREI is negative So a negative NREI value at the location specified in a resultsOutMaxNREL txt file could be interpreted as indicating that no more fish would be able to make a living in the remaining portion of the reach Other more conservative cut off points in terms of NREI requirements could be
38. e added to the array where we stored the contributions of dispersion Calculations proceed cross section by cross section down the stream A1 4 Model of entry The entry component of the model operates only on the concentrations of the bottom tier of streamtubes those following the bed of the stream Entry is the only component of the model where we deal explicitly with numbers of insects in addition to concentrations The inputs are a rate of entry RE No m s and a time Tab that an insect is allowed to spend near the bed T p allows us to avoid having large numbers of re entering insects overwhelm settling in very slow water and is consistent with Ciborowski s 1983 experimental finding that insects released near the bed spend a constant time in the flow before settling out The number of insects entering the flow between cross sections is modeled as RE Acor Max 1 Ti T The effects of entry are converted to concentrations and accumulated in the array storing the cumulative effects of dispersion and settling June 2012 Appendix 1 36 CAWTHRON CAWTHRON Appendix 2 Swimming cost options in the foraging model Appendix 2 Swimming cost options in the NREI model The foraging model code includes four options for calculating swimming costs The models and their parameters and their origins are presented in the following sections e A polynomial model fitted to Brett and Glass s 1973 swimming cost data for sockeye salmon
39. ed is within defence radius of any other fish then it will not be proposed as the best location on that cross section and another will be evaluated until one is found outside the defence distance of any other fish This of course requires the positions of previously placed fish to be available to the foraging model which is ensured by the script writing these positions to a FishPosns txt file that is then read in by the foraging model 7 ACKNOWLEDGEMENTS The R2DStreamTubes software is built on the River 2D Bed code generously provided by the River 2D group Steffler P Ghanem A Blackburn J 2003 University of Alberta Canada The foraging model software relies on the Generic Polygon Clipping code written by Alan Murta for area calculations and hit testing The GPC software is Copyright 1997 1999 Advanced Interfaces Group Department of Computer Science University of June 2012 28 CAWTHRON CAWTHRON Manchester This software is free for non commercial use It may be copied modified and redistributed provided that the copyright notices which appear within the library source files are preserved on all copies The intellectual property rights of the algorithms used reside with the University of Manchester Advanced Interfaces Group The progress bar displayed during the drift model run is a freely available VC object from CodeProject with no restrictions on use The Matlab programs for displaying model output
40. edata readLines openfile close openfile rm openfile NREI max val NREI_Val crossSec as double substr NREIFiledata 3 11 1000 and the best fish position Note that the NREI threshold for placement is only managed in this file not in the executable so a fish position will be reported even if it s a negative or too low NREI this script then has to accept or reject that suggestion based fon the NREI value at that position relative to the threshold if NREI_Val crossSec gt NREI_Limit fFishCount fishCount 1 fishpos fishCount c as double strsplit NREIFiledata 1 1 1 3 as double strsplit NREIFiledata 1 1 4 June 2012 Appendix 3 44 CAWTHRON write fishCount FishPosns txt for fishwrite in 1 fishCount write paste sep fishpos fishwrite 1 fishpos fish write 2 append TRUE FishPosns txt make a numbered backup of initialConc drift file file copy initialConc txt paste initialConc as character crossSec txt sep Write depleted concentration to initialConc for the next drift run note that the grid may be reversed if working from Rhyhabsim whereas the output concs never are so we reverse them here if necessary file remove initialConc txt openfile file initialconc txt open w if a fish is placed use the depleted concentrations for the next drift run if NREI_Val crossSec gt NREI_Limit NREIFile
41. einigung fur Theoretische und Angewandte Limnologie 28 988 991 Stewart D J 1980 Salmonid predators and their forage base in Lake Michigan a bioenergetics modelling synthesis PhD thesis University of Wisconsin Madison Stewart D J Weininger D Rottiers D V Edsall T A 1983 An energetics model for lake trout Salvelinus namaycush application to the Lake Michigan population Canadian Journal of Fisheries and Aquatic Sciences 40 681 698 Videler J J 1993 Fish swimming Chapman and Hall London U K Wankowski J W J 1979 Morphological limitations prey size selectivity and growth response of juvenile Atlantic salmon Salmo salar Journal of Fish Biology 14 89 100 June 2012 31 CAWTHRON Appendix 1 Drift transport model description Appendix 1 32 Appendix 1 Drift transport model description The invertebrate drift model calculates the settling entry vertical dispersion and lateral dispersion of drifting invertebrates in the study area estimating their spatially explicit densities It builds on insights provided by Ciborowski 1987 The model is based on the eulerian frame of reference provided by the equal discharge streamtubes and the cross sections that divide these tubes into cells The model takes as input the densities of invertebrates in each streamtube cell at the upstream cross section and estimates the concentrations for each of the cells in the downstream cross sections The model as impleme
42. ely The X Sec gt Save endpoints and X Sec gt Read endpoints selections allow for a cross section or series of cross sections to be saved and subsequently read back into River2D June 2012 7 CAWTHRON CCF2P1 BCO1 cdg R2DStreamTubes lel File Edit View Display Bed X Sec StreamTubes Prey Query alia 2e alti Ei sella aiz ell id x 476052 598886 y 5125869 906085 z 581 756094 d 0 716360 Qx 0 000000 Num Figure 4 R2DStreamTubes exe interface window with boundary and cross sections displayed 3 4 Creating streamtubes The next step is to divide the cross sections into a number of streamtubes each conveying equal discharge This is done using the FlowTubes gt Draw menu item At this stage a dialog box asks you to specify the number of streamtubes you require across the stream width Figure 5 These streamtube divisions are subsequently drawn into the display window Figure 6 The number of streamtube divisions required in the vertical axis is specified later along with the velocity depth profile during the process of saving the StreamTubes output Dialog Number of Tubes Right to Left Cancel June 2012 8 Figure 5 Dialog box for specifying the number of horizontal streamtube divisions The nodes defining the horizontal division of the flovv can be edited interactively to eliminate unrealistic acute angles from the streamtubes e g see Figure 6 by selecting FlowTubes
43. es including their appropriate file extensions ssssssssrsssssssssessessessecessessesessesscssesesscsssssssssassesessesseseesess 2 Figure 2 Velocity vectors describing a 2D flow solution produced by River2D for a surveyed pool top converted to cross sections centre black and streamtubes centre blue using the StreamTubes model One cross sectional view through the StreamTubes grid is shown at the bottom not to scale 5 Figure 3 R2DStreamTubes exe interface window with boundary displayed and first cross section drawn 6 Figure 4 R2DStreamTubes exe interface window with boundary and cross sections displayed 8 Figure 5 Dialog box for specifying the number of horizontal streamtube divisions ssessesscesssssssesseseeees 9 Figure 6 R2DStreamTubes exe interface window with boundary cross sections and horizontal Streamtubes displayed in plan VicW ssscsscsssssssssssscssssssssscsssssssssesssssessssessessesssssssassssessessessssessassessesessessesesee 9 Figure 7 Dialog box with prompt for number of vertical streamtubes and tick box to specify constant velocity rather than default logarithmic velocity profile With depth ssscsssssssssssssssssssssssscsssssssssssssesesseee 10 Figure 8 Example of a StreamTubes output file See text for explanation Of values sesserssosssoeseosseoesesese 10 Figure 9 The lateral exchange of drift between streamtube cells in a cross section based on a series of Gaussian di
44. graphics the Portable Network Graphic png file type is recommended for best visual results A3 2 Displaying outputs using River 2D It is possible to use the triangulation and contouring functionality of the freely available River 2D software to display the output from the foraging model This involves using the outputs of the foraging model to create a pseudo bed file which than can then be displayed in River 2D The first step is to copy the X and Y values from a resultsOutNREL txt file into a new file Add the X and Y coordinates of the model boundaries captured using Save Endpoints As under the X Sec menu in R2DStreamTubes to the end of these columns This file can now be used to extract bed elevation values for the series of points using Read Query XY File under Query in R2DStreamTubes Next add a column before the X and Y columns containing a series of arbitrary node numbers running from 1 to n Then add another column after all the others containing the NREI values for the nodes from the original resultsOutNREI txt file Give the boundary nodes an NREI value of zero Finish the file off by typing no more nodes at the end Save it and change its file extension from txt to bed This file is now a pseudo bed file with NREI values in the column were Bed Roughness values would normally be It is ready for display in River 2D Open the flow solution used as the basis for this modelling in River 2D D
45. gt EditNode and then selecting and dragging a node with the mouse This should not be undertaken without careful consideration as moving a node can cause unrealistically high or low velocities by changing the cross sectional areas of streamtubes Trying to edit nodes when there are no streamtubes will generally crash the program CCF2P1_BCO1 cdg R2DStreamTubes loj x File Edit View Display Bed X Sec StreamTubes Prey Query Help Blot e Sele al eola JA al eleli x 476054 327469 y 5125870 955193 Figure 6 R2DStreamTubes exe interface window with boundary cross sections and horizontal streamtubes displayed in plan view 3 5 Subdividing vertically and saving streamtubes Select X Sec gt Save All as to save your StreamTubes in a format suitable for input to the drift model At this point a dialog box will prompt you for the number of vertical streamtube divisions required Figure 7 It also allows you to check a box if you want to use a constant velocity depth profile no change in velocity between the surface and the stream bed to extrapolate velocities through the water column Otherwise a logarithmic velocity profile Gordon et al 1992 with slower flow near the stream bed will be used as the default June 2012 9 ra CAWTHRON Enter Number of Yertical Tubes xi Cancel di Number of tubes bottom to top 5 Velocity is constant with depth a Figure 7 Dialog box with prompt for
46. hould preferably be derived from the results of drift sampling in the field June 2012 14 P multi driftclass initial conc Notepad File Edit Format Viem Help p 4 Deleatidium6_5mm 10 10 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 0 178 Deleatidium7_Smm 10 10 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 0 179 N 170 N 170 NAF NA7A N 170 NAV NA7 NAV NAV n 170 Figure 12 A drift density initialization text file to initialise the first cross section with concentrations for 4 drift types Note This program does not model drift
47. i cccccesvessscooscosessvcseseasovecsassnesanssseacsncsecssnsnsoeseseosssssvsiessenssssevitsbseses 6 MODELLING CARRYING CAPACITY 6 1 Manual approacieiieicgams tires sie ies time rosa tio et iaia anna sddsagstseresssebtyavivynesed 6 2 Automated approach sismes isesssviecoicscccssdtvnivesecs cosbsveacaedscossvasassevssedecoteasagetoveeecscanssnysvses 6 3 Defence radius ciseseisisccssters tdsees coesasaneves nit EnaA EE E AEE SEE OEFENE E ARNEE Eas 7 ACKNOWLEDGEMENTS scccsssssssssssssssssssscsssssssssssssssecsesessesscsesssssasssssssessessssessasssssssessesssoes 28 8 DISCLAIMER csssssssssssssssccsssssscsssssssscscsessssscsessssscssessssessesssesscsssssssassssossessesssssssassessssessesesoes 29 9 REFERENCES sessseseseseseeessresorcsceeoeseoroeeeeesesesesesoseeeeesesesoeoeoeerosereeoeorersreeeesesesesesoeesesesesesoseeenenonese 29 June 2012 CAWTHRON LIST OF TABLES Table 1 Parameters for the two swimming speed equations for steelhead rainbow trout from Fish Bioenergetics 3 0 Hanson et al 1997 s scccsccssssssssssscsscsscssssssssscssssssscsssssesssssssssescsssssssessssscssessssssessossosseess 39 Table 2 Parameters for the brown trout swimming cost equation from Hayes et al 2000 a b b were sourced from Elliott 1976 and b was sourced from Rand et al 1993 and Stewart 1980 cited from Rao 1968 39 LIST OF FIGURES Figure 1 Flow chart of modelling process Model names as well as their input and output fil
48. ile generated by CrossSections gt ConcentrationsDump from the drift model 2 CrossSections gt Dump for Foraging Model outputs a non annotated text file Figure 15 which is suitable for input to the foraging model The file contains the following the cross section at which the drift model run was initialized the total number of cross sections Then the following is repeated for each cross section the horizontal and vertical number of streamtubes X and Y coordinates of the first point on the cross section in plan view and the X Y of the last point on the cross section distances to consecutive horizontal streamtube divisions across the cross section a matrix of depth coordinates of the vertical divisions at each horizontal division a matrix of streamtube water velocities the number of drift classes and the name length and weight for each class of drifting invertebrates from the invertebrate parameters file For each invertebrate drift class a matrix of the density in each polygon in that cross section June 2012 16 CAWTHRON aixi File Edit Format View Help 010 1512 53 1473 64 1504 71 1485 38 0 1 64469 2 46566 3 19512 3 87187 4 57912 5 37234 6 4762 7 66603 9 23861 14 1061 0000000000 0 0263696 0 0346481 0 0420414 0 0411084 0 0414731 0 03793 0 0284187 0 0281482 C 0 0648438 0 0838683 0 10076 0 0986327 Q 0994645 0 0913765 0 0695656 0 0687537 C 0 104185 0 1341
49. isplay colour contours of Bed Roughness clipped to the waters edge Now use Mesh Edit gt Load Bed File to load your newly created bed file Toggle off the Bed Contours under the Display menu River 2D will now be displaying a colour contoured image of NREI values through the modelled reach Note The water boundaries may not display particularly accurately using this method June 2012 Appendix 3 42 CA ra ra CAW HRON Appendix 4 Control script This is the core of an R script that takes a line at a time from a StreamTubes output file and passes it the drift model then the outputs of that to the foraging model It examines the output of the foraging model and places a new fish if the highest NREI in the cross section just modelled is higher than the threshold set at the start of this script It then runs the drift model again either with the drift dispersed densities from the drift model of the previous cross section or the foraging depleted densities from the foraging model This script is provided as an inspiration only and will require customisation to the user s specific requirements NREI_Limit 1000 Place a fish at spots greater than this fishpos data frame NA NA This will store the fish positions names fishpos c x y fishCount 0 sforageRun ForageOnexs Specifies the foraging model to be used MaxNRETout resultsOutMaxNREI set the first initial conc file if file exists initia
50. lConc txt file remove initialConc txt a file exists streamtubes txt file remove streamtubes txt file exists fishPosns txt file remove fishPosns txt Fh Fh H a the top reach uses the first initial concentration he next reaches carry the altered concentration forward secn 1 file copy initialConcO txt initialConc txt dE rh C E if secn 2 file copy paste workingDir sectionDir 1 lowDir flown initialconc txt sep paste workingDir sectionDir 2 flowDir flown initialconc txt sep if secn 3 file copy paste workingDir sectionDir 2 initialconc txt sep paste workingDir flowDir flown initialconc txt sep f lowDir flown sectionDir 3 icstore file icstore txt wr velstore file velstore txt wr get the streamtubes data openfile file STREAMTUBES r set filename her flowDatl readLines openfile close openfile rm openfile nCrossSecs as double strsplit flowDat1 1l Ncolrow c as double strsplit flowDat1 2 as double strsplit flowDat1 3 1 lin 41 NREI_Val vector double nCrossSecs NREIstore matrix data 0 nrow 0 ncol 3 June 2012 Appendix 3 43 CAWTHRON one cross section at a time crossSec 1 while crossSec lt nCrossSecs grab a single cross section from the streamtubes txt data and write it out for drift bat target
51. language R and Matlab This allows multiple modelling runs to be automated with a set of configuration data passed to the model sequentially one after the other and the set of results archived systematically 2 1 StreamTubes model The StreamTubes model generates a mathematical grid representation of the study reach The irregular 3D grid of a StreamTubes representation provides a convenient way of describing flow through the complex topography of a modelled stream reach A stacked array of streamtubes each divided lengthwise into many cells allows spatial variation in water velocities to be described in X Y and Z directions The streamtube concept simplifies calculations relating to water flow and hence the downstream transport of invertebrate drift and subsequent prey capture by drift feeding fish The flow of water volume per time down a streamtube is conserved along its length from cell to cell while the cross section area of the streamtube and hence velocity of its flow may vary Variation in cross section area provides an indication of variation in water velocity along the tube Our R2DStreamTubes program is designed to construct a StreamTubes representation of flow from a two dimensional depth averaged flow output as generated by River 2D Steffler et al 2003 The River 2D surface is divided into a user defined number of cross sections Spanning the river each of these will have equal discharge through it As these will cross
52. low from cross section to cross section Figures 9 amp 10 See Appendix 1 for a more detailed description of the drift transport model including supporting equations Modelling of invertebrate drift dispersion is based on Rutherford s 1994 river mixing equations and estimates of taxon and size specific entry rates No m s and settling velocities m s Rutherford s turbulent mixing equations are rewritten as variations on the equation describing the Gaussian distribution where the standard deviation of lateral or vertical dispersion is predicted from water depth water velocity bed roughness and the distance between the cross sections By integrating this equation twice we derived an equation to predict the drift contribution that a given streamtube makes to each adjacent streamtube After calculating the effects of dispersion on drift density between two adjacent cross sections the model estimates the effects of settling and entry on invertebrate drift concentration This invertebrate drift transport model allows a significant advance in the way that spatial variation in invertebrate drift density can be dealt with in bioenergetically oriented models of stream fish habitat Until now these models have generally assumed uniform drift density even though it is known that this assumption is unrealistic Stark ef al 2002 Hughes et al unpublished manuscript There simply has not been an adequate way to predict spatial variability of
53. ment y Integrating this function yields a survival distribution which can be used to calculate the probability June 2012 Appendix 1 34 CAWTHRON that a drifting invertebrate would disperse farther from its initial location than displacement y S x y P x y dy y Integrating again we find the probability that an invertebrate randomly placed in a cell will disperse farther than lateral coordinate y 1 n en 120 1 M1 Sea XY 0 5 dy dy 0 5 S x y dy a ar ar In this equation yl and yO are respectively the maximum and minimum distances from the cell to lateral coordinate y as shown here eY _ gt Cell 0 eyo Cell 1 _tI_f_ ___ gt Lateral axis This equation is the basis of the model Using it we can predict the concentrations that disperse from a given cell across the stream and into and beyond laterally adjacent cells as the water moves between cross sections If S j x y is the proportion of the drift that disperses from cello at least as far as cell and S x y is the proportion that disperses past cell then Soon x y Seen X Y is the proportion of the drift in cello that ends up in cell at distance x downstream This computation is made for each pair of cells laterally and moving from cross section to cross section downstream The equation assumes unbounded flow and we account for the fact that the flow is not unbounded by using the mi
54. model output There are several potential outputs from the drift model The first two ways of extracting drift data are most frequently useful 1 CrossSections gt Concentrations Dump creates a lightly annotated text file Figure 14 For each cross section by drift class combination there are some meta data cross section number name of drift class number of tubes vertically and horizontally and the address in computer memory where the data structure was stored followed by a tab delimited matrix of predicted drift concentrations No m for each polygon of each cross section c Concs_dump Notepad 5 x File Edit Format View Help Kross section Targe deliatidium 5 10 address 3303296 Cross section 1 large deliatidium address 3303392 Cross section 2 address 3303488 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 0 4 5 10 0 37233 0 382501 0 384977 0 385443 0 385075 0 383666 0 399998 0 399991 0 399997 0 4 0 399999 0 399997 0 399999 0 399998 0 399999 0 4 0 399999 0 399998 0 399999 0 399997 0 399999 0 4 0 399999 0 399996 0 399995 0 399981 0 399993 0 399999 0 399998 0 39999 large deliatidium 5 10 0 355924 0 371639 0 375233 0 375742 0 375001 0 372552 0 390977 0 394612 0 395717 0 396009 0 395906 0 395343 Figure 14 An output f
55. mtubes sasira nn EE Ea EEE EKRENE EEE LENN EESE EEE EE 8 3 5 Subdividing vertically and saving streamtubes s ssssssssesesesesessstsesiseserssssenersresesesenesrseses 9 3 6 Querying data at points Or CrOSS SECtioOns s sseeeeeeseeeeeieesteieststeisststeststeteststertntsrisinreeretssre 10 4 INVERTEBRATE DRIFT MODEL cccssssssssssssssssssssssssessesscssesssscssessssesscssssssessessssessesseses 11 4 1 Purpose of the model serret cc scescaussoueecdacnagssaesecetecipeencsneh cu gcvaasoacasnses eodacuagutevieeysteagiy tastes 11 4 2 MechanicS se ssi icsvennosvesvisniey Senses eoa E EE AEREA NE A ENRE TARER 11 4 3 Drift model data requirementS sson ai an Er E tat ENEA tteme ssts 13 4 4 Modelling invert brate drifts icona kirsin nn resans E arsana ingere EnaA test met i 15 4 5 Saving the drift model output eesriie eeni aN nankor aana EAE F E EE Naana 16 4 6 Altering the model parameters s s ssssesssssssstssssssstsesesestsestrertrtrereresessrenestseesesrsesisenenesrnenes 18 4 7 Command line CONteOlccsessscieeitcsisacsnesessesteasssesassnesusvesasesevascendsexssesuaaess cossponadestvussisst sassienes cee 19 5 SALMONID FORAGING MODEL ccccscssssscsssscsssssscssssessssessssecsccssssssessssessessessesessessesseees 20 5 1 Purpose of the Model sc csc tsdssssssssessscebessvcccsovagwesncoess nan Esi N E E raaa EEEN EEO EA aE AEE Sa 5 2 Modeling NRELs iiiinin iiecii 6mHhnmmnmenmenimenemorminynmsns sen sits 5 3 Foraging model OUtpUts
56. nd then declining as a function of temperature with swimming cost ACTIVITY as an exponential multiplier This is a departure from Equation 2 in Bioenergetics 3 0 which used a constant activity multiplier ACT The exponential activity multiplier that we use is taken from Equation 1 fT V z e Xray ACTIVITY e108 where V RTM T RTM RTO X Z 1 1 40 Y 400 Z LN RQ RTM RTO June 2012 Appendix 2 38 CAWTHRON Y LN RO RTM RTO 2 RO approximates the Qio the rate at which the function increases over relatively low water temperatures RTO C is the optimum temperature for respiration where respiration is highest RTM C is the maximum lethal water temperature Table 1 Parameters for the two swimming speed equations for steelhead rainbow trout from Fish Bioenergetics 3 0 Hanson et al 1997 Parameter Value RA 0 00264 RB 0 217 RQ 0 06818 RQ Da RT 0 0234 RTO 22 RTM 26 Parameters sourced from Railsback amp Rose 1999 In our opinion the value for RTO 22 from this source may be unrealistically high for general application VVe suspect that it may be most appropriate for warm water acclimated rainbow trout populations on an energy rich diet However it may overestimate the optimal temperature for growth for colder water adapted populations especially when trout are feeding on invertebrates rather than a fish diet In our opinion further research
57. ne 2012 CA EO RON Figure 23 Example of the epsOut eps output from the foraging model sesesesesesesesesesosesososesosescsoeseosseereseese 24 Note that changes in the eps file format in different versions of the Windows operating system means that as of 2012 this output does not render correctly csccscssssssssessesscssccessesscssssessessesesssssessssessassssessesseseees 24 Figure 24 A resultsOutForaging txt output of the foraging MOdel sscsscsrssseccsrssssesscsssssssessessesessesseress 25 LIST OF APPENDICES Appendix 1 Drift transport model description scssssscsssssssecescsscessnsscssesssssesesscssesssssssseseseeees 33 Al 1 Values derived from the streamtube description 22033 A1 2 Model of dispersion ssscsscssssssssscsscssssssssssssssssscssesescssssssscssesesssnsesesssssessssssessoesseeees 33 A1 3 Model of Setbline vcccsissessecsessossvonscacnssensconeseseccssssosnnsseasonssasontccnesons ocoocensvausesonccansctsesnsascesesen 35 A1 4 Model of entry sscscsscsssssesssssssessssessesees 22038 Appendix 2 Swimming cost options in the NREI model s537 A2 1 Polynomial swimming cost model coho salmon 37 A2 2 Rainbow trout swimming cost models 2 m222222222222220220220222002022022000202022002000020200202000022 37 A2 3 Brown trout swimming cost mOdel cscessessccscscsessssssscssesssssessesssssessssssesseesserees 39 A3 1 Matlab programs for displaying model outpu
58. net energy intake and reach carrying capacity for drift feeding salmonids Ecological Modelling 207 171 188 Hayes J W Stark J D Shearer K A 2000 Development and test of a whole lifetime foraging and bioenergetics model for drift feeding brown trout Transactions of the American Fisheries Society 129 315 332 Hughes N F 1992 Selection of positions by drift feeding salmonids in dominance hierarchies model and test for Arctic grayling Thymallus arcticus in subarctic mountain streams interior Alaska Canadian Journal of Fisheries and Aquatic Sciences 49 1999 2008 Hughes N F Hayes J W Shearer K A Young R G 2003 Testing a model of drift feeding using 3 D videography of wild brown trout in a New Zealand river Canadian Journal of Fisheries and Aquatic Sciences 60 1462 1476 Hughes N F Hayes J W Shearer K A Stark J D Spatial variation of invertebrate drift density influence of stream discharge time of day taxon and body size in a pool on a New Zealand river Unpublished manuscript Hughes N F Kelly L H 1996 A hydrodynamic model for estimating the engergetic cost of swimming maneuvers from a description of their geometry and dynamics Canadian Journal of Fisheries and Aquatic Sciences 53 2484 4293 Jowett I G 1999 Rhyhabsim River Hydraulics and Habitat Simulation Version 3 30 computer program Kitchell J F Stewart D J Weininger D 1977 Applications of a bioenergetics model to yellow per
59. ng radials and predicted prey capture area interpolated from predicted foraging radius along each foraging radial based on prey reaction distance and water velocity Hughes et al 2003 5 2 Modelling NREI NREI modelling is undertaken by the command line program ForagingModel exe This model requires two input files June 2012 21 CAWTHRON 1 A cross section file which is the output from the drift model Figure15 This must be named xsIn txt 2 Fish characteristics a text file named paramsIn txt similar to the invertebrate parameters input to the drift model Figure 20 This file must contain in order e fork length of the fish m e fish weight g e minimum prey size m Hayes et al 2000 and Hughes et al 2003 estimated this as 1 15 of the fish length based on gill raker spacing after Wankowski 1979 e maximum prey size m typically based on the gape size 45 2 90 of the fish length e g Hayes et al 2000 and Hughes et al 2003 after Wankowski 1979 e probability of prey detection allows correction for overestimation of prey capture rate in the model height of focal point above bottom m number of interrogations per cross section water temperature C defense radius m paramsIn txt Notepad s Ioj xi Edit Format View Help 0 5 2500 0 00575 0 226 0 55 0 1 36 16 0 4 Figure 20 A paramsIn txt file specifying fish and habitat parameters 2 B Both of these file
60. nted provides for concurrent processing of an arbitrary number of drift categories which can vary in initial densities settling rates time near bed and entry rates The illustrations in the following explanation of the model are limited for clarity to processing a single drift category A1 1 Values derived from the streamtube description The use of equal discharge streamtubes provides a concise geometric description of water flow through the survey reach The polygons on a given cross section are either triangles or quadrilaterals The following quantities are calculated from the 3D description of the streamtubes and the known per cell discharge Q e area Acen water velocity between cells V 0 5 Q Acen Q Acet 1 s centroid Ceen distance to corresponding cell upstream Dx ICeen Ceen 1 time for water to move between cells T D V vertical height of the cell at the centroid Heen area of streambed forming the floor of the streamtube cell Aco i e between cross sections A1 2 Model of dispersion The dispersion component of the model is similar in the lateral y and vertical z directions The form of each analogous equation is the same and aside from the coordinate axis the only thing that changes is the value for the dispersion coefficient k and k respectively so we limit this discussion to lateral dispersion Rutherford 1994 gives an analytic solution for the mixing equation for constan
61. of drifting insects d directory baseDir base directory for input files o lt output gt output file default is hmult lt double gt std dev horizontal multiplier vmult lt double gt std dev vertical multiplier smult lt double gt settling rate multiplier emult lt double gt entry rate multiplier help print this help and exit version print version information and exit lt streamFile gt lt stream tube Input file gt lt paramsFile gt lt drift param Input file gt June 2012 19 CAWTHRON lt concFile gt lt initial concentrations Input file gt q lt queryInput gt query input file This help output details how all the customisations that can be specified through the graphical interface described above are specified via the command line A command is built up from the name of the executable appended with a number of flags some optional to specify input and output files and parameterisation such as the multipliers drift exe d won 0 xsIn txt flowQ10 txt invertParams txt initialConc txt For iterative execution as implemented in a script R Matlab etc a valid and complete command line call is saved as a batch file and that batch file then called from the script The script is responsible for the file management that ensures the correct input files are in the directory the model is run in and the output files are moved away for archive and further analysis 5 SALMONID FORAGIN
62. programs e or using the triangulation and contouring functions in River 2D The Matlab and River 2D options are discussed here A3 1 Matlab programs for displaying model outputs The following two purpose built matlab programmes allow drift and foraging model output to be visually displayed flowtubeplot m plotfish m Flowtubeplot m produces a series of images depicting both the drift concentrations and NREI values in the modelled reach as colour contour plots This program produces both plan view and cross sectional images with drift concentrations broken down by drift class and amalgamated Plotfish m allows predicted fish positions to be plotted onto the NREI output from a multi fish simulation A3 1 1 Using the Matlab programs These programs require four input files to be located in the same directory 1 A boundary file a text file defining the boundaries of the modelled reach made using Save Endpoints As under the X Sec menu in R2DStreamTubes named Boundary txt 2 A StreamTubes output file as used as input to the drift model named StreamTubes txt 3 A drift model output file as used as input to the foraging model named xsIn txt 4 An NREI output file named resultsOut txt The command flowtubeplot in Matlab will produce a graphic showing a plan view of the cross sections and streamtubes produced in R2DStreamTubes Double clicking on any of the cross sections will produce a series of graphical out
63. puts including both cross sectional for the cross section selected and depth averaged plan views for the entire reach of 1 the modelled NREI values 2 the modelled drift densities for a each drift class individually No m b the overall drift density summed over all drift classes No m c the total drift biomass ug m Returning to the original cross section and streamtube plot and double clicking on a new cross section will change the cross sectional graphics to the new cross section June 2012 Appendix 3 41 CAWTHRON Clicking outside of the modelled area will exit the program Note The plan view NREI output also shows the total NREI available from the entire reach This is derived by summing all the positive area weighted NREI values Entering the command plotfish in Matlab will plot the fish locations obtained from a multi fish simulation over the active plan view NREI graphic To do this additional input files are required One file must contain the X and Y locations as returned in the resultsOutNREIMax txt files for the multi fish model runs These data make up the first two columns of a tab delimited text file called locations txt A third column should contain the lengths of each fish m A second file called newfish txt which describes the shape of the fish icons used in the plot is also required These graphics can be exported in a format that can be inserted into other documents etc When exporting
64. r reach River_2D flow solution cdg RHYHABSIM or exe exe R2DSTREAMTUBES Streamtubes txt Drift density initialization txt DRIFT Invertebrate parameters txt exe Drift output txt Fish characteristics txt FORAGE exe NREI output txt NREI max txt Feeding Effeciency txt Dietary composition txt Swimming costs txt Figure 1 Flow chart of modelling process Model names as well as their input and output files including their appropriate file extensions 2 OVERVIEW OF MODELS The three model components StreamTubes drift transport and foraging are implemented in software for use by technically minded workers They are implemented June 2012 2 as VVindovvs programs They are run sequentially and produce considerable amounts of intermediate output The drift and foraging models have command line variants The extensive and variably formatted output in combination with the ability to work through a modelling investigation step by step slows the modelling process but on the other hand these features allow users to ensure that the programs are operating sensibly and producing outputs consistent with the inputs and to more easily troubleshoot and customize the software by editing and recompiling or the data processing pipeline Furthermore once correct operation has been determined it can be managed by script run in any language that can interact with the operating system including the Windows batch
65. respiration function is June 2012 Appendix 2 37 CAWTHRON R RA W 4 f T ACTIVITY 213565 86400 where R is energy cost Joules s The original units in Fish Bioenergetics 3 0 are g O2 g Maes specific rate of respiration In the above equation a coefficient of 13565 J g Oz is used to convert to energy units Joules s Elliott and Davidson 1975 and J d is converted to J s by dividing by 86400 W is fish mass g RA is the intercept of the allometric mass function i e specific weight of oxygen g O2 g d consumed by a 1 g fish at 0 C and zero swimming speed g g d RB is the slope of the allometric mass function for standard metabolism g g d f T is the temperature dependence function T is water temperature C ACTIVITY is an activity multiplier Fish Bioenergetics 3 0 gives two equations for calculating f T and ACTIVITY Equation 1 Exponential with swimming speed Stewart et al 1983 This equation predicts a continuous exponential increase in respiration costs as a function of temperature and swimming speed FT efn ACTIVITY eT where VEL swimming speed m s RQ approximates the Qio the rate at which the function increases over relatively low water temperatures RT is the coefficient for swimming speed dependence on metabolism s cm Equation 2 Temperature dependent with activity multiplier Kitchell et al 1977 This equation predicts respiration increasing to a maximum a
66. rroring method described by Rutherford where we reflect dispersion from the banks The equation also assumes constant flow and our model accounts for the fact that velocities vary between cells as follows First the effect of dispersion is calculated for each cell individually to predict the contribution that each cell would make to its neighbours if conditions in the neighbouring cells were the same as in the originating cell Second the dispersions predicted for each pair of cells are averaged The result is the amount of interchange predicted to occur between each pair of neighbouring cells given their different values of o and velocity The contributions of each cell to the concentrations of cells to the left and right including mirror cells are accumulated in an array and a similar process is followed for vertical dispersion A1 3 Model of settling Settling is considerably simpler than dispersion as it is modeled as a linear time dependent process The settling component of the drift model begins with knowledge of the drift concentrations at the cells upstream and the settling velocity V for the drift category The proportion of the concentration in the cell immediately upstream that settles into the next cell below or in the case of the bottom most cell the proportion removed from the water is taken as the minimum of 1 0 or Heen T V2 The June 2012 Appendix 1 35 contributions from and to each cell from settling ar
67. rst fish is placed at the location with the highest NREI in the whole reach and subsequent fish are placed only downstream of existing fish This can leave large empty areas of profitable habitat upstream of the dominant fish To allow such territory to be occupied requires the change that is made possible by the automated approach based on drift modelling and then forage modelling just one cross section at a time from the top to the bottom of the modelled reach The external script takes on responsibilities for the management of the input and output files and involvement in the fish placement decision It passes two cross sections of topography and flow data to the drift model at a time and initialises it with appropriate invertebrate density data The script picks up the outputs from the drift model and hands them to the forage model inspects the output from the forage model to determine whether a fish will be placed in that cross section before handing the next cross section pair to the drift model At each stage a pair of cross sections is passed to the drift model one as the source and one as the destination for drift dispersion June 2012 27 CAWTHRON A branch in the script is controlled by vvhether the foraging model reports NREI high enough to place a fish in the cross section just processed The decision to place a fish is implemented in the script to allow for easy control over the NREI threshold chosen for fish placement If a
68. s must be appropriately named and located in the same directory folder as a copy of the foraging model ForagingModel exe Double clicking the foraging model icon will open the command line window and automatically start the model running The program will automatically create several output files in the same folder in which it has been activated Alternatively the executable program can be launched from a command line window where the directory has been changed to the folder containing the input files of interest The output files will then be written to the same directory Alternatively the program may be launched by a script running in an environment such as R or Matlab etc 5 3 Foraging model outputs The output named xsOut txt is simply a duplicate of the input cross section file xsIn txt and is solely to verify that the data were properly read in This is useful given the inconsistency between different streamtube generating programs as to which bank a cross section is initiated on Aside from this there are six output files June 2012 22 CAWTHRON 1 resultsOutNRELI txt lists the X and Y coordinates and the NREI output for each sampled position in a tab delimited text file Figure 21 P resultsOutNREI Notepad lol x Fie Edit Format View Help forkLength forkLength 1511 82 1474 0 5 0 5 71 1 70225 1511 11 1475 77 1 33865 1510 4 1476 84 2 10833 1509 69 1477 1508 98 1478 1508 26 1480
69. s sectional view of the best foraging location It shows the water surface horizontal and vertical streamtube divisions and foraging area radials EERE m TES HL LA i an Figure 23 Example of the epsOut eps output from the foraging model Note that changes in the eps file format in different versions of the Windows operating system means that as of 2012 this output does not render correctly 5 resultsOutForaging txt provides information on the model s predictions of the size composition of the diet prey choice etc Figure 24 For every focal point by drift class combination it contains the X and Y coordinates of the fish s focal point the drift class index the total energy content of the drift class J the fish s encounter rate with the given drift class per second of search No s the capture rate of the drift class per second of foraging No s the handling interception time s for the drift class this is reaction distance maximum swimming velocity e the profitability J s of drift class 0 7 total prey energy handling time N B 0 7 conversion factor after Elliott 1982 e the fish s gross energy intake at this position J s June 2012 24 e the in diet index 1 if the fish should eat the prey 0 if not determined by the minimum and maximum prey size or by Charnov s 1976 model P resultsOutForaging Notepad File Edit
70. s were written by Ben Tuckey and Ben Knight Cawthron Institute The automated fish placement mechanism was developed with partial support from Eco Logical Research Incorporated Utah 8 DISCLAIMER Any software available is provided as is and no warranty express or implied is made of its suitability for any particular purpose 9 REFERENCES Addley R C 1993 A mechanistic approach to modelling habitat needs of drift feeding salmonids M Sc thesis Utah Sate University Logan Annear T and 15 other authors 2002 Instream flows for riverine resource stewardship Instream Flow Council US Brett J R Glass N R 1973 Metabolic rates and critical swimming speeds of sockeye salmon Oncorhynchus nerka in relation to size and temperature Journal of the Fisheries Research Board of Canada 30 379 387 Bovee K D 1982 A guide to stream habitat analysis using the instream flow incremental methodology U S Fish and Wildlife Service Biological Services Program FWS OBS 82 26 Instream flow information paper 12 248 p Charnov E L 1976 Optimal foraging attack strategy of a mantid American Naturalist 110 141 151 Ciborowski J J H 1983 Downstream and lateral transport of nymphs of two mayfly species Empheroptera Canadian Journal of Fisheries and Aquatic Sciences 40 2025 2029 Ciborowski J J H 1987 Dynamics of drift and microdistribution of two mayfly populations a predictive model Canadian Journal of Fisheries and Aquatic
71. ss section e centroids the XYZ coordinates of the centroids of each streamtube polygon in the given row and column e tubeHeights the average heights m of each streamtube in each column e total depth through centroid the depth m of water from the surface to the bed measured on a line through the centroids of the given streamtube polygons ioi xi File Edit Format Yiew Help Number of vertical tubes 5 Number of horizontal tubes 10 column width 1 65593 0 823772 0 729452 0 671193 0 715612 0 7655 column roughness 0 136227 0 16 0 16 0 16 0 16 0 16 0 16 0 15568 right bottom 1512 46 1473 58 0 1511 56 14 74 97 0 46899 1511 11 1475 66 0 594 left bottom 1511 56 1474 97 0 46899 1511 11 1475 66 0 594312 1510 71 1476 2 row 4 areas 0 0536407 0 0612848 0 0674337 0 0668178 0 0706579 0 073C velocities 1 41648 1 2398 1 12675 1 13714 1 07534 1 04072 0 83182 0 922438 0 7847 centroids 1511 86 1474 51 0 0215954 1511 33 14 75 33 0 0374046 1510 9 1475 9 7 tubeHeights 0 0431908 0 0748091 0 0927011 0 0995568 0 0987387 total depth through centroid 0 31266 0 534349 0 650489 0 694 799 0 6895 Figure 16 A cross section dump from the drift model 4 6 Altering the model parameters It is possible to alter the relative impact of the various parameters driving the distribution of drift within the model This function allows for sensitivity testing of the model DriftModel gt Set H Params brings up a dialog
72. streamtubes on either side The settling rate was set to zero in this model run and the arrow indicates the direction of flow 4 3 Drift model data requirements The drift modelling takes place in a drift model executable which may be interface driven or command line driven Note also that there are customisations of this program to handle specific situations such as StreamTubes files generated by R2DStreamTubes or by RhyHabSim This program requires three input files StreamTubes invertebrate parameters and initial invertebrate densities The filenames of these files can be specified when calling the program from the command line or a batch file e g drift exe d wam oO xsIn txt flowQ10 txt invertParams txt initialConc txt or specified interactively through the graphical user interface GUI if using that implementation of the program 1 StreamTubes a text file generated by R2DStreamTubes as described above see Figure 8 2 Invertebrate parameters a text file providing the following information on the invertebrates in the drift Figure 11 e number of length class taxon classes of drifting invertebrates Then for each drift class CAWTHRON June 2012 13 CAWTHRON a name for the length taxon class entry rate into drift from the bed No m s for the class settling velocity positive if sinking negative if rising m s length of invertebrate m dry weight of invertebrate g time spent near the
73. stributions Each different colour shaded area denotes the proportion of drift from a lateral position in source streamtube A exchanged into destination streamtubes B and C The total proportion of drift exchanged from streamtube A with streamtubes to the left and right is based on the average overlap of the destination streamtube with the Gaussian distributions centred on all possible lateral positions within column A see Appendix 1 ccscsssssscsscsssssscsssssssessscsssessssesscssesssssscsessessessessssessessesess 12 Figure 10 A plan view of modelled invertebrate drift dispersed down through a river reach For this illustration the centre four streamtubes were given an initial concentration at the top of the reach of zero so the invertebrate density in these streamtubes lower in the reach has dispersed in from streamtubes on either side The settling rate was set to zero in this model run and the arrow indicates the direction of flow 13 Figure 11 An invertebrate parameters text file cscssssssssssssssssssssssssessscssscsssssssssscssesessesssssssessessesessecsesees 14 Figure 12 A drift density initialization text file to initialise the first cross section with concentrations for 4 drift types 15 Figure 13 The DriftModel exe menu Dar sssscssssssesscsscsscssssssssscessscssssssscessssssssscssssssssssssssssssessssssessossesseass 15 Figure 14 An output file generated by CrossSections gt ConcentrationsDump from the drift model
74. t velocity unbounded flows El m Un 4k x S x y e A ATK xv This predicts the concentration of a dispersing vertical tracer plume injected into the flow at lateral position yo with tracer mass m and height H and travelling with velocity vx at downstream displacement x and lateral position y This equation is closely related I I I m to the normalized Gaussian function and when we let the constant represent the Hv rate of inflow of the tracer we can look at mixing as the product of that concentration i on P y 1 e790 1207 9 2k x and a normal Gaussian function y where o as ON2T v x demonstrated below June 2012 Appendix 1 33 CAWTHRON CAWTHRON First substitute 67 in the numerator gt 1 ya 2 P y gt ec Yo 207 2k x 1 y yo K2 oe Yo 4kyx 1 S rs Then substitute 67 in the denominator gt gt 1 v y yo x Py ee Os O 22 1 y 7 x P y e Oyo AR 2R X 27 V 1 a y y9 2 4k x P y Poy AR Ak x z ve Vv v y x P y x e x Yo 14k 47k xv Finally introduce the tracer inflow constant to give Rutherford s 1994 mixing equation e Hv j4ak xv gt v UV 4k yx v y yo 4ky x S x jo i al H ja xv This probability function calculates the likelihood that after a downstream displacement x a drifting invertebrate would be found to have lateral displace
75. tes resulting in the spatial density of invertebrate drift in the study reach The salmonid foraging model predicts spatially explicit net rate of energy intake NREI which can be easily converted to growth potential In this guide we focus on the way these three tools can be used together to predict how stream discharge and invertebrate drift concentration influence predicted feeding locations and the carrying capacity of a stream reach for large brown trout but the tools we describe could also be adapted to address a variety of other questions in river ecology TABLE OF CONTENTS 1 INTRODUCTION cssscssssssssssscssssssscssessssecscssssessnssesessessessssesscsessssasessossesssssssessassessssessesessesse 1 2 OVERVIEW OF MODELS 2 1 Stean Tubes Models nis actes ie ae cases EE E N RAS 22 Invertebrate drift model component 0 eee eeesseeseeeeecteeeeceseeseesecseeeeeesecseseeaeeseeaeeneseeesaees 4 2 3 Foraging model COMpPONENE 0 20 06 cissscbesstceeccasavesse seas a AEE EE EERE E ARa E LAAEN TA ER A 4 3 STREAMTUBES MODEL ccsssssssssssssssscssssessscsccsssesscssssessessesessecsccessessesssssssessessesssessesseseses 4 3 1 Purpose 0i he models grunnin a oE EERE EEEE E E EREE 4 3 2 Opening and displaying flow data s ssesseseseneseseesisessersrststsestsssrsrntstsistntntnrnrnerssrersesnerseseets 5 3 3 Creating CLOSS SECLIONS j ts t2 iscsscensssescssvasccssaesneesanssvcphssnenssdonssbsed EEEE he EERE A Re 6 3 4 Creating Strea
76. the process until the end of the modelled reach The optimal spacing for cross sections depends on the scale of interest in the model balancing resolution against processing time If for example you are interested in modelling the distribution of small fish with small foraging areas then cross sections should be more closely spaced As a starting point we have spaced cross sections at between one and two times the length of the fish we intend to model Once the series of cross sections has been constructed the X Sec gt Fix Crossing command should be employed to resolve any cross sections that intersect one another This is implemented by searching downstream when a cross section is found to intersect the cross section upstream of it one endpoint is moved to coincide with the endpoint of the intersected upstream cross section Thus the intersection is removed at the expense of having the average flow at something other than right angles to the revised cross section Implicit in this action is the realisation that in complicated hydraulic situations the positioning of cross sections is somewhat arbitrary It is noted that in complex river topographies sinuous reaches or those with islands peninsulas etc the algorithm may fail to correctly place cross sections If this is the case it may be necessary to manually divide the problematic reach into several sections creating a StreamTubes representation for each and modelling each separat
77. ts Al A3 1 1 Using the Matlab programs s sssessessersserseresseseees Al A3 2 Displaying outputs using River 2D sscccscsssssscccssscscssscsssscssssecsscssscscscsssceseceseees 42 June 2012 iii CAWTHRON 1 INTRODUCTION This is a guide to a suite of software tools for modelling river hydraulics invertebrate drift transport and foraging energetics of drift feeding salmonids A StreamTubes representation describes river flow using an array of streamtubes each of which conveys an equal proportion of the total discharge and the R2DStreamTubes tool generates one of those representations The invertebrate drift transport model predicts how the processes of entry transport dispersion and settling will determine the spatial distribution of invertebrate drift density in a river reach The salmonid foraging model predicts spatially explicit net rate of energy intake for fish feeding on drifting invertebrates In this guide we describe how these three tools can be used together to predict the effects of stream discharge and invertebrate drift concentration on predicted feeding locations and the carrying capacity of a stream reach for large brown trout The tools we describe could be adapted to address a variety of other questions in river ecology When used together these three software tools provide a process based modelling approach capable of predicting how changes in stream discharge bed topography water temperature
78. ture rate by about 45 presumably because the fish did not capture every prey item that entered their foraging areas An option is provided to correct for this over estimation See section 5 2 Modelling NREI When the foraging model is run the user specifies the number of points on each cross section at which the model will predict the fish s NREI Interrogation points are then spaced evenly across each cross section and combine to provide a two dimensional representation of predicted NREI i e a 2D plan view of the reach June 2012 20 CAWTHRON CAWTHRON Figure 18 Plan view of the foraging model showing the geometry of prey interception The fish is assumed to detect prey as they hit the surface of the hemispherical reaction volume with a radius equal to its reaction distance RD to the length class of the prey in question The fish intercepts prey at its maximum sustainable swimming speed Vmax and may only capture those prey it is able to intercept before they cross the line D E Under these conditions when water velocity is V the maximum lateral capture distance a G fe max h MCD is Hughes et al 2003 Fish position foraging radials and capture area ALO Depth m Width m Figure 19 Cross sectional view of the geometry used by the foraging model showing fish position foragi

user manual - Cawthron Institute

Contents

Download Pdf Manuals

Related Search

Related Contents