Bureau of Reclamation Banner
Reclamation Home             Reclamation Offices             Newsroom             Library             Projects & Facilities

Fish Habitats by Species Life Stage for Reclamation 2D Hydraulic Modeled Reaches

Metadata:


Identification_Information:
Citation:
Citation_Information:
Originator: USDI; Bureau of Reclamation
Publication_Date: 6/26/2008
Publication_Time: Unknown
Title: Bulltrout_Fry_1100cfs_Easton_Reach
Geospatial_Data_Presentation_Form: vector digital data
Online_Linkage: \\IBR1EYOUNGD01\C$\Data_2_SDE_prep\test47filegd.gdb
Description:
Abstract:
Habitat locations for fish species life stages developed from two dimensional hydraulogic modeling information for a specific reach and flow level listed in the data sets name.
Purpose:
This is part of a larger study known as the Yakima River Basin Water Storage Feasability Study (YRBWSFS). The modeling was conducted to support analysis of how fish habitats change in the reach specified in this data sets name, based on flow variations. The final data sets will be used to help determine how fish habitat will be effected by potential new water storage options.
Supplemental_Information:
This data set is part of a larger set of information.
Below is a listing of the modeled flows for each reach: Easton Reach - 250, 500, 700, 900, 1100, 1300, 1500, 1750, 2000 Kittitas Reach - 667, 800, 1032, 1288, 1700, 2311, 2770, 2864, 3146, 4000, 5000, 6500, 8000, 10000 Naches Reach - 250, 500, 720, 1000, 1500, 2000, 2680, 3000, 4000, 6000, 8000 Union Gap Reach - 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000 Wapato Reach - 300, 500, 750, 1491, 2000, 2450, 5000, 7500, 15000
For each flow level, habitat for specific fish species (bulltrout, coho, rainbow trout, spring chinook, steelhead) at different life stages (fry, holding, incubation, spawning, subyearling, subyearling overwintering, yearling) were developed. Not every combination of species and life stage were developed.
The following reaches in the Yakima River Basin have had similar data produced from Reclamation modeling: Easton, Kittitas, Naches The following reaches in the Yakima River Basin have had similar data produced from USGS developed hydraulic modeling: Union Gap, Wapato
A separate metadata record exists for the USGS modeled reaches, since a different 2D hydraulic modeling approach was used.
Time_Period_of_Content:
Time_Period_Information:
Range_of_Dates/Times:
Beginning_Date: 200010
Ending_Date: 200505
Currentness_Reference: ground condition
Status:
Progress: Complete
Maintenance_and_Update_Frequency: None planned
Spatial_Domain:
Bounding_Coordinates:
West_Bounding_Coordinate: -121.173811
East_Bounding_Coordinate: -121.035557
North_Bounding_Coordinate: 47.240085
South_Bounding_Coordinate: 47.187778
Keywords:
Theme:
Theme_Keyword_Thesaurus: None
Theme_Keyword: Habitat
Theme_Keyword: Fish
Theme_Keyword: Riparian
Theme_Keyword: In-Stream Habitat
Theme_Keyword: hydrology
Theme_Keyword: water
Theme_Keyword: species
Place:
Place_Keyword_Thesaurus: None
Place_Keyword: Pacific Northwest
Place_Keyword: Washignton
Place_Keyword: Yakima Basin
Access_Constraints:
The USDI Bureau of Reclamation (Reclamation) provides public domain spatial data without restrictions. Access to spatial data that are considered 1) works in-progress, 2) for internal use, or 3) confidential, sensitive, or private may be restricted. Access to such data may require a Freedom of Information Act (FOIA) request. For further information, contact the Geospatial Data Administrator (see Point of Contact info).
Use_Constraints:
The USDI Bureau of Reclamation (Reclamation) provides spatial data """"""""as-is"""""""" without warranty of any kind, expressed or implied, including but not limited to, any warranties of fitness for a particular purpose. The burden for determining fitness for use lies entirely with the user. In no event will the producers of Reclamation spatial data be liable for any damages arising from the use of, or the inability to use, these data. The user assumes all responsibility for spatial and attribute accuracy, completeness, validity, and appropriateness with regard to the suitability of these spatial data for any specific use or application, and such use or application is at the user?s own risk.
Point_of_Contact:
Contact_Information:
Contact_Organization_Primary:
Contact_Organization: Upper Columbia Area Office, USDI Bureau of Reclamation
Contact_Position: GIS Specialist
Contact_Address:
Address_Type: mailing and physical address
Address:
US Bureau of Reclamation Box 1749 1917 Marsh Road
City: Yakima
State_or_Province: WA
Postal_Code: 98907-1794
Country: USA
Contact_Voice_Telephone: 509-575-5848
Contact_Facsimile_Telephone: 509-454-5611
Contact_Electronic_Mail_Address: eyoung@pn.usbr.gov
Hours_of_Service: 0730-1600
Contact_Instructions:
Data_Set_Credit:
Denver Technical Service Center, USDI Bureau of Reclamation USGS; Columbia River Research Laboratory
Native_Data_Set_Environment:
Microsoft Windows XP Version 5.1 (Build 2600) Service Pack 2; ESRI ArcCatalog 9.3.0.1770

Data_Quality_Information:
Attribute_Accuracy:
Attribute_Accuracy_Report:
Bed elevation used in the modeling process is tied to the accuracies of the bathymetric LiDAR work conducted in 2004 and 2005 with a listed error or 0.25 meters (suspected accuracy is 0.1 meters).
The habitat classifications are tied to the water surface elevation, bed elevation, and water velocity from the modeling process.
For more information, see the process section of this document.
Logical_Consistency_Report: polygon topology exists
Completeness_Report:
Completely finished for the project areas. The habitat analysis was completed for the following reaches: Easton, Kittitas, Naches, Union Gap, Wapato
Positional_Accuracy:
Horizontal_Positional_Accuracy:
Horizontal_Positional_Accuracy_Report:
The horizontal accuracy value is based on the published accuracies for the equipment used when collecting the bathymetric LiDAR data (0.25 meters). This is the major data set used to create the three dimensional mesh for the modeling. The modeling output was then used to determine fish species habitat locations. Based on error analysis, it is believed that the accuracy value is closer to 0.1. However the modeling process undoubtedly added further error to the base data. Also, the propagation of the modeling point outputs to a continuous surface adds error, especially around the river edges.
Quantitative_Horizontal_Positional_Accuracy_Assessment:
Horizontal_Positional_Accuracy_Value: 0.25
Vertical_Positional_Accuracy:
Vertical_Positional_Accuracy_Report:
The horizontal accuracy value is based on the published accuracies for the equipment used when collecting the bathymetric LiDAR data (0.25 meters). This is the major data set used to create the three dimensional mesh for the modeling. The modeling output was then used to determine fish species habitat locations. Based on error analysis, it is believed that the accuracy value is closer to 0.1. However the modeling process undoubtedly added further error to the base data. Also, the propagation of the modeling point outputs to a continuous surface adds error, especially around the river edges.
Quantitative_Vertical_Positional_Accuracy_Assessment:
Vertical_Positional_Accuracy_Value: 0.25
Lineage:
Process_Step:
Process_Description:
This dataset was produced through a three step process: 1) Two dimensional hydraulic model created for the specific reaches. 2) Habitat location analysis conducted using the two dimensional hydraulic model outputs. 3) Habitat location analysis overlayed with general habitat classifications for added value and cross checking attachment to the main river system.
THE FOLLOWING IS A DESCRIPTION OF THE ORIGINAL TWO DIMENSIONAL HYDRAULIC MODELINGWORK FOLLOWS: The following process information is taken from a 2006 Reclamation technical report draft entitled "Identifying Salmonid Habitat With a Two-Dimensional Hydraulic Model", prepared by Robert C. Hilldale of the Sedimentation and River Hydraulics Group.
Description of Modeled Reaches: Five reaches were chosen for a 2-D hydraulic model to obtain a data set with higher habitat resolution than a 1D model. The reaches were chosen based primarily on habitat characteristics outlined by Stanford et al. (2002), who stated that restoration efforts in the Cle Elum, Kittitas, Naches, Gap-to-Gap, and Wapato Reaches provide the greatest potential for improvement to salmon and steelhead populations. It was decided by the Storage Study Technical Committee to replace the Cle Elum Reach with the Easton Reach, which is considered to be of biological significance. Three of the reaches were modeled by Reclamation (Hilldale & Mooney) and two reaches were modeled by the USGS (Hatton) (Figure 1). The reaches modeled by Reclamation are referred to as Easton (RM 193 to 203.5) and lower Kittitas (RM 149 to 153.5) on the Yakima River and Naches (RM 4 to 14) on the lower Naches River. The reaches modeled by the USGS are referred to as the Gap to Gap reach (RM 109 to 118) and Wapato (RM 82.5 to 91.5) on the Yakima River. The USGS is responsible for reporting results for the Gap to Gap and Wapato reaches.
The Easton Reach of the Yakima River begins just downstream of Easton Dam near River Mile 203.5 and continues downstream to the Interstate 90 Bridge at River Mile 193 (Figure 2). The upstream end of this reach (RM 199 to RM 203) is characterized as anastomosed, with the remaining portion of the river being single-thread. A few locations in this reach are vertically controlled by bedrock. Flow rates in this reach are typically much less than downstream of the Cle Elum River mouth. Flows modeled in this reach range from 250 to 2,000 ft3/s. The characteristic slope of this reach is 0.24%. This reach contains more large woody debris than the other two reaches modeled by Reclamation.
The lower Kittitas Reach begins two miles downstream of the Irene Reinhart boat launch and terminates near the head of the Yakima Canyon (Figure 3). The Reaches Report (Stanford et al., 2002) identifies this area as the reach of the Yakima River with the highest habitat channel complexity. The range of flows modeled in this reach were 540 to 10,000 ft3/s. The characteristic slope for this reach is 0.25%. Significant side channel habitat exists throughout the reach.
The Naches Reach begins at the Naches Bridge in the town of Naches (RM 14) and terminates near the Highway 12 twin bridges (RM 4) (Figure 4). This reach of the Naches River is subject to numerous irrigation diversions and returns, which were not accounted for in the model. It was decided that the additional effort of accounting for the many diversions and returns would not significantly alter the analysis of habitat in this reach. In the lower Naches River, there are many locations where the water surface varies across the cross section. This typically occurs at locations that have split channel morphology, where riffles exist on both sides but in different locations longitudinally. Many of these locations were noted during a raft trip down the reach. A check of the modeled water surface elevations at these locations verified that this type of feature was properly duplicated in the model. The flows modeled for this reach ranged from 250 to 8,000 ft3/s.
Data Sources: The data used to build the model terrain and bathymetry was obtained from two sources. Terrestrial LiDAR (Light Detection and Ranging) was collected aerially in November, 2000. The underwater portion, or bathymetry, was collected with water penetrating LiDAR in September, 2004 for the Easton and lower Kittitas reaches and May, 2005 for the Naches reach. The point data from both LiDAR sets and contour lines generated from the terrestrial LiDAR were used to construct a continuous surface of above and below water terrain using Arc GIS. In many locations break lines had to be added along contours and adjusted to create a smooth transition from under water to above water terrain where the two sets of points met. However the two data sets merged quite well with the exception of a few locations on the Naches River. In those locations the river had migrated in the time between the terrestrial and bathymetric data collection, making it necessary to eliminate some of the terrestrial data points and use the bathymetry data. The complete set of bathymetric and terrestrial data points and contours were used to create a Triangulated Irregular Network (TIN). This surface was then converted to feature points for interpolation to the mesh, discussed in the next section. The spot spacing for the bathymetry data is at a resolution of no greater than 6.5 feet. This is the greatest resolution provided by the LiDAR Bathymeter. Hilldale (2006) discusses more information regarding the bathymetric LiDAR data.
Mesh Generation: The finite element mesh used for the model was created using the Surface-water Modeling System (SMS) software package. The mesh combines structured and unstructured regions. The structured portion of the mesh represents the channeled portions of the terrain. The unstructured portion represents the overbank areas of the terrain (Figure 5). The structured portions of the mesh represent the channel using rectangular cells with the long dimension coincident with the upstream/downstream direction. The unstructured portion of the mesh consists of three, four or five sided polygons. At greater distances from the channel the unstructured mesh resolution decreases (cell size increases) in order to reduce the overall number of mesh cells. This type of mesh greatly reduces computing time while maintaining sufficient resolution in areas of greater interest. The resolution of the structured mesh is generally consistent throughout each model and typically varies in the width (short dimension) as the active channel changes width. Structured meshes require the same number of nodes across the channel at each end of a region. Where there are significant changes in active channel width, regions were created in such a way that the number of nodes across the channel could be adjusted to maintain a mostly consistent resolution. For all reaches the cell sizes in the active channel varied from approximately 6 to 10 feet in the lateral direction and 10 to 20 feet in the streamwise direction. The mesh cell size is related to the resolution of the survey data.
Hydraulic Model: The following excerpt is a comprehensive description of the GSTAR-W model, taken from the User's Manual for GSTAR-W (Lai, 2006b).
GSTAR-W, Generalized Sediment Transport for Alluvial Rivers and Watersheds, is a two-dimensional (2D) hydraulic and sediment transport model for river systems and watersheds. It has been developed primarily for use by Reclamation engineers to solve various hydraulic and sedimentation problems; and it has been applied successfully to many projects at Reclamation.
GSTAR-W is a 2D model that may be used to predict water flow and sediment transport for river reaches or water runoff and sediment delivery for a watershed. GSTAR-W adopts an approach for coupled modeling of channels, floodplains, and overland flow. Major features include the following:
Hybrid Zonal Modeling: GSTAR-W divides a watershed or river reaches into modeling zones. A zone may represent a 1D river reach or a 2D feature that may be solved with suitable models and algorithms. This layered hybrid approach facilitates the use of most appropriate models and solvers for each zone; it also extends the model to larger spatial and time scales.
Geometry Representation: The arbitrarily shaped element method (ASEM) of Lai (2000) is adopted for geometry representation. This unstructured meshing strategy is very flexible and facilitates the implementation of the hybrid zonal modeling concept. It essentially allows the use of most existing meshing methods available. For example, it allows a natural representation of a channel network in 1-D or 2-D, as well as the surroundings (flood plains or watersheds). With ASEM, a tight integration between watershed and channel system is achieved and a truly mesh-convergent solution may be obtained.
Major capabilities of GSTAR-W are listed below: -GSTAR-W solves the 2-D form of the diffusive wave or dynamic wave equations. The dynamic wave equations are the standard St. Venant depth-averaged equations; -Both diffusive wave and dynamic wave solvers use the implicit scheme so that solution robustness and efficiency may be achieved for a majority of applications; -Both steady or unsteady flows may be simulated; -Unstructured or structured 2-D meshes, with arbitrary element shapes, may be used with GSTAR-W. In most applications, a combination of quadrilateral and triangular meshes works the best. A Cartesian or raster mesh is a special mesh that may also be used by GSTAR-W; -All flow regimes, i.e., subcritical, transcritical, and supercritical flows, are simulated simultaneously; -Solution domain may include a combination of main channels, floodplains, and overland; -Both steady and unsteady sediment transport may be simulated with the nonequilibrium approach for nonuniform sediment transport; -Sediment transport module includes more than 10 non-cohesive sediment transport capacity formulae that are applicable to a wide range of hydraulic and sediment conditions. -Fractional sediment transport with bed sorting and armoring;
GSTAR-W is a 2-D model and it is particularly useful for problems where 2D effect is important. Examples include flows with in-stream structures, through bends, with perched rivers, and for multiple channel systems. A 2-D model may also be needed if one is interested in local flow velocities and eddy patterns.
The 2-D channel network portion of GSTAR-W was used for modeling the reaches in this report using a diffusive wave solution with an implicit scheme. The diffusive wave solution assumes the convective and diffusive transports of water are in equilibrium. Eddies and flow separation are not considered, resulting in a loss of the ability to calculate micro- and meso-scale flow conditions that might occur in the vicinity of large in-stream boulders or woody debris. The scale of the survey data, on the order of 6.5 ft, does not provide proper resolution to define such meso-scale stream structure, much less the micro-scale. Although the diffusive wave solver can calculate sub-, super-, and trans-critical flow, a hydraulic jump is transitionally smoothed. For the purpose of this modeling effort, these properties are not crucial. For the diffusive wave solver, the Manning roughness coefficient should be interpreted as the energy loss coefficient. Additional losses due to eddies, separations, and hydraulic jumps are lumped together within this coefficient. These roughness values are on the order of traditional 1-D models.
The outputs available from this model are spatially rectified values of x-velocity, y-velocity, magnitude velocity, depth, water surface elevation, bed elevation, and Froude number. This data can be manipulated and presented in a number of ways, including shape files in Arc GIS.
Roughness: Various roughness values were assigned to specific regions of the model via polygons while creating the mesh. In all models there were four primary roughness categories used to represent the various conditions; main channel, littered gravel bars, side channels, cultivated floodplain, and forested regions. The Manning's roughness of the side channels and littered gravel bars increased by 0.005 over the main channel roughness. It is common for debris to collect and small vegetation to grow in side channels, slightly increasing the roughness. The same applies to gravel bars where low vegetation growth and forest litter can slightly impede flow. The forested portions of the floodplain were given the highest roughness. Manning's roughness values used were: main channel = 0.03 - 0.035; side channel/littered gravel bar = 0.035 - 0.045; cultivated field areas = 0.04 - 0.045; forested areas = 0.055 - 0.07.
Boundary Conditions: The models include three types of boundary conditions; upstream, downstream, and a slip boundary. The upstream boundary is the incoming flow rate. The downstream boundary is a fixed water surface elevation. The slip boundary is essentially a friction-free 'wall' built around the model. Ideally the flow should not meet the no-slip boundary.
The downstream boundary condition requires knowledge of the water surface elevation at various flow rates. A rating curve can be developed over a wide range of measured flow rates and corresponding water surface elevations. In the absence of adequate measured data, a 1-D hydraulic model can be constructed for the downstream most portion of the reach, from which a stage-discharge rating curve can be developed. The latter method requires at least one or two known water surface elevations to verify the 1-D model results.
The downstream boundary condition for all three models used water surface elevation values determined for each flow with a 1-D hydraulic model verified with surveys at two different flow rates.
Model Verification: Verification of the models was performed by matching surveyed water surface elevations with modeled water surface elevations at various locations throughout each reach for a minimum of two different flow rates. Surveys of water surface elevations were performed with RTK GPS surveying equipment at accessible locations along each reach.
Below is an array of verification data comparing water surface elevations. The mean value and standard deviation are statistics on the differences between modeled and surveyed water surface elevations (modeled value minus surveyed value). Flows at the Naches Gage at Naches during the 2680 CFS survey fluctuated 80 ft3/s. Additionally, flows fluctuated spatially throughout this same reach by 63 ft3/s due to irrigation diversion and return. An average of the estimated flows at each location was used for the modeled flow rate. Easton at 250 CFS: mean error = -0.02; standard deviation = 0.33; number of locations = 7 Easton at 500 CFS: mean error = 0.18; standard deviation = 0.27; number of locations = 4 Kittitas at 1032 CFS: mean error = 0.23; standard deviation = 0.49; number of locations = 4 Kittitas at 3146 CFS: mean error = -0.04; standard deviation = 0.43; number of locations = N/A Naches at 720 CFS: mean error = 0.01; standard deviation = 0.36; number of locations = 5 Naches at 2680 CFS: mean error = 0.36; standard deviation = 0.33; number of locations = 4
Habitat via Froude Number: A major component of this study was to use the model output to identify habitat types, i.e. pools, riffles and glides, at various flow rates. Jowett (1993) performed an analysis whereby habitat types were numerically determined using a variety of methods to increase replicability and predictability in river studies. Jowett evaluated habitat using the Froude number, slope, velocity/depth ratio and combinations of these values. Although it was found that a velocity/depth ratio best described these habitats, the difference in success between the Froude number and velocity depth ratio was small. For the present study, it was determined that the Froude number better defined the habitat types when compared to field data.
Using the model output at all flow rates, the Froude number (Fr = V/(square root of gh)), where V is depth averaged velocity, g is the gravitational constant and h is the flow depth, was used to determine the areas comprising pools, glides and riffles. Determining the Froude values at which the habitat changes from one type to another was initially done by experiment, beginning with the values determined by Jowett (1993). The break in habitat Froude classification was then adjusted to match field surveys of identified habitat types in all reaches modeled. Generally, the Froude number changes very little from one mesh cell to another, creating clusters of mesh cells corresponding to pools, glides or riffles.
The following piecewise function was used for habitat determination: Fr < 0.09 = pool 0.09 <= Fr <= 0.42 = glide Fr > 0.42 = riffle
The break in Froude number between pools and glides used in this study differs from that used by Jowett (1993), who used Fr = 0.18 but is very similar to that used by Reuter et al. (2003), Fr = 0.10. The break between glides and riffles used in this study was determined to be Fr = 0.42, similar to Jowett (1993), who used Fr = 0.41. Reuter further defined a habitat category of races, 0.2 < Fr < 0.4, with riffles defined as having a Froude number greater than 0.40, similar to the findings of this study.
A TIN was created in Arc GIS to display the results. In the process of creating a TIN, neighboring cells are interpolated between discrete habitat cell values. These interpolations consist of one cell in each dimension bounding each cluster of habitat classifications. Habitat features naturally include transitions along the boundaries of pools, glides and riffles that exist in the stream. For calculations of habitat area, these transition zones are neglected.
Field Verification of Determining Habitat via the Froude Number: Field verification of the Froude habitat classification was performed from a raft using a hand-held Trimble GPS to mark the beginning and end of visually identified pools, glides and riffles throughout the reach. Joel Hubble, a fisheries biologist for Reclamation (Upper Columbia Area Office), assisted with the field habitat identification. The intent was to continuously map the reach, however there were instances where identifying the habitat was unclear. Identifying pools, riffles and glides is a subjective and qualitative process and led to some unidentified lengths. Another cause of unidentified portions of the reach can also be attributed to the difficulty in marking the beginning or end of a specific feature while floating in a raft. There were occasions where a feature type was not realized until the raft had already passed the true beginning of the feature.
It was noted during the float surveys and data processing that the habitat features (pool, glide and riffle) in the Naches and Kittitas reaches are much more easily defined than those in the Easton reach, from the standpoint of both field identification and modeling. The habitat features in the Easton reach are much more discontinuous. Evidence of this is shown in the average length of each feature identified in the field. Features in the Easton reach average 196 feet in length while features in the Naches and Kittitas reaches average 397 and 384 feet, respectively. This result is probably a function of scale. A smaller channel with lower discharge is likely to have smaller features.
Below is an array showing the percent of correctly classified features for habitat type: Easton: all features = 73%; riffles = 63% (69 surveyed in the field); glides = 87% (66 surveyed in the field); pools = 64% (19 surveyed in field) Kittitas: all features = 85%; riffles = 83% (17 surveyed in the field); glides = 91% (15 surveyed in the field); pools = 50% (2 surveyed in field) Naches: all features = 81%; riffles = 87% (34 surveyed in the field); glides = 77% (31 surveyed in the field); pools = 0% (1 surveyed in field)
Overall, the modeled classification agreed very well with the field verification. The number of pools identified by the model is not an appropriate representation of the total amount of pool habitat in each reach. The model can only account for geomorphically created pool habitat and does not account for pool habitat created by woody debris. The float trips of all three reaches identified a significant contribution to pool habitat by large wood in the channel. It is not feasible that large wood in a stream be surveyed to the detail required for numerical modeling when the model covers many miles of river channel. The aerial survey of the bathymetry flown for this model has a horizontal spot spacing on the order of 6.5-x-6.5 feet, meaning that features less than 6.5 feet in size are not accurately represented. The survey did not provide the level of detail required to account for large woody debris and associated pool habitat. Another important consideration is that the number of pools identified in the field habitat survey does not include the entire pool habitat of each reach. Pool habitat exists in the side channels and backwater areas that were not accounted for during the raft survey.
When pools, riffles, and glides are evaluated over various flow rates, their classifications begin to change. The length of pools becomes shorter with increasing flow as the features upstream and downstream encroach on the pool from each end until the entire pool feature becomes a glide. In some areas, a series of short riffles separated by short glides becomes one long riffle. With regards to glides and riffles, the greater tendency is for glides to transition into riffles with increasing flow, although this is not absolute. The ultimate determining parameter regarding this type of habitat is the water surface slope.
In addition to habitat identified via Froude number, flow depth and velocity are also provided for use as input to the EDT model.
Conclusion: The use of GSTAR-W in combination with terrestrial and bathymetric LiDAR has been shown to be effective in determining aquatic habitat for input to EDT.
The combination of bathymetric LiDAR with the terrestrial methods provided a more detailed representation of the terrain than traditional boat survey methods. The resolution captured geomorphic channel features that cross section techniques might miss.
GSTAR-W allows contiguous 2D hydraulic simulation over lengths and resolutions beyond the capabilities of most other 2D models. Two-dimensional results provided depths and velocities over and around complex channel features including laterally varied riffles, pools, and glides as well as side channels.
Froude number classification of pools, riffles, and glides provides an objective and repeatable means of quantifying aquatic habitat. Field verification at different flow rates provided support for the method. Classification did not include features formed by woody debris or features smaller than the mesh resolution.
THE FOLLOWING DESCRIPTION IS FROM THE USGS PERSONNEL RESPONSIBLE FOR THE HABITAT LOCATION ANALYSIS 1) Hydraulic habitat shapefiles a) Delphi exercise completed in May, 2006 to define suitable ranges of depths and velocities for each pertinent life stage of each target species: i) Bull trout ii) Coho iii) Fall chinook iv) Resident rainbow trout v) Spring chinook vi) Steelhead b) Two different types of reclassification tables developed from the Delphi definitions: i) Step 1 - Basic depth and velocity reclass tables developed to include and classify each break point of the variable from all the Delphi classifications combined. ii) Step 2 - Individual depth and velocity reclass tables developed for each life stage, based on the classes delineated in Step 1. [Note- the two-step reclassification was necessary because of the overlapping, but non-identical habitat suitability criteria developed by the Delphi panel.] c) Shapefiles containing hydraulic data for each simulated discharge obtained from Robb Hilldale. Each point shapefile was triangulated (TIN) and converted to depth, velocity, water surface, and water surface gradient grids, respectively. TINs and grids were clipped using the clip polygon shapefiles provided by Robb. d) Individual depth and velocity grids reclassified using the reclass tables developed under step b1, and combined into a single depth/velocity grid for each flow. e) Hydraulic habitat grids for each life stage developed by reclassifying the combined grids with the reclass tables defined in Step b2. f) All grids were converted to polygons, and the items "Target" (a code for the species and life stage) and "Flow" (showing the discharge pertaining to the shapefile) added to the attribute tables. 2) Shoreline buffered hydraulic habitat for fry, all species. a) Purpose - the general consensus of biologists we consulted, including members of the Delphi panel, suggested that newly-emerged fry tended to occupy shallow, low-velocity areas along the shorelines, that shallow, slow areas elsewhere (i.e., in the middle of the channel) were not utilized, and therefore, should not be considered to be viable habitat for this life stage. b) Development of the shoreline buffer polygon. i) The depth grids for each simulated flow were reclassified to distinguish wet and dry cells. ii) The reclassified grids were converted to polygon shapefiles, and all polygons representing dry areas were removed. iii) The remaining "wet" polygons were converted to polylines, and the polylines buffered by five feet to create polygons of wet areas within five feet of the water's edge. c) The hydraulic habitat (e.g., shallow, slow) polygons for the fry life stage of each species were intersected with the appropriate shoreline buffer polygon for each flow, resulting in polygons representing suitable hydraulic conditions within five feet of the shoreline. 3) "Feeding station" habitat for juvenile salmonids (summer rearing only). a) Purpose - Many biologists have observed that juvenile and resident adult salmonids select areas of streams having a combination of relatively slow water adjacent to a zone of relatively fast water. This combination provides an energetically efficient "feeding station" wherein the slow water area can be occupied by the fish without expending much energy to maintain position, and the nearby zone of higher velocity provides a higher delivery rate of drifting food items. The purpose of this exercise was to depict this phenomenon. Methods described below should be considered experimental at this point, and may be subject to modification (or not used in the DSS). b) Delineation of velocity shear zones. i) Velocity gradients utilized by salmonids estimated from observations of actively feeding juvenile rainbow in several streams in Colorado during the early 1990s. (1) Only data for rainbow trout between 7 and 30 cm TL used in the analysis. (2) Velocity gradients calculated as the difference between the velocity occupied by the fish and the adjacent velocity, divided by the distance between the location of the fish and the location of the adjacent velocity: Vslope = (Vadjacent - Vfocal point)/Distance (3) Data for non-positive velocity gradients were removed from the data set (approximately 10% of the observations). (4) Remaining data sorted by velocity gradient, in ascending order. The break-point for a minimally suitable velocity gradient was arbitrarily set at the 10% probability level for occupied locations. In other words, 90% of the observed fish utilized a velocity gradient equal to or greater than the gradient at the break-point. The velocity gradients for observations in this trimmed data set ranged from a minimum of 13% to a maximum of 250%. ii) Velocity grids for each simulated flows were used as input to the Slope_3d algorithm (Percent rise option) in ArcGis to create grids of velocity gradients. The velocity gradient grids were then reclassified according to the criteria defined in the previous step. iii) The reclassified velocity gradient grids were converted to polygon shapefiles, which were then expanded by a five foot buffer. iv) Intersections were performed using the buffered velocity gradient polygons and the hydraulic habitat polygons for sub-yearling and yearling life stages. The result was a series of polygon shapefiles representing areas of suitable depths and focal point velocities within five feet of a velocity shear zone having a gradient of 13% to 250%. Notes: The five-foot buffer around the velocity gradient polygon was entirely arbitrary. The observations from the Colorado data indicated that most of the fish locations were within two feet of the measured adjacent velocity, and relatively few were more than four feet away. The five-foot buffer was used in this experiment to ensure at least some overlap between the velocity shear area and the suitable hydraulic habitat. Redefining the buffer width and re-running the intersections would be a relatively trivial matter.
The adjacent velocities for many of the observations from the Colorado data were described as "overhead." The distance used to define the velocity gradients for these locations was set at 1 foot. This is probably not a terrible assumption, but true overhead velocities can not be described by a 2-d model. To achieve this level of detail would require a 3-d model, and probably would constrain the analysis to much smaller areas.
The "spot size" (i.e., the area represented by a data point) of the Lidar data used to describe the bathymetry of the Easton, Kittitas, and Naches reaches was about 2 m in diameter. Consequently, the level of detail in the hydraulic models for these reaches is likely to be somewhat coarser than for the Union Gap (and presumably Wapato) reaches, where the bathymetry was defined using higher precision survey techniques. Although the simulations for the upper three reaches are lacking in fine detail, it does appear that the velocity gradient grids were sufficient to depict shear zones at the tailouts of riffles and around point bars. Any feature smaller than about 2 m in diameter (e.g., individual boulders), however, is not likely to be detected by this model.
THE FOLLOWING IS A DESCRIPTION OF THE GENERAL HABITAT CLASSIFICATION DEVELOPMENT: 1) All flow levels for a specific reach were combined using the ArcGIS 'update' tool. 2) Any non-attached wetted area was eliminated from the dataset. 3) Reclamation biologists classified the habitat into the following categories: backwater, beaver pond, braid, groundwater, main, pond outlet, side channel, wetland, by overlaying ortho-photography with the combined modeled flow levels. 4) The results from step three above were combined with the USGS species life stage specific analysis to incorporate the general habitat classifications into an attribute.
Process_Contact:
Contact_Information:
Contact_Organization_Primary:
Contact_Organization: USDI; USGS; Columbia River Research Laboratory
Contact_Address:
Address_Type: mailing and physical address
Address: 5501-A Cook-Underwood Rd.
City: Cook
State_or_Province: WA
Postal_Code: 98605-9717
Country: USA
Contact_Voice_Telephone: (509) 538-2299
Contact_Facsimile_Telephone: (509) 538-2843
Hours_of_Service: 0730-1600
Contact_Instructions:
Cloud_Cover: 0

Spatial_Data_Organization_Information:
Direct_Spatial_Reference_Method: Vector
Point_and_Vector_Object_Information:
SDTS_Terms_Description:
SDTS_Point_and_Vector_Object_Type: G-polygon
Point_and_Vector_Object_Count: 3478
SDTS_Terms_Description:
SDTS_Point_and_Vector_Object_Type: Label point
Point_and_Vector_Object_Count: 55
SDTS_Terms_Description:
SDTS_Point_and_Vector_Object_Type: GT-polygon composed of chains
Point_and_Vector_Object_Count: 55
SDTS_Terms_Description:
SDTS_Point_and_Vector_Object_Type: Point
Point_and_Vector_Object_Count: 4

Spatial_Reference_Information:
Horizontal_Coordinate_System_Definition:
Planar:
Map_Projection:
Map_Projection_Name: Lambert Conformal Conic
Lambert_Conformal_Conic:
Standard_Parallel: 45.833333
Standard_Parallel: 47.333333
Longitude_of_Central_Meridian: -120.500000
Latitude_of_Projection_Origin: 45.333333
False_Easting: 1640416.666667
False_Northing: 0.000000
Planar_Coordinate_Information:
Planar_Coordinate_Encoding_Method: coordinate pair
Coordinate_Representation:
Abscissa_Resolution: 0.000328
Ordinate_Resolution: 0.000328
Planar_Distance_Units: survey feet
Geodetic_Model:
Horizontal_Datum_Name: North American Datum of 1983
Ellipsoid_Name: Geodetic Reference System 80
Semi-major_Axis: 6378137.000000
Denominator_of_Flattening_Ratio: 298.257222
Vertical_Coordinate_System_Definition:
Altitude_System_Definition:
Altitude_Resolution: 0.000100
Altitude_Encoding_Method:
Explicit elevation coordinate included with horizontal coordinates

Entity_and_Attribute_Information:
Detailed_Description:
Entity_Type:
Entity_Type_Label: Bulltrout_Fry_1100cfs_Easton_Reach
Entity_Type_Definition: Polygon attribute table
Entity_Type_Definition_Source: Denver Technical Services Center, USDI Bureau of Reclamation
Attribute:
Attribute_Label: OBJECTID
Attribute_Definition: Internal feature number.
Attribute_Definition_Source: ESRI
Attribute_Domain_Values:
Unrepresentable_Domain:
Sequential unique whole numbers that are automatically generated.
Attribute:
Attribute_Label: Shape
Attribute_Definition: Feature geometry.
Attribute_Definition_Source: ESRI
Attribute_Domain_Values:
Unrepresentable_Domain: Coordinates defining the features.
Attribute:
Attribute_Label: HABITAT
Attribute_Definition: Description of the general habitat types.
Attribute_Domain_Values:
Enumerated_Domain:
Enumerated_Domain_Value: backwater
Enumerated_Domain:
Enumerated_Domain_Value: beaver pond
Enumerated_Domain:
Enumerated_Domain_Value: braid
Enumerated_Domain:
Enumerated_Domain_Value: groundwater
Enumerated_Domain:
Enumerated_Domain_Value: main
Enumerated_Domain:
Enumerated_Domain_Value: pond outlet
Enumerated_Domain:
Enumerated_Domain_Value: side channel
Enumerated_Domain:
Enumerated_Domain_Value: wetland
Attribute:
Attribute_Label: habitat
Attribute_Definition: Length of feature in internal units.
Attribute_Definition_Source: ESRI
Attribute_Domain_Values:
Unrepresentable_Domain: Positive real numbers that are automatically generated.
Attribute:
Attribute_Label: Shape_Length
Attribute_Definition: Length of feature in internal units.
Attribute_Definition_Source: ESRI
Attribute_Domain_Values:
Unrepresentable_Domain: Positive real numbers that are automatically generated.
Attribute:
Attribute_Label: Shape_Area
Attribute_Definition: Area of feature in internal units squared.
Attribute_Definition_Source: ESRI
Attribute_Domain_Values:
Unrepresentable_Domain: Positive real numbers that are automatically generated.
Overview_Description:
Entity_and_Attribute_Overview: Contains attributes on the fish habitat classifications

Distribution_Information:
Distributor:
Contact_Information:
Contact_Organization_Primary:
Contact_Organization: Upper Columbia Area Office, USDI Bureau of Reclamation
Contact_Position: GIS Specialist
Contact_Address:
Address_Type: mailing and physical address
Address:
US Bureau of Reclamation Box 1749 1917 Marsh Road
City: Yakima
State_or_Province: WA
Postal_Code: 98907-1794
Country: USA
Contact_Voice_Telephone: 509-575-5848
Contact_Facsimile_Telephone: 509-454-5611
Contact_Electronic_Mail_Address: eyoung@pn.usbr.gov
Hours_of_Service: 0730-1600
Contact_Instructions:
Resource_Description: U. S. Bureau of Reclamation's Yakima River Basin Database
Distribution_Liability:
The USDI Bureau of Reclamation (Reclamation) provides spatial data ""as-is"" without warranty of any kind, expressed or implied, including but not limited to, any warranties of fitness for a particular purpose. The burden for determining fitness for use lies entirely with the user. In no event will the producers of Reclamation spatial data be liable for any damages arising from the use of, or the inability to use, these data. Reclamation liability is limited to providing a replacement copy for requested data sets that are not readable upon receipt.
The user assumes all responsibility for spatial and attribute accuracy, completeness, validity and appropriateness with regard to the suitability of these spatial data for any specific use or application, and such use or application is at the user?s own risk.
Spatial data provided by the Reclamation are public domain, and the recipient may not assert any proprietary rights thereto nor represent data sets as other than government-produced data.
Standard_Order_Process:
Fees:
No fees for electronic download. Processing, media, and shipping fees may apply for CD-ROM requests.
Ordering_Instructions:
Contact Geospatial Data Administrator or GIS Coordinator for custom data requests.
Turnaround:
Processing time and delivery will vary depending on the size and complexity of the request.
Standard_Order_Process:
Fees:
Actual costs to produce cartographic products, including labor, plotting costs, and shipping.
Ordering_Instructions:
Contact the Geospatial Data Administrator or GIS Coordinator to request non-digital data products.
Turnaround: Processing time will vary with the specific request.
Custom_Order_Process:
Contact Geospatial Data Administrator or GIS Coordinator for custom data requests.
Available_Time_Period:
Time_Period_Information:
Single_Date/Time:
Time_of_Day: unknown

Metadata_Reference_Information:
Metadata_Date: 20081201
Metadata_Review_Date: 20010814
Metadata_Contact:
Contact_Information:
Contact_Organization_Primary:
Contact_Organization: Upper Columbia Area Office, USDI Bureau of Reclamation
Contact_Person: REQUIRED: The person responsible for the metadata information.
Contact_Position: GIS Specialist
Contact_Address:
Address_Type: mailing and physical address
Address:
US Bureau of Reclamation Box 1749 1917 Marsh Road
City: Yakima
State_or_Province: WA
Postal_Code: 98907-1794
Country: USA
Contact_Voice_Telephone: 509-575-5848
Contact_Facsimile_Telephone: 509-454-5611
Contact_Electronic_Mail_Address: eyoung@pn.usbr.gov
Hours_of_Service: 0730-1600
Contact_Instructions:
Metadata_Standard_Name: FGDC Content Standards for Digital Geospatial Metadata
Metadata_Standard_Version: FGDC-STD-001-1998
Metadata_Time_Convention: local time
Metadata_Access_Constraints: none
Metadata_Use_Constraints: none
Metadata_Security_Information:
Metadata_Security_Classification: Unclassified

Generated by mp version 2.9.6 on Mon Dec 01 12:05:44 2008