Project Purpose
The project objective is to identify potential segments of Texas waterways to be classified as "National Wild and Scenic Rivers". According to the Wild and Scenic Rivers homepage of the National Parks Service, only one river reach in Texas currently holds this designation. Authorized in 1978, this river reach is "a 191.2-mile strip on the American shore of the Rio Grande in the Chihuahuan Desert…It begins in Big Bend National Park and continues downstream to the Terrell-Val Verde county line."
This initial screening for potential river segments will be based on the following criteria:
The approach for identifying the river segments free of the affects of upstream dams will be to subdivide the 13 major river systems in Texas into subwatersheds using dam locations as outlets. The number of dams within each watershed will then be counted and the watersheds with the fewest dams will be selected. The next step will be to identify the river segments within each watershed that drain a significantly large area. These segments will be further queried to find those that flow through forested regions.
Data Sources
The data used for this analysis were obtained
from the dataset: "Hydrologic Modeling in Texas Using GIS", August 1997,
compiled by Francisco Olivera, Juling Bao, and David Maidment, Center for
Water Research in Water Resources, University of Texas at Austin. A number
of data layers from this source were used including: DitchDEM, Flow Direction,
Flow Accumulation, Reservoirs, Stream, Land Use and a raster map of the
study region. Additionally, several steps developed in the HEC-PREPRO hydrologic
processing module were used to analyze this data. More information about
the data can be found here.
More information about the data processing module can be found here
and here.
Information about specific coverages used and new sets generated can be
found in the data dictionary.
The following tasks and procedures were used to complete the project:
TASK 1: Identify Regions where Dams are Infrequent
Project Procedures
Procedure 1: Create a point shape file of dam locations in Texas Top
It was important that dam locations overlay on the stream grid coverage for the PRE PRO module to work properly. To ensure a good match, a new point shape coverage was created using the following procedure:
Procedure 2: Set Parameters for the PRE-PRO model Top
In order to use the PRE-PRO package to create
new watersheds based on the dam locations, several themes must be present
in the PRE-PRO project shell. Some of these are already present in the
"Hydrologic Modeling" dataset and others were generated. The table below
summarizes the functions that were used in the PRE-PRO module along with
their associated data layers. The color of the text identifies the origin
of the data layer:
| Function | Input Layer(s) | Output Layer(s) |
| Stream Segmentation (Links) | FlowAccumulationGrid[FlowAccum]
StreamGrid[Streams] |
LinksGrid[Links] |
| Add Outlets | LinksGrid[Links]
OutletsGrid[Outlets13] AdditionalOutlets[Dams.shp] StreamGrid[Streams] FlowDirectionGrid[FlowDir] |
ModifiedOutletsGrid[Modoutlets]
ModifiedLinksGrid[Modlinks] |
| SubWatershed Delineation | FlowAccumulationGrid[FlowAccum]
ModifiedOutletsGrid[Modoutlets] |
WatershedGrid[Damshed] |
|
|
Using the HEC-PREPRO Tools, the Links theme was created by the following procedure:
Procedure 3: Create Sub-Watershed based on Dam locations Top
The next step is to use the Add Outlets and Sub-Watershed delineation tools to define watershed regions upstream from dams. The Add Outlets function modifies both the Links and Outlets13 grids. Additional outlets are added to the Outlet grid from a user defined point shape file. The Links grid—a segmented version of the Streams grid—is also modified by the point shape file. The new modified Outlets grid was used to run the Sub-Watershed Delineation function. To complete this procedure, the following steps were used:
When this procedure was completed, 79 sub-watersheds were created. Of these only 66 were associated with a "dam outlet". The remaining 13 were associated with the original outlets of the major river systems of Texas. Closer inspection of the resulting watersheds revealed that dams along the main channel of rivers were favored over those on tributaries. It would be impossible this process to create a one to one relationship between a "dam outlet" and a sub-watershed. Additionally, no information about the presence or absence of a dam immediately upstream of a watershed was generated. These two drawbacks to the use of HEC-PREPRO to meet the first criterion defined for this project lead to the following procedure.
Procedure 4: Count Number of Dams in Each Sub-Watershed Top
The delineation of sub-watershed to locate "Wild" or undammed river stretches was only partially successful. Only 66 sub-watersheds from a total of 196 dam locations were created or roughly 33 percent. As a result there was not a one-to-one relationship established between sub-watershed and dam. This relationship would enable the selection of watersheds based on the size of watershed or the length of the flowpath upstream. The limited success of the approach was that a number of watersheds were created that contained only one dam at the outlet and were not downstream of another watershed. The following procedure was completed to count the number of dams in each watershed to limit the search to these watersheds:
Procedure 5: Select Sub-Watersheds with minimal dam influence Top
From inspection, the Water79.shp file shows graphically which regions of Texas are most dammed. Those lightly shaded have only one dam, while the darkest have 13-15. No information is associated with the themes about the number of dams upstream or their distance upstream. If desired the "Wild" watershed could be selected manually, by looking for upstream/downstream dam relationships. The following procedure was used to "Wild" watershed:
Procedure 6: Identify River Segments that Drain Threshold Area Top
Once the "Wild" watersheds have been created in Procedure 1-5, the river network to be queried needs to be established. Under criterion 2, only rivers that have a significant amount of flow should be used in this exercise. Its time to "separate the rivers from the streams". The amount of flow in each river segment is going to be roughly related to the amount of area that that segment drains. The values in the flow accumulation grid were used screen out those stream segments that don’t receive significant flow. From several inspections, the threshold value that separates a river from a stream was determined to be 256860. The following procedure was used to remove the streams:
Procedure 7: Extract River Segments within Sub-Watersheds from TASK 1 Top
With the river network establish and the "Wild" watersheds selected, the union of the two will produce the ‘Wild" River. The following procedure was used:
Procedure 8: Extract River Segments that flow through scenic areas Top
With the Wild River segments established, the next step is to establish those portions of the river that flow through scenic countryside. A Qualitative decision was made to equate scenic areas with forested areas—more specifically those with an Anderson Land Use code of "4". The following procedure was used to identify these stretches:
Procedure 9: Displaying the Data Top
The third criterion has been met and the WildScenicRivers coverage created. The only remaining task is to identify exactly what rivers were selected and how to find them. The following procedure was used:
The final output of Procedures 1 through 9 was three river segments in Texas:
The three river segments selected are reasonable candidates for further investigation. It is important to note, however, that two of them are downstream of dams (Nueces and Pecos). This occurred because the watershed approach to identifying wild river segments did not account for upstream dams immediately upstream. This problem was the result of a limit of the problem solving approach. An alternate approach that might yield better results would be to analyze the river network using an Avenue script. This approach would generate information about upstream and downstream network relationship and calculate flow distances. An additional portion of the project that could be refined, is the inclusion of more dams in the analysis. The USGS listed 301 dams in Texas, only 197 were included here based on the reservoir locations.
Created 5/1/98, Last modified 5/7/98, djs dscott@mail.utexas.edu