EXECUTIVE SUMMARY

The National Estuary Program was established through the Federal Water Quality Act of 1987 to identify and protect significant estuaries in our nation. Twenty-eight estuaries are now funded and administered by the United States Environmental Protection Agency (EPA). Corpus Christi Bay was formally nominated for the program in 1992 by Governor Ann Richards (Texas Sea 1994).

The Corpus Christi Bay National Estuary Program is designed to coordinate the activities of all regulatory authorities along with input from the non-government sector. The project includes 75 miles of south-central Texas coastline; 550 square miles of water including all bays and saltwater bayous, Aransas Bay, Corpus Christi Bay, Baffin Bay, and Upper Laguna Madre; twelve Texas counties including McMullen, Live Oak, Bee, Refugio, Aransas, San Patrico, Jim Wells, Duval, Nueces, Klegberg, Kenedy and Brooks; and includes the Padre Islands (Corpus 1996).

• Introduction
  
• Project Procedure
  1. "Burning in" Stream Network
  2. Connecting the Stream Network and Bay System
  3. Annual Precipitation in the Basin
  4. Create a Runoff Grid
  5. Linking Land Use to Expected Mean Concentration (EMC) of Pollutants
  6. Estimating Annual Loadings
  7. Estimating Equilibrium Constituent Concentrations in the Bay System
  
• Conclusions
  
• Future Work
  
• Data Dictionary
  
• References
  

The objective of this research project is to complete a Geographic Information System (GIS) assessment of total constituent loadings for the Corpus Christi National Estuary Program (CCBNEP). The tasks required to accomplish the project objectives consist of calculating the constituent loadings applied to the land surface and the resulting loads in the river network and the bay system. Along with the land surface loadings, point sources and atmosphere deposition are added to the loadings grids. Following the determination of the constituents loadings are determined in the bay system, the equilibrium concentration level of each of the constituents is calculated.

The first portion of the project has been completed to date and includes data collection for the base map for the total constituent loadings study. The data collected includes political boundaries, hydrologic boundary data, River Reach Files (RF1), Digital Elevation Maps (DEM), land use data and corresponding event mean concentration (EMC) data, Digital Line Graphs (DLG), and precipitation data. A paper entitled Completing a Base Map for a Geographic Information System (GIS) Assessment of Nonpoint Source Pollution for the Corpus Christi Bay National Estuary Program(CCBNEP) was written by Ann Quenzer in December of 1996 to describe the process in which the base map data is collected.

This report will discuss the second portion of the project which includes the GIS assessment of nonpoint source pollution for the Corpus Christi Bay National Estuary Program. The report includes a discussion of the steps needed to download and prepare the digital database for an analysis of the nonpoint source loadings. Also included is a brief narrative and actual Arc/Info and UNIX commands needed to establish the database. Commands entered into the computer by the reader are in bold type, whereas computer prompts and messages are in italics. This report will discuss the steps needed to calculate constituent loadings on the land surface and the resulting river network and bay system loadings. The report also includes a discussion of the steps needed for the calculation of the equilibrium concentrations in the bay system. Total nitrogen will be used as an example of constituent loading throughout the report.

To complete the second portion of the project, all other constituent loadings need to be calculated on the land surface, in the river system, and finally, in the bay system. The other constituents include: ammonium, nitrate-nitrite, organic nitrogen, total phosphorus, ortho-phosphate, oil and grease, malathion, diazinon, and metals (as data allows).

The third portion of the project will include in the model point source data and atmospheric deposition which affect the Corpus Christi Bay System.

The report also contains a data dictionary which discusses each of the grids and coverages used.

1. "BURNING IN" THE STREAM NETWORK

To begin the assessment of nonpoint pollution, it is necessary to initiate the "burning in" process which involves the Avenue script AGREE written by Ferdi Hellweger. The script requires the stream vector coverage and the Digital Elevation Model (DEM) to operate properly. The program AGREE reconditions the surface elevation of the DEM to match those of the stream vector coverage. The AGREE program runs within the ArcView environment and can be downloaded from the Internet site agree.txt (Hellweger 1997).

Download the Digital Elevation Model (DEM) and Digital Line Graph (DLG) files as discussed in the report entitled Completing a Base Map for a Geographic Information System (GIS) Assessment of Nonpoint Source Pollution for the Corpus Christi Bay National Estuary Program (CCBNEP) in the sections Downloading DEM data from the Internet and Downloading Digital Line Graphs from the Internet. Both sets of data are also obtainable from the United States Geological Survey's Internet site: http://edcwww.cr.usgs.gov/doc/edchome/ndcdb/ndcdb.html (USGS 1996).

Download the ArcView project which contain the Avenue scripts used to calculate the equilibrium concentrations in the bay system. The project is called balance.apr and is downloaded from the Center for Research in Water Resources (CRWR) ftp site:

Address: ftp.crwr.utexas.edu
Login: anonymous
Password: your e-mail address
Directory: /pub/crwr/gishydro/balance
Transfer Mode: binary
Program: balance.apr

Note: If you are new to ftp you might want to read the page entitled Getting Data From CRWR's Anonymous FTP Site.

Once the data and project are downloaded, begin the project by starting ArcView 3.0.

$arcview3 &

Open the project balance.apr by clicking on the File/Open Project, highlighting balance.apr in the window, and click OK. Click on the pull down menu File/Extensions and add the Hydrology and Spatial Analysis Extensions by checking the appropriate boxes.

Click on the Project Window and Open a new view. By going to the View/Properties, change the name of the view to Agree.

Add the DEM grid to the view by clicking on , change the Data Source Type to Grid Data Source, highlight dem, and click OK. Once the theme has been added, make the dem theme active and check the box next to the theme to view the DEM of the area.

By using the Arc command Describe on the DEM grid, it was found that the cell size is 100m, the elevation ranges from 0 meters to 289 meters, and the total number of cell in the study area is 5,854,170 making the study area 58,541,700 square kilometers.

Add the stream coverage to the view by clicking on , change the Data Source Type to Feature Data Source, highlight dlg, and click OK. Once the theme has been added, make the dlg theme active and check the box next to the theme to view the rivers in the area.

Click on the Project Window, click on the Scripts window and open a new script window (labeled Script1). From the Menu Bar, select Script/Load Text File. Click on agree.txt, and click OK. Using Script/Properties change the name of the script to Agree. The program will calculate a new grid called, elevgrid, which contains the modified elevations within the buffer zone and the original elevations outside of the buffer zone. Within the script, there is a line that states to which file the result will be saved:

avenue.savedataset("/home/ferdi/quenzer/trialrun/elevgrid".asfilename)

Edit this path name in the Script so that the result will be saved to the correct directory.

Compile the script by clicking on the check button, and run it on top of the Agree view. Make the View active, make the DEM and DLG themes active, and then click on the Agree script window and hit the Running man next to the check button to initiate the script. A message box will appear prompting a selection of the variables used in AGREE for the grid calculations. The variables include the elevation grid, the vector grid, the buffer distance, the smooth distance, and the sharp distance. The buffer distance describes the spatial extent of the surface reconditioning. The smooth distance is used to calculate how far the cells in the vector grid are dropped or raised to calculate the smooth grid. The sharp distance drops the cells in vector grid after creating the smooth grid. This is equivalent to digging a ditch or building a wall (Hellweger 1997).

The DEM file was used for the elevation grid, the DLG file for the vector coverage, 500 meters for the buffer distance, -10 meters for the smooth distance, and -300,000 meters for the sharp distance.

Once the new grid is calculated, change its name to gridagree.

Arc: copy elevgrid gridagree
Arc: kill elevgrid all

Fill the AGREE grid by making gridagree active, and clicking on the Hydrology Extension/Fill pull down menu. The new grid which is added to the view should be renamed to gridfilla.

Arc: copy fdr1 gridfilla
Arc: kill fdr1 all

The following figure shows the DEM and river network for the study region.

The following figure shows the DEM, the DLG and the "burned in" streams from the AGREE method at a threshold of 1000. The DLG river network is blue, and the "burned in" river network is red. The method agrees with the DLG file.

The bay system outline was developed from the River Reach Files (RF1) downloaded from the United States Geological Survey (USGS 1996) Internet site: http://h2O.er.usgs.gov/nsdi/wais/water/rf1.HTML (USGS 1996). The process for downloading these files are outlined in the report entitled Completing a Base Map for a Geographic Information System (GIS) Assessment of Nonpoint Source Pollution for the Corpus Christi Bay National Estuary Program (CCBNEP) in the chapter Downloading RF1 data from the Internet.

The bay segmentation was originally taken from the report entitled Corpus Christi Bay National Estuary Program, Ambient Water, Sediment and Tissue Quality of Corpus Christi Bay Study Area: Present Status and Historical Trends, Summary Report, (Ward 1996). It was decided that the original segmentation used in the report was too detailed for the total constituent loadings project due to the time that would be involved in trying to calibrate the equilibrium concentration model. The bay coverage was segmented and calibrated by Ferdi Hellweger and was downloaded from a database at the Center for Research in Water Resource at the University of Texas. A report entitled CCBPARAM - Corpus Christi Bay Water Quality Model Parameter Estimation to further explain this process.

Click on the Project Window and Open a new view. By going to the View/Properties, change the name of the view to Connect.

Add gridfilla to the view by clicking on , change the Data Source Type to Grid Data Source, highlight gridfilla, and click OK. Once the theme has been added, make the gridfilla grid active and check the box next to the theme to view the DEM of the area. Add the bay coverage to the view by clicking on , change the Data Source Type to Feature Data Source, highlight bay, and click OK. Make the bay theme active and check the box next to the theme to view the bay coverage of the area.

Download the Avenue script CONNECT from this site: connect.txt. The script is used to connect the river system to the bay system by creating a sink at the centroid of the bay polygon. However, this creates an instability at the sink and another program PICKLOAD should be run in order to extract the values from the centroid. This program will be used later in the report.

Click on the Project Window, click on the Scripts window and open a new script window (labeled Script1). From the Menu Bar, select Script/Load Text File. Click on connect.txt, and click OK. Using Script/Properties change the name of the script to CONNECT. The program will calculate a new grid called, gridconnect, which contains the modified elevations within the bay system and the original elevations outside of the bay system.

Compile the script by clicking on the check button, and run it on top of the Connect view. Make the View active, make the gridfilla and bay themes active, then click on the Connect script window and hit the Running man next to the check button to initiate the script. A message box will appear prompting a whether the temporary grid should be saved. Saving the temporary data sets is unnecessary. The newly calculated grid will be added to the Connect view called gridconnect. The new theme can be viewed by clicking on the box next to the gridconnect theme.

The following figure shows the connect river network and bay system.

A flow direction needs to be calculated on the connected grid by making gridconnect theme active, and clicking on the Hydrology Extension/FlowDirection pull down menu.

A detailed explanation of the Oregon data and the downloading procedure can be read in the report entitled Completing a Base Map for a Geographic Information System (GIS) Assessment of Nonpoint Source Pollution for the Corpus Christi Bay National Estuary Program (CCBNEP) in the section Obtaining Precipitation Data

Seann Reed of the Center for Research in Water Resources combined the two data sets for his research entitled Spatial Water Balance of Texas. The Oregon State University grid is used for the inland precipitation data, and the University of Delaware grid is used for the Gulf of Mexico precipitation data. The combined grid, obtained from Reed, are for the entire state of Texas, are in the Texas State Mapping System projection, and have a cell size of 5Km (Reed 1996).

For the CCBNEP research, the Grid Setwindow command is used to reduce the analysis window to the map extent of the buffered coverage. A smaller precipitation grid which contain the values from the large precipitation grid is defined within this window. Finally, the cells are resized to 100m using the Grid command Setcell. The procedure is as follows:

Arc: grid
Grid: display 9999
Grid: mape cchuctsms
Grid: linecolor 3
Grid: arcs cchuctsms
Grid: setwindow cchuctsms allann
Grid: setcell 100
Grid: precip = allann

Open a new View. Call this view Precip/Runoff. Using the button, add the grid precip from your directory. The coverages bay and outline should be added to the view in arc format. This is done in the Add Theme Window by clicking on the little folder next to the theme name. You will then see the theme expanded with a polygon, arc, labelpoint and node option. Highlight the arc option and click OK.

When the themes show up in the legend of the view, double click on the precip symbol to bring up the Legend Editor window. In the Legend Editor, change the color scheme of precip to get a ramped effect. Click on the arrow next to the Color Ramps box, scroll down, and select the Precipitation color scheme. When you are finished, select in the Legend Editor window and then close both the Legend Editor and Color Palette. Finally, select for the precip, bay and outline themes in the View Legend. The following figure shows the distribution of annual precipitation in the Corpus Christi Bay National Estuary Program's study region.

From this description of the grid, the range of precipitation values (min to max) was identified over the basin. The mean annual precipitation ranges from 539 mm/yr to 1034 mm/yr with a mean value of 701.556 mm/yr.

A detailed explanation of this process is discussed in Seann Reed's paper:
Spatial Water Balance of Texas. An exponential function was fit to the data in drier areas with mean annual rainfall less than Po. It was found that a linear function yields a better fit to the wetter watersheds with rainfall above Po.

The rainfall/runoff relationship is (Reed 1996):

Q = 0.00064 * P (mm/yr) * exp(0.0061 * P)    if P < Po
Q = 0.51 * P (mm/yr)  - 339.1                if P >= Po

Where: Q = runoff (mm/year)
P = precipitation (mm/year)
Po = 801 mm/yr

Download the Avenue script rogrid.txt which was written to calculate the runoff for the CCBNEP study site. The script inserts the precip grid into the rainfall/runoff equations discussed earlier to calculate the runoff grid (mm/yr). The procedure is as follows:

Click on the Project Window, click on the Scripts window and open a new script window (labeled Script1). From the Menu Bar, select Script/Load Text File. Click on rogrid.txt, and click OK. Using Script/Properties change the name of the script to Rogrid. Within the script there is a line that states to which file the result will be saved:

rogrid.savedataset("/home/ferdi/quenzer/trialrun/runoff".asfilename)

Edit this path name in the Script so that the result will be saved to the correct directory.

Compile the script by clicking on the check button, and run it on top of the Precip/Runoff view. Make the View active, then click on the Rogrid script window and click on the Running man next to the check button to initiate the script. A message box will appear prompting a selection the theme within the view (Precip) which will be used as the precipitation input.

Two error messages may appear when running the Rogrid script. This is a machine dependent error, and does not effect the results of the calculation.

The runoff grid should be clipped to the study area, so there isn't runoff in the Gulf of Mexico. The procedure includes converting the outline coverage into a grid using the Grid Polygrid command. Next, the cells outside of the outline are set to No Data and the cell within the outline are given the same value as the runoff grid.

Grid: outlineg = polygrid (outline)
Grid: runoff2 = con (oulineg ==2, runoff, outlineg)

Once the runoff grid is calculated and added to the view, change the legend colors the same way as the precipitation grid was changed. A different color scheme may be used if desired.

The following figure shows the expected runoff in the study region.

Using the Arc Describe command the range of runoff in the CCBNEP study area was found. Runoff values ranged from 11mm/yr to 176 mm/yr with a mean value of 43.723 mm/yr.

The land use/land cover files were obtained from the Texas Natural Resource Information System's Internet site: http://www.tnris.state.tx.us/ftparea.html under the heading Land Use/Land Cover (ascii data). These files use a Anderson Land Use Code classification system, in which major land use types are broken out into 9 categories:
1 = urban
2 = agriculture
3 = rangeland
4 = forest
5 = water
6 = wetlands
7 = barren land
8 = tundra
9 = ice and snow.
The second digit distinguishes subcategories of these principal categories, e.g.
11 = urban residential
12 = urban commercial
13 = urban industrial, etc.

This land use classification the United States was made in the late 1970's and land use has changed in the years since then, particularly as cities have grown. But, the LULC files are still the standard land use classification of the United States taken as a whole.

More information about the land use/land cover files is contained in the GIS Hydrology Home Page.

Once the data is downloaded from the Internet, imported, joined, and projected as discussed in the report entitled Completing a Base Map for a Geographic Information System (GIS) Assessment of Nonpoint Source Pollution for the Corpus Christi Bay National Estuary Program (CCBNEP) written in December 1996, in the section Obtaining Land Use Data.

For the CCBNEP study, only the land uses within the study region were used. The land uses outside of this boundary can be trimmed away using the clip command:

Arc: clip lucorptsms landuse outline poly

(This may take a couple of minutes.)

The Clip command uses the outline polygon coverage as a "cookie cutter". In English, this command says, "Clip the lucorptsms coverage with the outline template to create the landuse coverage which is made up of polygons.".

In Arcview, close the Rogrid Script window and the Precip/Runoff View, and open a new View. Use View/Properties to name it Land/Load, and add the outline polygon land use coverage of the study area as a theme. To display the individual land uses, double-click on the landuse theme symbol and, in the Legend Editor, change the Legend Type to Graduated Color and the Classification Field to landuse . Click on the Classify button and change the number of classes to 8. Click on the Value buttons to change the values of each category. Click on the Label buttons to change the labels of each category. Use the Fill Palette to color-code each of the symbols into land use categories. The following color scheme works pretty well, but feel free to use your own:

        Value   Label           Symbol Color

	0 	Unknown 	White (or transparent)
	10-17 	Urban  		Red
	20-24 	Agriculture 	Tan
	30-33 	Rangeland 	Yellow
	40-43 	Forestland 	Green
	50-54 	Water  		Medium Blue
	60-62 	Wetlands 	Light Blue 
	70-76 	Barren  	Gray

Close the Fill Palette and select in the Legend Editor before closing it. Now select for the landuse theme in the view and note where all the different land use categories in the basin are.

Below is what the land use coverage should look like:

Pollutant concentrations need to be assigned to each cell in order to calculate loadings of pollutants in the bay system. This study will use Event Mean Concentration values obtained from a previous CCBNEP analysis, Characterization of Nonpoint Sources and Loadings to the Corpus Christi Bay National Estuary Program Study Area (Baird, 1996). This study developed EMC values from water quality analysis performed at the Oso Creek and Seco Creek USGS Stream Gauges. The Oso Creek gauge is located west of Corpus Christi and represents agricultural land use. The gauges located on Seco Creek are northwest of Hondo, TX, and represent range land uses. EMC values for 18 pollutants were listed in this study and can be downloaded from the following text file emc.txt and is shown in the following figure.

The values in this table are typical concentrations of pollutants found in runoff water from each particular land use. The values are compiled from many field studies done by the US Geological Survey and other organizations.

Download and open the emc.txt file in a text editor window. Expand the window so that all columns in the table are visible. Note that there are different EMC values for each of the Urban Land Use codes (residential, commercial, industrial, transportation, mixed) but only one EMC value for the Agriculture, Range land, and Barren Land Use categories. Minimize the table when done viewing.

The research project is interested in estimating annual loadings of nutrients to the Corpus Christi Bay system, however this report will use total nitrogen loadings as an example.

Open a new file emca.dat in the text editor. This is an edited version of the emc table previously viewed which has two columns of numbers representing, respectively: landuse and total nitrogen.

Join these EMC values to the landuse polygon attribute table in order to associate them with land use. From the Arc prompt, open the Tables module and define the following attribute data fields in a new table called attrib.dat.

Arc: tables
Enter Command: define attrib.dat

      1

     10

This completes the definition of the table. To check that the tables were created correctly, look at the directory of available tables:

Enter Command: dir

 TYPE NAME                            INTERNAL NAME    NO. RECS LENGTH EXTERNL
------------------------------------------------------------------------------
  DF  ATTRIB.DAT                       ARC0000DAT                   47		14
  DF  PRECIP.BND                       ARC0001DAT	            4      	32    	XX
etc ....

A table template was created to which data can be added.

The values from the table have been prepared in a data file (emca.dat). These values can be added into the new table using the "add from" command in tables:

Enter Command: add from emca.dat
Enter Command: quit
Arc: list attrib.dat

Now use the Arc Joinitem command to join the EMC values to the appropriate polygons in the landuse coverage:

Arc: joinitem landuse.pat attrib.dat landuse.pat landuse landuse

  Joining landuse.pat and attrib.dat to create landuse.pat

In English, this command says, "join the landuse polygon attribute table (pat) with the corresponding data from the attrib.dat table, creating a new landuse.pat table. Use the landuse field as the item with which to relate the items from the two tables and add the new data after the landuse field in the landuse attribute table."

Now, convert the land use theme to a equivalent grid showing the Total Nitrogen concentrations in mg/l over the landscape. The Avenue script concgrid.txt has been written to calculate the new grid.

Download the concgrid.txt file. Open a new script and call it Concgrid. Click on Script/Load Text File and load in the concgrid script.

**Remember to change the pathname for the saved data files to fit your directory.**

Open the Land/Load view. Click on Analysis/Properties and check that the analysis extent is Same as EMC, the Analysis Cell Size is 100. Compile the script. Finally, run the script on top of the Land/Load view (This means click on the Land/Load View, click on the Concgrid Script Window, then click on the Running Man in the Script GUI tool bar). Pick EMC as the land use theme and tn as the concentration field. The concentration grid which is calculated by the script is added to the view.

The figure below shows the total nitrogen concentration on the land surface.

The loadings grid is found by multiplying the concentration grid by the runoff grid. Download the Avenue script loadgrid.txt which has been written for you to calculate the annual nonpoint source land surface loadings grid.

Open a new script and call it Loadgrid. Click on Script/Load Text File and load in the script loadgrid.txt .

**Remember to change the pathname in the script from /maidment to your directory for the saved data files to be placed your directory.**

Compile the Loadgrid script. Run the script on top of the Land/Load view (this means click on the Land/Load View, Click on the loadgrid Script Window, then hit the running man in the Script GUI tool bar).

Pick concgrid as the concentration grid and runoff as the runoff grid.

The script makes a conversion of mg/l (concentration grid) * mm/yr (runoff grid) to the Kg/d (loading grid). This is a more conventional way of looking at loadings. The conversion constant of 1/36525 is because the input units are mm/yr for runoff and mg/l for concentration, the output units are Kg/d for loading, there are 1,000,000 mg/Kg, each cell has a 10,000 m2 area, and there are 365.25 days/year.

In effect, for each cell, the load is computed as

Load = Runoff * Concentration * Cell Area (1)

The following figure shows the land surface loading for the Corpus Christi Bay National Estuary Program's study region.

WATER