Guidelines for AGNPS-GRASS Input Interface

The major objective of the AGNPS-GRASS input interface is to minimize the user interaction in preparing the input data for the AGNPS model and to minimize the number of user supplied/developed GIS database layers. Of the 22 different data required by the model for each cell (appendix 2), the interface will prepare the input data with only 8 (appendix 2) basic GIS database layers supplied by the user and with minimal user interaction. There are 5 (appendix 2) parameters needed for the whole watershed which will be obtained from the user.

Name of the AGNPS-GRASS Input Interface Program: aggrs_input

Procedures to follow while executing the Input Interface:

First start GRASS 3.1 from a shelltool window with proper mapset where the data resides

Execute aggrs_input program from the shelltool window where GRASS is running

A Watershed Input menu will appear on the shelltool window requesting the following data/information from the user:

Watershed Description: Enter a description about the watershed. This information optional and will appear in both input and output of the AGNPS model run.

Rainfall Amount in inches: Enter the total amount of rainfall in inches.

Erosivity Index Value: Enter the erosivity index value for that particular storm/rainfall event.

Enter the cell size in meters: The length of the side of a cell in meters is entered. The square of the cell size will be the area of each cell.

Enter a file name to save in AGNPS format: Enter a file name with .dat as the extension to save the AGNPS model input file created by the AGNPS-GRASS input interface tool. This will be stored in the current working directory.

Enter the watershed name: The name of the watershed should be entered here. All the input layers should have the same watershed name with the proper extension (as X in appendix 1). Please refer to appendix 1 for the valid extensions and valid category labels/values that are allowed for each layer.

Then hit the Esc key to continue.

Error Messages: If any of the category labels/values for the input layers is not valid, an error message is printed and the program quits. Run support on the layer where the category label is wrong to correct it. The program support helps to modify the supporting files of a map, which include header, category, color, and history. The wrong/misspelled category labels can be corrected by running support on the layer to be corrected and choosing the category option. If there is any problem with the flow direction layer, error messages are printed for those cell numbers that had problems and the nature of those problems (like circularity, sinks or holes, or flow directions pointing at each other). Use Dedit on the aspect map to correct the problems. The Dedit program helps to display and edit the aspect map with arrows on it, which should enable one to correct the flow direction on a cell by cell basis.

For each run of this input interface tool, a new temp_cell_num layer is created and stored in the current mapset. In this map, the cells are numbered from top to bottom and left to right. This is a useful layer for editing and interpreting the error messages due to conflicting flow direction for any particular cell. While using the Dedit program to correct the errors, this map allows one to locate the cell number where conflicts occur.

Once the input interface completes its run without errors, it saves the data in the AGNPS format. The AGNPS model can be executed with this data. The name of the executable file program to run the AGNPS model is agrun. Type agrun from the shelltool window where GRASS is running. The program will ask for the input file name. Enter the name of the file that was saved by running the AGNPS-GRASS input interface, with the .dat extension. The model will output the results of the run with a .nps extension. Both input and output ASCII files are needed to execute the AGNPS-GRASS output interface tool (the Visualization tool).

Appendix 1

Watershed boundary
(X.wshd)
Should have category values greater than 0 within the watershed and 0's outside of the watershed boundary.
Elevation
(X.elev)
Elevation data in meters should be the category values for each cell. Also make sure at least 1 row and column of data around the watershed boundary exists to estimate slope and aspect at the boundary of the watershed.
Aspect
(X.asp.cell_size)
The aspect map should have values between 1-8 as category values and should not have any sinks, circularity or more than one outlet. Cell_size is the number that the user entered for question 4 in the interface input menu. For example, if the watershed is modeled for a 100 meter cell size, then the aspect map name should be X.asp.100
Soils
(X.soils)
If any of the soil attribute (K factor, clay, sand or hydrological soil group) maps do not exist, then the program looks for X.soils map to extract information from the Soils-5 data base. (currently this is not working fully).
USLE K factor
(X.K)
The USLE K factor map should contain the value of the K factor as the category label for each cell. This is obtained from the soil type map.
Hydrological soil group
(X.hyg)
The category labels allowed are either A, B, C, or D and should appear as category labels for each cell. This is obtained from the soil type map.
Sand
(X.sand)
The percentage of sand content should be the category label for each cell. This is obtained from the soil type map.
Clay
(X.clay)
The percentage of clay content should be there as a category label for each cell. This is obtained from the soil type map.
Land use
(X.luse)
The land use map category labels allowed are fallow, row crops, small grain, legumes, rotation meadow, close-seeded legumes, pasture, range, meadow, woods, woodland, grass waterway, hard surface, farmsteads, roads (dirt), water, or marsh.
Management Practice
(X.mgpr)
The management practice map category labels allowed are straight row, contoured, or contoured and terraced.
Nutrient
(X.nut)
The cell value should correspond to the level of fertilizer application, i.e 1, 2, 3 or 0. Where 1-low, 2-medium, 3-high and 0-urban, water or marsh. (refer fertilizer level table).
Machinery
(X.mach)
This map is used to find the fertilizer availability factor for the top 1 cm depth of soil. The category labels allowed are large offset disk, moldboard plow, lister, chisel plow, disk, field cultivator, row cultivator, anhydrous applicator, rod weeder, planter, smooth, or no till. The category label for urban, water, marsh or farmsteads can be no till or smooth (refer fertilizer availability factor table).
Channel slope
(X.chsl)
A channel slope length map has to be prepared by the user. Percentage of channel slope should be the value of the category in that cell. If this layer is missing, the user has the choice to either enter the name of layer which has the channel slope values or can accept the default value of 50% of the overland slope value as the channel slope value (recommended by the AGNPS manual).
Slope length factor
(X.slen)
A slope length factor layer in feet has to be prepared by the user for each cell. The value should be the value of the category in that cell.
USLE C factor
(X.C)
The USLE C factor map should contain the value of the C factor as the category label for each cell.

Fertilization level

----------------------------------------------------------
|Level of      | Assumed fertilization (lb/acre) |       |
|fertilization |          N         P            | Input |
|--------------------------------------------------------|
|None                     0         0            |  0    |
|Low                     50         20           |  1    |
|Medium                 100         40           |  2    |
|High                   200         80           |  3    |
----------------------------------------------------------

Fertilizer availability factors according to tillage practice

--------------------------------------------------------------
Tillage practice          Fertilizer availability factor (%)
--------------------------------------------------------------
Large offset disk                      40
Moldboard plow                         10
Lister                                 20
Chisel plow                            67
Disk                                   50
Field cultivator                       70
Row cultivator                         50
Anhydrous applicator                   85
Rod weeder                             95
Planter                                85
Smooth or no till                     100
--------------------------------------------------------------
Appendix 2: AGNPS cell input parameters
box;
l.
1. Cell number
2. Number of cell into which it drains
3. SCS curve number
4. Average slope %
5. Slope shape factor
6. Average field slope length
7. Average channel slope
8. Slope length factor
9. Mannings roughness coefficient for channel
10. Soil erodibility factor (K) from USLE
11. Cropping factor (C) from USLE
12. Practice factor (P) from USLE
13. Surface condition constant (based on land use)
14. Aspect (one of 8 directions indicating drainage from cell)
15. Soil texture (sand, silt, clay, peat)
16. Fertilization level (zero, low, medium, high)
17. Incorporation factor (% fertilizer left in top 1 cm of soil)
18. Point source indicator (indicates existence of a point source input within a cell)
19. Gully source level (estimate of amount, tons, or gully erosion in a cell)
20. Chemical oxygen demand factor
21. Impoundment factor (indicating presence of an impoundment terrace system within the cell)
22. Channel indicator (indicating existence of a defined channel within a cell)

The 8 basic input GIS layers required for extracting the data by the AGNPS-GRASS input interface tool are:
  1. Soils
  2. Elevation
  3. Land use
  4. Management practice
  5. Fertilizer or nutrient inputs
  6. Type of machinery used for land preparation
  7. Channel slope
  8. Slope length factor
Five pieces of data are required for the total watershed:
  1. Watershed identification/description
  2. Precipitation (inches)
  3. Erosion Index (EI-value) for that storm/rainfall event
  4. Area of each cell (acres)
  5. Outlet cell number
An important implication of this set is that AGNPS does not accommodate non-uniform storms, i.e. it uses a lumped modeling approach for its rainfall.

For each watershed element, AGNPS requires the following 22 input data values (its distributed parameter information):

  1. Cell number
  2. Number of the cell into which it drains
  3. SCS curve number
  4. Average land slope (%)
  5. Slope shape factor (uniform, convex or concave)
  6. Average field slope length (feet)
  7. Average channel slope (%)
  8. Average channel side slope (%)
  9. Mannings roughness coefficient for the channel
  10. Soil erodibility factor (K) for USLE
  11. Cropping factor (C) for USLE
  12. Practice factor (P) for USLE
  13. Surface condition constant (factor based on land use)
  14. Aspect (one of 8 possible directions indicating the principal drainage direction from the cell)
  15. Soil texture (sand, silt, clay, peat)
  16. Fertilization level (zero, low, medium, high)
  17. Incorporation factor (% fertilizer left in top 1 cm of soil)
  18. Point source indicator (indicates existence of a point source input within a cell)
  19. Gully source level (estimate of amount, tons, of gulley erosion in a cell)
  20. Chemical oxygen demand factor
  21. Impoundment factor (indicating the presence of an impoundment terrace system within the cell)
  22. Channel indicator (indicating existence of a defined channel within a cell)