Evaluating Initial Abstraction and Loss Models in HEC-HMS

Dr. Nelson


Model Setup

  1. Delineate Judy's Branch Watershed
    You may use an existing delineation, but it is probably easier to start from scratch with your DEM Data

  2. The NRCS provides county-level (SSURGO) and state-level (STATSGO) soils data that includes hydrologic soil group attributes.  We want you to visit the NRCS sites through the GSDA web page and you may choose to download the data for Judy's Branch (Madison County). These data sets are large and can take some time to download so you can use my copies of the data previously downloaded SSURGO, and Illinois STATSGO. You should know enough about the NRCS site that you can download data for your projects later on.

  3. Load the shapefiles into WMS and then right-clicking on the layer and choose the option to join the NRCS data (Note that this option is not available in the Wizard currently).

  4. Download the land use shape file(s) for Madison County (Judys Branch is in this county of Illinois).  You can follow the links on the GSDA website for downloading landuse (use the WebGIS link which is the first one in the list).  Click on the button under Web GIS, click on Land use, select coordinate system, select state and select county. These files are small enough that each team can download separately.

  5. Create WMS soils and land use coverages from the shape files downloaded.

    Notes: You will have two separate soil coverages, one from the SSURGO data and one from the STATSGO data that you will use to compare.  When you load the shape files you will discover that they are Geographic NAD83 and you will have to convert to UTM NAD83 Zone 16 (or 15 depending on what you have beenw working in). Be sure to map soil polygons to the soil coverage and land use polygons to the land use coverage.  It is also recommended that you select only the land use and soils polygons overlapping the watershed (particularly for the case of SSURGO data which has thousands of polygons, but many of which are not over your watershed).

    For the land use data you will need two shapefiles.  You can load both, convert the coordinates and then select the polygons overlapping the watershed and WMS will map the appropriate polygons to a single coverage from both shapefiles.

  6. You can use Calculators | Compute Coverage Overlay (in the hydrologic modeling module) for the drainage and land use coverages to determine which land use types overlap your watershed boundary.  For these land use types determine CN values for all four hydrologic soil groups for the three antecedent moisture conditions (pg. 158 Wanaliesta - CE 431 Hydrology Text Book and WMS TR55 documentation).  You can define these CN values in WMS from the attributes of the land use polygons or importing a text file that you create (using an existing one as a template).

  7. Use the Compute Coverage Overlay for the drainage and soil coverages to determine which hydrologic soil groups overlap your watershed boundary.  For these hydrologic soil types determine appropriate Green and Ampt Parameters. The Green and Ampt parameters are discussed in page 42 of the HMS Reference Manual. You can also see this document which you will find to be useful.


In this assignment we will be looking at different parameters in HMS that affect infiltration and perform some of the sensitivity analyses between different methods, and parameters used to define them. You may use any starting and ending dates but make sure that your simulation runs long enough to capture the complete hydrograph..

 Processes:

1. Curve number sensitivity

Here you will see how the variation in curve number affects the runoff. The CN variation is done on the basis of antecedent moisture condition of soil. (pg. 158 Wanaliesta - CE 431 Hydrology Text Book, Appendix 1 of the HMS reference manual, and WMS TR55 documentation). You will compare the runoff hydrographs from different CN cases.


2. Initial abstraction sensitivity

Here you will see the effects of initial abstraction on runoff hydrograph. Use normal condition CN and simulate with Ia of 0, 0.1S, 0.2S, and an appropriate value as suggested on page 39 of your reference for estimating initial abstractions. You will compare the hydrographs from different initial abstractions.


3. Soil Type sensitivity

You have downloaded two different soil types with the county level and state level resolutions. In this part we will see how the soil resolution affects the runoff. So, determine CN from i) SURGO soil and ii) STATSGO soil. Use these curve numbers and simulate. you will compare the hydrographs from two different soil types.


4. Initial and Constant-rate Method (Single Basin)

This is the most basic loss rate model.  You can estimate parameters relatively easily, however it is difficult to know the right values without performing some kind of calibration.  We'll use this model to compare to the SCS and Green and Ampt methods.


5. Green and Ampt Method (Single basin)

Another method to consider the infiltration is the Green and Ampt method. The parameters are determined based on the soil and land use information. As already explained use the references stated above and estimate Green and Ampt Parameters and simulate. You will compare the results form Green and Ampt with the results from Single basin HMS (With SSURGO soil Normal condition, Ia = 0.2S).  Be sure to save this model as you will use it to adjust % impervious in a comparison with GSSHA in the next assignment.

Note:  If you get 0 runoff try adjusting moisture deficit to 0.0 (saturated), or decrease hydraulic conductivity until you see some runoff - be prepared to comment on what this means?


6. Spatial variation sensitivity

In this section, you will divide the watershed into two sub basins. Your watershed should look somewhat like this. You have to route the hydrograph from right basin to the outlet. Use the following methods.


7. Gridded SCS Method with MODClark

Finally, you will use the distributed model to see the effects of lumped and distributed parameter models on runoff. Use 100 by 100 MODClark grids and compare the results with Single basin and two basin HMS models with SSURGO soil and Ia = 0.2S.


To turn in with modeling exercises:

To turn in with losses review (but at least think about these now):


******