Evaluating Effects of Channel Properties in GSSHA

CEEn-531 Dr. Nelson


After considering all the losses from the surface, overland flow passes through the streams to the outlet of the watershed. In most of the natural streams, the cross section of the stream and the flow are both irregular and non prismatic. In many cases such cross sections are approximated with some prismatic sections like trapezoidal, rectangular or circular. In this assignment, we will compare how this approximation affects the results. 

In previous assignment on GSSHA you varied surface roughness based on the land use for the overland flow. But, in this assignment, you will vary the roughness of channel and analyze the effects on the outflow.

Model Set up

Here you will use two different models, one with density of streams as 0.2 sq mile and another with 1.0 sq miles. In order to review the process of building a model "from scratch" you need to create two separate models using the data from the links below.  This is the same data we've been using all semester, it's just collected here for your convenience. 

The stream density defines the number of streams in your model. Drainage density is defined when you delineate your basin from the DEM data using TOPAZ in WMS.   In the "Delineate Watershed" dialog of the wizard, you will need to enter 0.2 mi^2 or 1.0 mi^2  in the field labeled "Stream threshold value"

1. Comparison of stream density with trapezoidal channel

Here you will see how the stream density affect the results. So, you will compare the models with stream density of 0.2 and 1.0 sq miles. Both these models will have trapezoidal sections.

Create index maps for soils and land use from the land use and SSURGO shape files you can download.

Use the same GSSHA mapping table as you did in the surface runoff assignment (Link).

Define the channel as follows:

The stream cross sections are not of same size all over the basin. As you move downstream, streams increase in cross section because of more accumulation of water. So, Define the channel as follows

For Low density model (1 sq mile) follow this link.

For High density model (0.2 sq mile) follow this link.

Compare the results from low density model with the results from high density model.

2. Comparison of trapezoidal and natural cross section in the model with drainage density of 1.0 sq mile.

Changing the Stream cross section to Natural Section:

In the low density model you used for part 1 of this assignment, now you will change the stream cross section to more realistic 'break point cross section'. Download this file which has the cross section information and follow this link to enter the cross data. 

Your low density model should have the following data. It is already defined just VERIFY if everything is entered properly.

Compare the results from this simulation with the results from Part 1 low density model (i.e. model with natural cross section and trapezoidal cross section)

3. Sensitivity of Manning's n for the stream:

As the flood hydrograph flows through the streams, the roughness of the channel has a significant role on the outflow hydrograph. In this assignment, we will see how sensitive is Manning's n to the model.

Open the low density model you used for part 1 of this assignment (the model with trapezoidal cross section). Now use the following values of Manning's n for the model and rerun it. You have already run the model with Normal values (which is the model you developed in part 1 of this assignment). So, you have to create two different scenarios of Manning's n variation.

Arc Number

Manning's n 

Normal Value

Increased by 50%

Decreased by 50%





















Save the model and rerun it for both cases

Compare the results from Normal n value (Part 1 of this assignment) with both 50% increased and 50% decreased cases.

To turn in Now:

To turn in with Assignment 26: