TOPMODEL
 

    TOPMODEL is a variable contributing area conceptual model in which the predominant factors determining the formation of runoff are represented by the topography of the basin and a negative exponential law linking the transmissivity of the soil with the distance to the saturated zone below ground level (Franchini et al, 1995). The original steady state TOPMODEL formulation is essentially based upon two equations written in lumped form at the cathment scale: the first one representing the overall mass balance and the second one representing a storage-discharge relationship. These equations are generally derived by the process of averaging the point equations over increasing size pixels up to the catchment scale (Todini, 1995).
The version of TOPMODEL I will be working with has been adapted by Dr. Tarboton and applies the classical TOPMODEL over each subwatershed in a river network.


Term Project Objectives


 

1. Collect and prepare data to ruun TOPMODEL over the Mt. Sterling River Basin.
2. Calibrate Model using multi-objective criteria
3. Compare  model output to existing streamflow and  soil moisture data.

Data Preparation

Stream network and Elevation data:

 

• Prepare DEM
• Fill sinks in DEM
• Run TAUDEM for Stream Network and Watershed delineation

Elevation data was downloaded from the USGS data distribution website; http://seamless.usgs.gov/ . NED  data sources have a variety of elevation units, horizontal datums and map projections.Hence a number of processing steps have to be undertaken to allow for correct analysis using Arc/Info software. The processing steps undertaken in this project are as outlined below.

 

1. Prepare the DEM

            The downloaded data was in Geographic coordinates, and therefore the first step was to project the data to UTM zone 17. I created a folder named 'Ohioutm' in my working directory. I then proceeded to open ArcToolbox , project wizard, (coverages and grids) and selected 'project my data to a specified coordinate system'. I chose NAD 1983  as the datum for my dataset. I then selected Ohioutm as my output dataset with 'cubic' as the desired resampling method and saved it as a grid named NED. I then opened ArcMap and added Ohioutm\ned to the project while clicking ok to allow project to build pyramids.

   2. Filling Sinks in the DEM
                   This was accomplished using the Arc/Info fill command. The following sequence of Arc commands was executed in commandline Arc/Info Workstation.
 

• w ( to see which folder or geodatabase is set as the workspace)
• w i:Ohioutm (to change workspace to where the data is)
• lg (to list grids in the workspace)
• grid (to invoke the grid function subset)
• fill nde nedfel ( to fill sinks in the grid ned, and write output to nedfel)
• quit ( to leave grid)
• quit (to leave Arc).
   

This procedure creates a new pit filled grid, nedfel, which is then used as the base DEM for TauDEM.

   3. Run TauDEM for Watershed and Streamnetwork Delineation
       
 

• Once the fill sinks operation has been completed, the prepared DEM, Nedfel, can now be added to TauDEM and 'Do all watershed and network delineation steps' selected.
• Before this can be accomplished however, a shape file containig outlet points needs to be created. The procedure followed is as outlined below;

• Check stream delineation settings and ensure that the threshold of the cells required to create a first order stream should be large enough that you do not have too much detail, but small enough that streams are not significantly different.
• Check with 'constant drop analysis' to determine drainage density based upon topographic texture and different channel delineation thresholds.

( After Dr. Tarboton and The Utah Water Research Laboratory, 2000).

Vegetation data:

Land Use Land Cover data was downloaded in ASCII file format from USGS, EROS website. http://edc.usgs.gov/products/landcover/lulc.html
This data consists of historical land use and land cover data based primarily on the manual interpretation of 1970's and 1980's aerial photography. The following data transformation procedure was used to  prepare the original ASCII file for input into  ArcMap  project:
 

• Open ArcToolbox, conversions and convert ASCII file to grid.
• Add converted grid to ArcMap project and using Spatial Analyst, set options, working directory, Analysis output = same as data frame, Extent, same as Base DEM.
• Go to Spatial Analyst, Raster Calculator and evaluate 'veg' in calculator box.
• Right click on output calculation to make permanent.

The vegetation in the area is predominantly Row Crops with some Pasture/Hay.

Soils Data:

This was obtained from STATSGO soils data for Ohio.This dataset is a general Soil  association map developed by the National Cooperative  Soil  Survey and  distributed  by the  Natural Resources Conservation Service  (NRCS).  It consists of  a  broad based inventory of soils  and  non soil areas that occur in a repeatable pattern on the landscape and that can be cartographically shown at the scale mapped. The soil maps from STATSGO are compiled by generalizing more detailed soil survey maps. The spatial component of the STATSGO database is archived and distributed in Arc/Info export file format *.e00.

The soils data processing involved the following steps:
 

1. 
    

• Unzip STATSGO  soils data for Ohio
• Open ArcToolbox and export to shapefile, export from interchange file
• Define projection; STATSGO data is provided with the following projection parameters:
   
  • Projection                      Albers equal area
  • Units                              meters
  • 1st standard parallel        29  30  00
  • 2nd standard parallel       45  30  00
  • Central meridian             -96  00  00
  • Latitude of Origin            23   00  00
  • false Easting                      0.00
  • false Northing                   0.00
• Add this data to ArcMap project.

   
     2.   Create Tables to Relate Grid Layer toPolygon Data by Map Unit Identifier (MUID)
 

• Add both the soil texture grids and soils shape file to a viewer,
• Using Spatial Analyst, summarize by zones from the soils shape file. Pick the field 'MUID' to define zones then select soil texture grids to summarize. repeat for all layers.

This process extracts the majority soil-code from each raster layer of soil texture that occurs in each polygon of the STATSGO shape file.
 

• Edit the soil tables

Since the majority field gives the texture class for the matching polygon, delete all fields except MUID, Zone_code and majority.

Use Table/Properties to set thee alias for the field majority to designate the texture class depth range. Now each majority column will be labled by the layer to which it corresponds. Change the name of the majority field to the layer name with which it corresponds.

Join all layers into one table

Join the resulting table to other tables with MUID as the join field. Export resulting table as delimited text, Ohtxttab.txt.

Join the soil layer table to the soil polygon layer

Add the Ohtxttab.txt to the ArcMap project. Join the table using MUID field to the attribute table of the soils shape file.

Convert Polygon to Raster

Set Spatial Analyst options such that Analysis Output will be saved in coordinate system of the active data frame and extent and cell size are the same as the base DEM. Using Spatial Analyst, convert features to raster from soils shape file, using the field 'zone_code' from joined attribute table. Output result as as Ohiosoil.

Precipitation Radar Data

NEXRAD Stage III radar data was used in this modeling exercise. The NEXRAD Stage III products offer high quality hourly rainfall estimates with an approximate resolution of 4Km by 4Km cells. This data provides much more information about how weather systems behave in space and time than can be inferred from raingauges alone. Stage III data was created specifically for the NWS river forecast centers which need rainfall estimates over a much larger area than covered by an individual radar. Stage III mosiacs together Stage II estimates from multiple radars onto a subset of the national HRAP grid covering the river forecasting area of responsibility. (http://www.nws.noaa.gov/oh/hrl/papers/ams/ams9-1.htm).
Hourly NEXRAD Stage III products are in binary format and have the following naming convention; 'xmrgMMDDYYhhz'. Each day, these xmrg products are compressed then tarred into a daily file in the form 'SiiiMMDDYYRFCID.tar'. At the end of the month these daily files are tarred into a monthly file and posted on the web in the form 'SiiiMMYYRFCID.tar', where:
 

The data were downloaded from the following site (http://dipper.nws.noaa.gov/hdsb/data/nexrad/nexradiii.html)

Make_raindat program is used to read and convert raw radr data to TOPMODEL ready data. It promts the user to select a beginning and ending monthly dataset. It takes the compressed monthly data, decompresses to the daily data,  decompresses to the hourly data, reads the xmrg file from binary and outputs a beginning and ending month.dat file. this output file should be renamed rain.dat and copied into the model runs directory. make_raindat.exe creates latlong.txt which is then converted into an event theme in ArcCatalog, allpoints.shp. This shape file is then used to create a buffer over the watershed of interest and from this, radar_pts.shp is created. The .dbf from this file is then exported to a txt file radar_pts.txt. When this file is  present, make_raindat.exe will read these locations and write output for only these locations.