OHD/HL - Threshold Runoff: Baseline

1. READ THIS IF you have already received ThreshR data
2. Recommended Procedures

2.1 Changes in unit graph parameter specification

2.1.1 Adding Cp and Ct or Ct2 Fields
2.1.2 Sources of information for Snyder parameter estimates
2.1.3 Populating Cp and Ct fields with data
2.1.4 Length Correction Option

2.2 Other Changes/Notes

1. READ THIS IF you have already received ThreshR data (AB, MB, MA, and CB), otherwise proceed to Section II.

If you have not started using ThreshR or you just wish to start again from scratch, I can resend your complete data set and you won't have to do any of the steps in this section.

• Change the filename of the grid "slopeff" to slopepct.  To do this, open a new empty project in ArcView. Open a new, empty View and choose File à Manage Data Sources and then use the Rename option in the File à Manage Data Sources dialog.

• The C Program for determining basin connectivity has been updated. Apparently I was using C++ libraries that are available here at OH, but not at the RFCs. I recoded the program so it only uses C. For those of you who have already downloaded your data, you can simply replace the executable "idupstream" located in your data directory with the file ftp.nws.noaa.gov/oh/gis/idupstream (use binary ftp transfer to download this file).  This program  is only needed if you choose the StreamLink basin delineation option and use the ThreshR à Determine Connectivity item with the C Program option turned on.

• Some data corrections have been made.

• If you have run ThreshR using v. 1.1, and you wish to continue working with the same main View you were using, you should run ThreshR-Utility à Clear View Tag and then rerun ThreshR à Setup before proceeding.

• I recommend that you run ArcView/ThreshR on one of your AWIPS Workstations ("ws") rather than on the data server "ds." Your ArcView/Spatial Analyst license should be a floating license so that you can run ThreshR on any of your AWIPS workstations.

• Follow the instructions from Workshop 2 using rfcbound.shp (in lieu of wsusq.shp) to define your analysis area and noting the changes specified below. The results will yield gridded ThreshR values representative of basin sizes ranging from 20-35 mi². You may also wish to run the same procedures (still using the Headwaters1 basin delineation option) with larger basin sizes to get a feel for how ThreshR values vary with basin size. A range of cumulative basins (headwaters + non-headwaters) can be examined using the Streamlink basin definition option, although running the Headwaters1 option twice may be quicker than running the Streamlink option once. If you run the Streamlink option, it is probably safest not to select any of the advanced options although these should work.  Although technically feasible, it is not advisable to attempt to define basin boundaries smaller than 20 mi² using the DEM resolution that is being delivered with AV-ThreshR v. 1.11. As we discussed at the workshop, ThreshR values are somewhat sensitive to basin drainage area, yet there is no consistent trend across the country as to whether ThreshR values increase or decrease with drainage area.

The interface for unit graph calculations has been simplified to allow for more flexibility in specifying Cp and Ct values. In addItion, the option to use two different formulations for computing time to peak (lag) is also included. These two lag equations were discussed in Lecture 4 at the July 25-27 Workshop, but are reproduced here for clarity in discussion.

Equation 1. 

(Snyder, 1938.)

Equation 2. 

The new dialog interface for unit graph calculations looks like this:

In order for the calculations to proceed, a field named "cp" and either "ct" (If Original Lag Equation is selected) or "ct2" (If USACE Lag Equation) must exist in the Subbasin Theme (shd3.shp in this case). If the Calculate button is clicked and these fields cannot be found then the following message will appear: "Must add cp and ct or ct2 field to shd1.shp."

2.1.1 Adding Cp and Ct or Ct2 Fields

• Make the subbasin attribute table active and select Table à Start Editing
• Select Edit à Add Field. Define a numerical field (Width = 5 and Decimal Places = 3 should suffice for Cp, Ct, or Ct2.

• Repeat for Ct or Ct2

2.1.2 Sources of information for Snyder parameter estimates

• Literature (see ranges below)
• Back calculate from know unit graph as done in Workshop 1.
• Call your Corps of Engineers District Offices

Snyder, 1938
Cp: 0.56 - 0.69
Ct: 1.8 - 2.2
Geographic Region: Appalachian Highlands

Buil, 1968
Cp: 0.77 - 2.45
Ct: 1.65 - 7.99
Geographic Region: Lower Missouri River in Kansas and Missouri

Miller, Kerr, and Spaeder, 1983
Cp: 0.24 - 0.67
Ct: 1.01 - 4.33
Geographic Region: Pennsylvania

HEC, 1989
Cp: 0.22 - 0.82
Geographic Region: Allegheny River Basin

USACE, 1959
Ct2: 0.35 for valley drainage, 0.72 for foothill drainage, 1.2 for mountain drainage.
Geographic Region: In the vicinity of Los Angeles, CA

Carpenter and Georgakakos, 1993
Cp: 0.35 - 0.51
Ct2: 0.39 - 0.47
Geographic Region: Two streams in Iowa and one in Virginia

Examples:

Define values for selected records using the Field Calculator . The field for which you wish to set values must be active. [Cp] was active for the case shown below in which all selected subbasins are assigned the value Cp = 0.5.

Subbasin selections can be made either graphically using  or by using a tabular query . The tabular query option could be used, for example, if you wish to set different ct or ct2 values for different ranges of channel slopes (CHSL). This might be one way to distinguish between valley bottom (ct2 = 0.35), foothills (ct2 = 0.72), and mountains (ct2 = 1.2). The dialog below illustrates the selection of all subbasins with CHSL (channel slope) less than 30 ft/mile. Use of this dialog could be followed by another Field Calculation () to set values for selected basins. Graphical selection would be useful if you have an idea about parameter ranges for specific river basins.

Another useful button when setting table values is , which toggles between selected and unselected records.

If you have been able to prepare a grid of Cp or Ct values like we used in Workshop 2 then you may wish to populate the Cp, Ct, or Ct2 fields using a grid. The menu item ThreshR - Utility à Polygon Average from Grid allows you to do this.

When you are done editing your Cp and Ct attribute fields, click Table à Stop Editing.

2.1.4 Length Correction Option

An option to turn off or on the empirical length correction option is now available. This option is turned on by default. These corrections are used due to limitations with the DEM resolution (400-m cells) being used. Lengths and slopes measured from the 400 m DEM are adjusted in an effort to be closer to those used by the USGS to develop empirical relationships. The empirical adjustment equations are similar to (but not identical to) those presented in Lecture 6 from the July 25 Workshop).

2.2 Other Changes/Notes
 

• When using ThreshR --> Q2, Q5 etc., you should now get a message if the equations for the return period you have selected are missing for a particular state. For example: In eastern Colorado, the USGS has not published the 2-year equations. The message you get is: "2 Year Equations May be Missing for CO"

• Although we have expended considerable effort in checking the regression equation database, errors still may crop up.  In several cases, errors in the data tables can be detected by displaying ThreshR results and looking for unusually high, unusually low, or missing values (-1).  Let us know if you see any problems.

• -1 is now assigned to regions without available regression equations for the return period selected (rather than 0).

• A change to the Setup program now requires the user to specify the output directory as well as the input directory. This automatically sets the project working directory.

• I'd be interested to know whether you feel that the Q5 may be a better representation than Q2 for "flooding flow" in your area. I'm hoping to make some comparisons of Q2 and Q5 with known bank full flows (i.e. at locations with rating curves) to get a better feel for this.

• Known data problems are described in the data release notes.

• In some regions, the USGS regression equations may be applicable only to certain ranges of area or other parameters. Checking these applicability limits is not currently built into the ThreshR software.  In areas I have tested, there are few problems with the basin sizes of interest for ThreshR size basins. A few potential problem areas are mentioned in the data release notes, although this list is not exhaustive.

• For the parameter values added to your subbasin attribute table, constants added or subtracted, or unit conversions required by the USGS equations have already been made. Thus, use caution in interpreting these values.  The CHLN and CHSL values listed in the table do not reflect any empirical length adjustments.  These adjustments are hardcoded into ThreshR --> Compute Q2, Q5, etc. and ThreshR --> UG Peak Flow.

Buil, J. A, "Unit Graphs for Nonuniform Rainfall Distribution," Journal of the Hydraulics Division, Proceedings of the American Society of Civil Engineers, 235 - 257, January, 1968.

Carpenter, T.M., and K.P. Georgakakos, "GIS-Based Procedures in Support of Flash Flood Guidance," Iowa Institute of Hydraulic Research (IIHR) Report No. 366, November 1993.

Hydrologic Engineering Center, "Real-Time Flood Forecasting and Reservoir Regulation for the Allegheny River Basin," Project Report No. 89-5, June 1989.

Miller, A.C., S.N. Kerr, and D.J. Spaeder, "Calibration of Snyder Coefficients for Pennsylvania," Water Resources Bulletin, AWRA, pp. 625 - 630, August 1983.

Snyder, F.F., Synthetic unit-graphs, Trans. Am. Geophys. Union, vol. 19, pp. 447-454, 1938.

U.S. Army Corps of Engineers, Flood hydrograph analysis and computations, Engineering and Design Manual, EM 1110-2-1405, U.S. Government Printing Office, Washington, D.C., August 31, 1959.