Maryland

The reports below document the regression equations available in StreamStats, the methods used to develop the equations and to measure the basin characteristics used in the equations, and the errors associated with the estimates obtained from the equations. Users should familiarize themselves with these reports before using StreamStats to obtain estimates of streamflow statistics for ungaged sites.

Dillow, J.J.A., 1996, Technique for Estimating Magnitude and Frequency of Peak Flows in Maryland: U.S. Geological Survey Water Resources Investigations Report 95-4154, 55 p.
Carpenter, David H., Hayes, Donald C., 1996, Low-flow characteristics of streams in Maryland and Delaware: U.S. Geological Survey Water Resources Investigations Report 94-4020, 113 p.

Interactive Map

StreamStats has been implemented only for the Gunpowder and Patapsco River Basins in Maryland as a demonstration project. This area is enclosed by an aqua-colored line on the map of Maryland that is shown initially in the StreamStats user interface. A black line separates a yellow hatched area in the southeastern part of the implemented area from the rest of the area. The hatched area is part of the Western Coastal Plain Region of Maryland, which extends beyond the boundaries of the Gunpowder and Patapsco River Basins. Low-flow regression equations were not developed for this Region. Peak-flow regression equations were developed for this Region, but they were not implemented in StreamStats for the reasons described below.

The percentage of forested land area is used as an explanatory variable in the peak-flow regression equations developed by Carpenter and others (1996). The National Land Cover Database of 2001 (NLCD - http://www.mrlc.gov/mrlc2k_nlcd.asp) was used to measure the percentage of forested land area for StreamStats. It is not known how the values that appear in Carpenter and others (1996) were determined. Comparisons were made between the published percentages of forested land area and the values measured by StreamStats for 25 stations that were used previously to develop the peak-flow regression equations. Of the stations, 21 were in the Piedmont Region and 4 were in the Western Coastal Plain Region. The comparisons showed large differences between the two sets of measured values. The differences in results may be attributable to the differences in the source material and methods used for the measurements, but they also may be due to actual changes in land use between when the original source material (probably areas shown as green on USGS topographic quadrangles) and the NLCD were developed.

The comparisons for the Piedmont Region yielded an average difference between the published values and the values determined by StreamStats of +0.8 percent, with differences for individual stations ranging from +72.6 to -42.7 percent. The peak-flow regression equations for the Piedmont Region of Maryland are not highly sensitive to the entered values of the percentage of forested land area. The small average difference in the values indicates that, on average, there is little bias in the StreamStats measurements. In addition, estimates of the 100-year flood flows determined from the equations were consistently within 10 percent of the published values. As a result, the peak-flow regression equations have been implemented in StreamStats for the Piedmont Region, but actual standard errors of prediction for the estimates are likely larger than those provided in the StreamStats output.

The comparisons for the Western Coastal Plain Region yielded an average difference of -55.8 percent, with differences for individual stations ranging from -29.2 to -80.0 percent. The peak-flow regression equations for the Western Coastal Plain Region were much more sensitive to the entered values of the percentage of forested land area than the equations for the Piedmont Region, and the average difference of -55.8 percent indicates substantial bias in the measurements for this region. As a result, the peak-flow regression equations were not implemented in StreamStats for the Western Coastal Plain Region.