Flow Estimation

Flow estimates are based on the 1:24,000 scale NHD stream data. In several cases, the original data was modified to add connectivity to perform network traversal. In other cases, disconnected headwater stream segments with no obvious connection to the rest of the network were deleted.

In addition, underground passages and ephemeral stream segments were added to connect sinking streams and large depressions found in the karst region running though Greenbrier County. These additions were guided by dye trace results reported by Dasher and Boyer (2000) and in Dasher (2012).

All flow values were calculated using regional equations from Wiley (2008) and Wiley and Adkins (2010). This involved calculating the drainage area, longitude centroid of the drainage area, and mean precipitation within the drainage area for each stream segment in the state.

Calculating a drainage area for every stream segment using typical GIS tools was not computationally practical on a state-wide scale. Therefore, an alternative routine was developed that incrementally built up drainage areas from the direct drainages to each individual stream segment that occurred upstream of a particular stream segment. These segment-level drainage areas were easily calculated using Esri's watershed command.

The input elevation model was a hydrologically-conditioned 9-meter elevation grid resampled from 3-meter source data. The source data was a seamless mosaic of the 1/9 arc second National Elevation Dataset, supplanted by lidar data where available. Hydrological conditioning enforced drainage to the modified NHD stream network. The conditioning method was based on the technique presented by Nagel, et.al. (2011) which essentially lowers the elevation values near streams by a value that is inversely proportional to the distance from a stream segment. The method described by Nagel was modified to use an exponential term, so that the the conditioning became significantly more pronounced in close proximity to the stream line. For drainage areas that fell outside the elevation model--primarily streams that flow into the state--existing watershed boundaries developed by the NRCS and USGS were used.

Finally, precipitation means were calculated using the 1971-2000 PRISM 30-year normal dataset, which was used to develop the regional equations.

References

Dasher and Boyer, 2000. Recent Spring Creek Area Dye Tracings Greenbrier County, West Virginia. West Virginia Speleological Survey Monograph #2.
Dasher, 2012. The Caves and Karst of West Virginia. The West Virginia Speleological Survey Bulletin #19.
Nagel, D; John M. Buffington, and Charles Luce 2011. A Simple Drainage Enforcement Procedure for Estimating Catchment Area Using DEM Data. Presented at the NW GIS User Conference, Oct. 17-21, Boise, ID.
Wiley and Atkins, 2010. Estimation of Selected Seasonal Streamflow Statistics Representative of 1930–2002 in West Virginia. USGS Scientific Investigations Report 2010–5185.
Wiley, 2008. Estimating Selected Streamflow Statistics Representative of 1930–2002 in West Virginia. USGS Scientific Investigations Report 2008–5105.