Flood response for the watersheds of the Fernow Experimental Forest in the central Appalachians

Abstract

AbstractWe examine flood response of high‐gradient, forested central Appalachian watersheds through analyses of rainfall, streamflow, and piezometer observations from the Fernow Experimental Forest near Parsons, West Virginia. Analyses focus on hydrologic processes that control the “upper tail” of flood distributions. The largest flood peaks in the Fernow are an order of magnitude smaller than record floods in the central Appalachian region (for basins of comparable drainage area). We examine flood distributions in the Fernow using extreme value distributions (Generalized Extreme Value and Generalized Pareto distributions) and compare them to other watersheds in the central Appalachians. To examine the role of antecedent soil moisture on flood response, we installed a network of 415 crest‐stage piezometers on two headwater watersheds (0.30 and 0.14 km2) of the Fernow. Observations show pronounced heterogeneity of subsurface saturation even within the unchannelized swales of headwater watersheds. Shallow perched water tables over large portions of a watershed occur infrequently in forested central Appalachian basins, but may play an important role in extreme flood response. Fernow watersheds include “treated” and control watersheds with stream gaging records extending back to 1951. We examine nonstationarites in flood frequency in the Fernow and show that forest management practices have had relatively minor impacts on flood frequency.