Regionalization of disturbance‐induced nitrogen leakage from mid‐Appalachian forests using a linear systems model

Abstract

AbstractThe ‘leakage’ of nitrate‐nitrogen (nitrate‐N) to surface waters is a common (but not universal) response of forest ecosystems to both human‐induced and natural disturbances. There are several reported examples of the transient leakage of nitrate‐N to surface waters from eastern US forests that have sustained outbreaks of defoliating insects, such as the introduced gypsy moth (Lymantria dispar) larva. Previous research has suggested that annual nitrate‐N leakage from disturbed forests can be modelled using an empirically derived unit nitrogen export response function (UNERF) model. The model represents annual nitrate‐N export as a linear deterministic process in both space and time and is analogous to a unit hydrograph. The goal of the present study was to verify and apply a regionalized, lithology‐based UNERF model that references the geographic distribution of bedrock class and the timing and extent of gypsy moth defoliation of forests in the non‐glaciated highlands of the Chesapeake Bay watershed. Despite an inability to verify the model for most individual watersheds within the study area, the model was able to reproduce the statistical distribution of annual nitrate‐N export to streams that comprised our target population. During water year 1991 (the year following peak defoliation) the model results indicated that regional annual nitrate‐N export had transiently increased by nearly 1500% from a baseline rate of about 0·1 kg ha−1 to a peak value approaching 1·5 kg ha−1. We thus conclude that natural vegetation disturbance is an important mechanism by which dissolved nitrogen is leaked from forested lands to small streams, rivers, and Chesapeake Bay. The present study also illustrates how simple, empirically derived linear systems approaches like the UNERF model can be successfully applied to problems where regionalization is a primary goal. Copyright © 2004 John Wiley & Sons, Ltd.