A linear model of the effects of disturbance on dissolved nitrogen leakage from forested watersheds

Abstract

The leakage of dissolved nitrogen (N), primarily in the form of nitrate, from forested watersheds in the mid‐Appalachian region has important water quality ramifications for small acid‐sensitive streams and for downstream receiving waters such as Chesapeake Bay. Dissolved N leakage is a common, well‐documented response of forested watersheds to forest management practices such as clear‐cutting. Patterns of N leakage from mid‐Appalachian watersheds during the late 1980s and early 1990s have also been shown to display considerable temporal and spatial synchrony with outbreaks of defoliation by the gypsy moth (Lymantria dispar) larva (a nonnative forest insect pest). This evidence suggests that forest disturbance may be an important contributor to N leakage in the mid‐Appalachian region. A logical first step in testing this hypothesis is evaluating the ability of a simple, unit nitrogen export response function (UNERF) model to explain temporal changes in annual N export from gaged forested watersheds in the years following disturbance. Annual N export data from seven such watersheds were analyzed as part of the study: two small (<0.5 km2) watersheds subjected to deforestation and five larger (1.6–12.6 km2) watersheds subjected to repetitive defoliation by the gypsy moth larva. Several forms of linear UNERF models, parameterized by deconvolution of annual time series of N export using linear programming or by a least squares method, were generally found to be minimally biased and to explain high percentages (38–98%) of the total variation in annual N export. Despite their neglect of spatial and temporal ecosystem nonlinearities these linear models appear reasonably robust, making them at least as useful as their more complex nonlinear brethren for purposes of regionalization.