|John D. Aber
Complex Systems Research Center
University of New Hampshire
Durham, NH 03824
Department of Civil and Environmental Engineering
Syracuse, NY 13244
NITRATE CONCENTRATIONS in streams at Hubbard Brook show a complex pattern of change at seasonal, annual and decadal scales. In an earlier paper (Aber and Driscoll 1997) a simple model of water, carbon and nitrogen dynamics was shown to capture many of the major features of the nitrate record on watershed 6 when driven by accurate monthly climate data and an accurate description of previous disturbance history. Post-hoc validations of models are often less convincing than predictions of future events. The ice storm disturbance offers the opportunity to test model predictions made before the validation data are available.
In this study we used the PnET-CN model (Aber et al. 1997, Aber and Driscoll 1997) parameterized for vegetation, soil and disturbance history on watershed 6. Ice storm damage was simulated as a fractional removal of aboveground tree biomass. Because final and accurate data on the extent of damage were not available by the time these simulations were run, we tested two different levels of removal, 20% and 30%. Results were initially presented at the Hubbard Brook science meeting in the summer of 1998.
Figure 1 summarizes the predicted changes in stream water nitrate concentration over time for the two levels of damage. These values are net of other changes in nitrate values due to long-term recovery from previous disturbances or short-term responses to interannual climate variability.
Two stages of response can be seen. The first is an immediate increase in nitrate loss due to reduced photosynthesis and NPP as a result of loss of canopy mass. This response is predicted to peak in 2000 at 0.5 and 0.9 mg/L NO3-N for a 20% and 30% canopy loss, respectively, and then decline. A longer-term increase of 0.07 to 0.1 mg/L NO3-N is predicted to occur between 2005 and 2050 due to long-term changes in C:N ratios, foliar N concentration and altered NPP. While the initial response should be clearly visible in the measured nitrate record, the longer-term response, because it is smaller and continuous over several decades, would be very difficult to separate from other sources of variation in the nitrate record.
|Figure 1. Net change in nitrate-N loss (mg/L) predicted by the PnET-CN model due to ice storm damage removing 20% and 30% of canopy mass for all of watershed 6.|