Long-term Decreases in Stream Nitrate: Successional Causes Unlikely; Possible Links to DOC?

Abstract

Huntington (2005) poses the hypothesis that ‘‘immobilization of inorganic N, in combination with region-wide recovery from past disturbances may explain the decrease in stream water nitrate in recent decades’’ that we observed in 28 White Mountain, New Hampshire streams (Goodale and others 2003). We focus our response on Huntington s suggestions, and note that a range of additional factors not discussed here can have marked impacts on long-term patterns of NO3 ) loss (for example, N deposition, insect outbreaks, climate variation and extreme events such as soil frost). Below, we address Huntington s suggestion in two parts, first considering the role of region-wide recovery from historical disturbances and then considering recent changes in environmental factors that might have increased soil N immobilization through increased belowground productivity. We then introduce a hypothesis that suggests that changes in stream NO3 ) might be partly linked to observed changes in dissolved organic carbon (DOC). We agree with Huntington that historical disturbances have had substantial impacts on forest carbon and nitrogen dynamics in the White Mountains of New Hampshire (see for example, Bormann and Likens 1979; Covington 1981; Thorne and Hamburg 1985). We have previously considered in some depth the role of the region s disturbance history as a major driver of patterns of organic matter accumulation and watershed N retention (see for example, Vitousek and Reiners 1975; Aber and Driscoll 1997; Goodale and others 2000; Goodale and Aber 2001). We agree that re-accumulation of soil organic matter in re-growing forests provides important sinks for atmospheric CO2 and N deposition across large parts of the eastern U.S. (see for example, Huntington 1995; Hooker and Compton 2003). However, we do not believe that forest re-growth and soil re-accumulation explain the approximately 70% decrease in stream NO3 ) concentration (mean decrease of 25 lmol/L) that we observed between 1973)4 and 1996–97 (Goodale and others 2003). First, we emphasize that we observed large decreases in stream NO3 ) concentration in both aggrading successional systems and in several oldgrowth watersheds that had no evidence of standlevel disturbance by historical logging or fire (see Goodale and others 2003 Figure 6; see also Martin and others 2000). Second, the observed decreases in stream NO3 ) between the 1970s and the 1990s would require a synchronous increase in the rate of soil re-accumulation during this time period across all watersheds. Few successional factors could cause an increase in soil re-accumulation rate, and such factors are unlikely to occur simultaneously in both old-growth and successional watersheds.