FLUXES AND FATES OF NITROGEN IN SOIL OF AN UNPOLLUTED OLD‐GROWTH TEMPERATE FOREST, SOUTHERN CHILE

Nitrogen retention and recycling are topics of enduring interest in ecosystem ecology, yet we lack a mechanistic field‐tested model of how these processes work in unpolluted, old‐growth temperate forests. Forests of the Cordillera Piuchué Ecosystem Study (CPES) in southern Chile provide an opportunity to examine nitrogen cycling and retention in a forest that is virtually free of human disturbance. We applied 15 N pool dilution and pulse‐chase tracer techniques as complementary approaches within small plots to understand flows of inorganic nitrogen in the surface soil of an evergreen mixed‐angiosperm forest. We also followed separate pulses of 15 NH 4 + and 15 NO 3 − for two years to gain insights into how short‐term mechanisms of inorganic nitrogen cycling translate into long‐term patterns of ecosystem nitrogen retention. Strong consumption appears to limit losses of NH 4 + and NO 3 − from this forest, and predominantly by the same mechanisms for both forms of nitrogen. As a result, the extent of 15 NH 4 + and 15 NO 3 − retention were also similar, yet ∼44‐fold higher rates of gross NH 4 + production leads to the dominance of NH 4 + over NO 3 − in soil and stream waters. Microbial biomass played a key role in the short‐term assimilation of 15 N tracers, but retention was only transient. Turnover of 15 N through microbial biomass was rapid and appeared to be only weakly retained in soil exchangeable pools, fine roots, and soil organic matter, resulting in substantial losses of 15 N from soils within weeks of tracer additions. Assimilation of 15 N into fine roots was a much larger sink (13%) than has been reported for other forested ecosystems (1–3%), and the transport of 15 N from microbial biomass to aboveground sinks in vegetation may explain the observed loss of 15 N from surface soils over time. Losses of 15 N from microbial biomass did not enter the extractable pool of dissolved organic nitrogen (DON), suggesting that DON losses do not originate directly from active microbial turnover, and also that microbial activity may not exert as much control over hydrologic losses of DON as compared to losses of NH 4 + and NO 3 − . Our results also suggest an additional rapid and extremely transient (1 d) mechanism of NO 3 − retention via incorporation into extractable‐DON. The long‐term retention of 15 N at the whole‐plot level did not differ significantly between 15 NH 4 + and 15 NO 3 − treatments, and averaged 65% after two years. The lack of an appreciable change in 15 N recovery for ∼1.5 yr following the initial assimilation, redistribution, and loss of 15 N suggests that the majority of 15 N was not recycled over the long term through inorganic nitrogen pools and microbial biomass via mineralization/immobilization pathways. Instead, long‐term retention of inorganic 15 N appeared to be dominated by rapid and possibly direct assimilation into a slow‐turnover pool of soil organic matter. Elevated 15 N contents in fine‐root and microbial pools for up to two years after 15 N additions, however, also indicated sustained biotic retention of inorganic nitrogen. Our results suggest very similar retention of NH 4 + and NO 3 − that is dominated by rapid assimilation and turnover through microbial biomass in the short term (weeks), and transfer from microbial biomass into nitrogen‐conserving plant (and to a lesser extent soil organic matter) pools in the long term (years). These processes result in efficient long‐term retention of nitrogen in unpolluted old‐growth temperate forests.