Inputs, Outputs, and Accumulation of Nitrogen in an Early Successional Moss (Polytrichum) Ecosystem

Measurement of total ecosystem nitrogen accumulation in a moss (Polytrichum) ecosystem after 13 yr of primary succession upon exposed sands of glacial origin in New Hampshire, USA, revealed an N accumulation rate of 10.1 kg · ha— ¹ · yr— ¹. Comparison of the measured accumulation with a mass balance analysis of measured inputs (bulk precipitation inorganic N plus biological N₂ fixation plus windblown coarse particulate matter) minus outputs (hydrologic losses plus N₂O and N₂ emissions) revealed that 35% of the inputs required to satisfy the observed rate of N accumulation had not been measured. These unmeasured sources include bulk precipitation organic N, dry deposition, and dew inputs. Total N inputs, including the unmeasured sources, were 10.5 kg · ha— ¹ · yr— ¹. Bulk precipitation was the dominant N input, comprising 58% of the annual input. Dinitrogen fixation and coarse particulate organic matter together represented only 7%; unmeasured sources accounted for the remainder (35%). The predominance of bulk precipitation as an input source to the moss ecosystem is in contrast to many other primary successional systems where N₂ fixation is the dominant source of N. Nitrogen outputs from the ecosystem were small (0.40 kg · ha— ¹ · yr— ¹), with most coming from hydrologic export (0.29 kg · ha— ¹ · yr— ¹) and a smaller fraction (0.10 kg · ha— ¹ · yr— ¹) from N₂O effluxes. No N₂ losses were detected. The moss ecosystem was extremely efficient at retaining bulk precipitation N inputs. Additions of ¹ ⁵N—labelled simulated rainfall showed that soil and belowground live moss biomass retained most of the input. The amount of N retained by each component of the ecosystem was related directly to the amount of biomass (living or dead) in the component. Soil organic matter was also important in retaining nitrogen inputs. In a short—term microcosm experiment, soil with a high organic matter content (2.71%) retained 47% of inorganic N in simulated rainfall, compared to only 27% retained in soil with a low organic matter content (0.51%). The moss vegetation exerted control over the N retention efficiency of the ecosystem. Removal of mosses resulted in a short—term retention of N inputs, soon followed by losses that exceeded inputs. Nitrogen retention again resumed following recolonization of the soil by mosses. Development of the moss ecosystem appears to have been enhanced by two positive feedback mechanisms. First, establishment of the mosses led to biomass and soil organic matter accumulation, enhancing N capture and retention, and thus increasing biomass accumulation. Second, development of aboveground biomass further enhanced collection of N in dry deposition and dew, increasing total inputs.