Litter, warming and plants affect respiration and allocation of soil microbial and plant C, N and P in arctic mesocosms

In a mesocosm experiment, we studied decomposition rates as CO2 efflux and changes in plant mass, nutrient accumulation and soil pools of nitrogen (N) and phosphorus (P), in soils from a sub-arctic heath. The soil was incubated at 10 degreesC and 12 degreesC, with or without leaf litter and with or without plants present. The purpose of the experiment was to analyse decomposition and nutrient transformations under simulated, realistic conditions in a future warmer Arctic. Both temperature enhancement and litter addition increased respiration rates. Temperature enhancement and surprisingly also litter addition decreased microbial biomass carbon (C) content, resulting in a pronounced increase of specific respiration. Microbial P content increased progressively with temperature enhancement and litter addition, concomitant with increasing P mineralisation, whereas microbial N increased only in the litter treatment, at the same time as net N mineralisation decreased. In contrast, microbial biomass N decreased as temperature increased, resulting in a high mobilisation of inorganic N. Plant responses were closely coupled to the balance of microbial mineralisation and immobilisation. Plant growth and N accumulation was low after litter addition because of high N immobilisation in microbes and low net mineralisation, resulting in plant N limitation. Growth increased in the temperature-enhanced treatments, but was eventually limited by low supply of P, reflected in a low plant P concentration and high N-to-P ratio. Hence, the different microbial responses caused plant N limitation after litter addition and P limitation after temperature enhancement. Although microbial processes determined the main responses in plants, the plants themselves influenced nutrient turnover. With plants present, P mobilisation to the plant plus soil inorganic pools increased significantly, and N mobilisation non- significantly, when litter was added. This was presumably due to increased mineralisation in the rhizosphere, or because the nutrients in addition to being immobilised by microbes also could be absorbed by plants. This suggests that the common method of measuring nutrient mineralisation in soils incubated without plants may underestimate the rates of nutrient mobilisation, which probably contributes to a commonly observed discrepancy of measured lower rates of net nutrient mineralisation than uptake rates in arctic soils. (C) 2004 Elsevier Ltd. All rights reserved.