Carbon pulses but not phosphorus pulses are related to decreases in microbial biomass during repeated drying and rewetting of soils

Drying and rewetting cycles are known to be important for the turnover of carbon (C) in soil, but less is known about the turnover of phosphorus (P) and its relation to C cycling. In this study the effects of repeated drying-rewetting (DRW) cycles on phosphorus (P) and carbon (C) pulses and microbial biomass were investigated. Soil (Chromic Luvisol) was amended with different C substrates (glucose. cellulose, starch: 2.5 g C kg(-1)) to manipulate the size and community composition of the microbial biomass, thereby altering P mineralisation and immobilisation and the forms and availability of P. Subsequently, soils were either subjected to three DRW cycles (1 week dry/1 week moist) or incubated at constant water content (70% water filled pore space). Rewetting dry soil always produced an immediate pulse in respiration, between 2 and 10 times the basal rates of the moist incubated controls, but respiration pulses decreased with consecutive DRW cycles. DRW increased total CO(2) production in glucose and starch amended and non-amended soils, but decreased it in cellulose amended soil. Large differences between the soils persisted when respiration was expressed per unit of microbial biomass. In all soils, a large reduction in microbial biomass (C and P) occurred after the first DRW event, and microbial C and P remained lower than in the moist control. Pulses in extractable organic C (EOC) after rewetting were related to changes in microbial C only during the first DRW cycle; EOC concentrations were similar in all soils despite large differences in microbial C and respiration rates. Up to 7 mg kg(-1) of resin extractable P (P(resin)) was released after rewetting, representing a 35-40% increase in P availability. However, the pulse in P(resin) had disappeared after 7 d of moist incubation. Unlike respiration and reductions in microbial P due to DRW, pulses in P(resin) increased during subsequent DRW cycles, indicating that the source of the P pulse was probably not the microbial biomass. Microbial community composition as indicated by fatty acid methyl ester (FAME) analysis showed that in amended soils, DRW resulted in a reduction in fungi and an increase in Gram-positive bacteria. In contrast, the microbial community in the non-amended soil was not altered by DRW. The non-selective reduction in the microbial community in the non-amended soil suggests that indigenous microbial communities may be more resilient to DRW. In conclusion, DRW cycles result in C and P pulses and alter the microbial community composition. Carbon pulses but not phosphorus pulses are related to changes in microbial biomass. The transient pulses in available P could be important for P availability in soils under Mediterranean climates. (C) 2009 Elsevier Ltd. All rights reserved.