The fate of phosphorus in a seasonal, lowland, Amazonian ecosystem: Biological cycling versus geochemical sequestration.

摘要

The aim of this research was to determine the relative importance of biological and geochemical cycling of phosphorus (P) in highly weathered tropical soils and to determine the roles of soil texture, water availability and soil P availability as controlling factors of both cycles. Over 80% of the soils in the Neo-Tropics are classified as Oxisols and Ultisols, the most weathered soil orders. These soils commonly have low nutrient concentrations, with P considered the primary limiting nutrient to net primary productivity (NPP). However, these soils support highly productive tropical lowland rainforest ecosystems, suggesting that the biota successfully compete for limited P resources and maximize the efficiency with which they use the P they do acquire: Phosphorus availability in highly weathered soils is constrained by both the relatively small size of the total soil P pool and the capacity of the soils to occlude P into geochemically and physically protected forms (reviewed in Chapter 1). I hypothesized that distribution of P in soil, microbial and plant pools and patterns of root and microbial P cycling would differ in soils with different geochemical P sorption capacities (i.e. sands versus clays). Results from a fertilization study (Chapter 2) showed that both the total amount of P retained and the distribution of the P in the abiotic and biotic pools varied significantly with soil texture and with season. This suggests that, on short time scales (1 year) ecosystem P cycling is dominated by biological processes and is sensitive to climatic and environmental variations. I further hypothesized that decomposition, a major P flux in low P systems and a key process in the biological P cycle, would be sensitive to soil P availability. A common substrate decomposition experiment (Chapter 3) showed that rates of leaf tissue mass decay did not vary with soil texture or P fertilization, but that rates of P immobilization during the decay process were strongly correlated to fertilization levels. While it appears that decomposition is not P limited in this system, the microbial biomass does clearly respond to increases in P availability. With the strong patterns in P pools with season seen in the two previous experiments, I hypothesized that seasonal changes in soil moisture, particularly the rapid increase in soil moisture at the onset of the wet season, would have large effects on the labile and organic soil P pools. Results from set of soil wetting experiments (Chapter 4), demonstrated that water addition effects on soil and microbial P pools were dependent on both the soil water content and the fertilization level but not on soil texture. Supporting my previous conclusions, these results suggest that the fate of labile P in highly weathered soils in the seasonal tropics is largely determined by biological mechanisms, which in turn, are under the control of climatic and edaphic factors. While the large proportion of soil P held occluded forms indicates that geochemical sorption is an important, and possibly dominant process in P cycles in highly weathered soils over the long term (centuries), biological P cycling appears to be more effective at retaining labile P on short time scales (1 year).