Phosphorus Cycling in Sediments of Deep and Large Reservoirs: Environmental Effects and Interface Processes
Jue Wang, Jijun Gao, Qiwen Wang, Laisheng Liu, Huaidong Zhou, Shengjie Li, Hongcheng Shi, Siwei Wang
Abstract
Although the sediment–water interface of deep and large reservoirs is recognized as a dominant source of internal phosphorus (P) loading, the quantitative hierarchy of environmental drivers and their interaction thresholds remains poorly resolved. Here, we integrate 512 studies to provide the first process-based synthesis that partitions P release fluxes among temperature, pH, dissolved oxygen, salinity, sediment properties, and microbial activity across canyon, valley, and plain-type reservoirs. By deriving standardized effect sizes from 61 data-rich papers, we show that (i) a 1 °C rise in bottom-water temperature increases soluble reactive P (SRP) flux by 12.4% (95% CI: 10.8–14.0%), with sensitivity 28% lower in Alpine oligotrophic systems and 20% higher in warm monomictic basins; (ii) a single-unit pH shift—whether acid or alkaline—stimulates P release through distinct desorption pathways,; and (iii) each 1 mg L−1 drop in dissolved oxygen amplifies release by 31% (25–37%). Critically, we demonstrate that these drivers rarely act independently: multi-factor laboratory and in situ analyses reveal that simultaneous hypoxia and warming can triple the release rate predicted from single-factor models. We further identify that >75% of measurements originate from dam-proximal zones, creating spatial blind spots that currently limit global P-load forecasts to ±50% uncertainty. To close this gap, we advocate coupled metagenomic–geochemical observatories that link gene expression (phoD, ppk, pqqC) to real-time SRP fluxes. The review advances beyond the existing literature by (1) establishing the first quantitative, globally transferable framework for temperature-, DO-, and pH-based management levers; (2) exposing the overlooked role of regional climate in modulating temperature sensitivity; and (3) providing a research agenda that reduces forecasting uncertainty to <20% within two years.