Decoding lake water dynamics to optimize watershed agriculture through isotopic analyses of memory effects and hydrological connectivity

Abstract

Understanding the complex spatiotemporal dynamics of lake water systems is critical in the context of intensifying global environmental changes. In this study, a novel stable isotope analysis method, combined with Bayesian mixing models, is applied to investigate hydrological connections and water source contributions in a specific area on the southwest shore of Dianchi Lake, China, at a monthly scale. The study reveals a significant “memory effect,” where 56.76% of the lake water volume in the current month is associated with the lake water volume in the prior month, with notable seasonal variations. The relative contributions of precipitation, surface water, soil water for different agricultural land uses, and groundwater to the lake water balance are assessed. The hydrological processes in Dianchi Lake are significantly influenced by agricultural land use, with greenhouse soils contributing less water than open field soils. Water outflow, rather than evaporation, is the primary factor reducing the nearshore lake volume, highlighting the impact of human activities. The dependency of water source contributions on meteorological factors is also examined, with seasonal and weather effects on lake water dynamics and agricultural water availability observed. By integrating isotope data with meteorological records and advanced modeling techniques, a quantitative framework for evaluating hydrological changes in lake catchments is developed. The developed approach enhances our understanding of lake water system dynamics and can enhance agricultural water management strategies, water resource allocation, irrigation planning, and climate change adaptation in agricultural watersheds.