Spatio-temporal variation of fish taxonomic composition in a South-East Asian flood-pulse system

The Tonle Sap Lake (TSL) is a flood-pulse system. It is the largest natural lake in South-East Asia and constitutes one of the largest fisheries over the world, supporting the livelihood of million peoples. Nonetheless, the Mekong River Basin is changing rapidly due to accelerating water infrastructure development (hydropower, irrigation, flood control, and water supply) and climate change, bringing considerable modifications to the annual flood-pulse of the TSL. Such modifications are expected to have strong impacts on fish biodiversity and abundance. This paper aims to characterize the spatio-temporal variations of fish taxonomic composition and to highlights the underlying determinants of these variations. For this purpose, we used data collected from a community catch monitoring program conducted at six sites during 141 weeks, covering two full hydrological cycles. For each week, we estimated beta diversity as the total variance of the site-by-species community matrix and partitioned it into Local Contribution to Beta Diversity (LCBD) and Species Contribution to Beta Diversity (SCBD). We then performed multiple linear regressions to determine whether species richness, species abundances and water level explained the temporal variation in the contribution of site and species to beta diversity. Our results indicate strong temporal variation of beta diversity due to differential contributions of sites and species to the spatial variation of fish taxonomic composition. We further found that the direction, the shape and the relative effect of species richness, abundances and water level on temporal variation in LCBD and SCBD values greatly varied among sites, thus suggesting spatial variation in the processes leading to temporal variation in community composition. Overall, our results suggest that fish taxonomic composition is not homogeneously distributed over space and time and is likely to be impacted in the future if the flood-pulse dynamic of the system is altered by human activities.洞里萨湖（Tonle Sap Lake, TSL）是典型的洪泛脉冲系统，作为东南亚规模最大的天然湖泊，它同时也是全球最具规模的淡水渔业产区之一，维系着数百万民众的生计。然而，受水利基础设施（水电、灌溉、防洪与供水）建设加速以及气候变化的双重影响，湄公河流域正经历快速变迁，这对洞里萨湖的年度洪泛脉冲动态造成了显著改变。此类改变预计将对鱼类生物多样性与种群丰度产生显著影响。本研究旨在刻画鱼类分类组成的时空变异特征，并揭示这类变异背后的核心驱动因素。为此，我们采用了一项覆盖6个监测点位、为期141周的社区渔获监测项目所收集的数据，该周期完整涵盖了两个水文循环。针对每一周，我们以点位-物种种群群落矩阵的总方差来估算β多样性（beta diversity），并将其拆解为点位对β多样性的局部贡献（Local Contribution to Beta Diversity, LCBD）与物种对β多样性的贡献（Species Contribution to Beta Diversity, SCBD）。随后我们开展多元线性回归分析，以验证物种丰富度、种群丰度与水位是否能够解释点位与物种对β多样性贡献的时间变异特征。研究结果表明，由于点位与物种对鱼类分类组成空间变异的贡献存在差异，β多样性呈现出显著的时间变异。我们进一步发现，物种丰富度、种群丰度与水位对LCBD与SCBD值时间变异的作用方向、函数形态与相对效应在不同监测点位间存在显著差异，这表明驱动群落组成时间变异的过程存在空间异质性。总体而言，本研究结果表明，鱼类分类组成在时空分布上并非均质，若该系统的洪泛脉冲动态因人类活动发生改变，其未来或将受到显著影响。