Disentangling and ranking the influence of multiple stressorson macroinvertebrate communities in a tropical river syste

Abstract

Spatial contrasts in lotic macroinvertebrate assemblages have been ascribed to physico-chemical, hydrological, and physical habitat conditions. The objective of this study was to disentangle and rank the influences of multiple stressors on benthic macroinvertebrates in the Manyame Catchment, Zimbabwe. Sampling sites were chosen according to their relative location from dams (upstream of dams; 0–10 kmdown stream of dams; 11–20 km downstream of dams; and >20 km downstream ofdams). Benthic macroinvertebrates, physico-chemical, physical habitat structure; and sediment grain size data were collected at12 sites in two seasons and paired with long-term hydrological data. Principal component analysis (PCA) was employed to identify key variables that describe the major sources of variation while minimizing redundancy. Subsequently, generalised linear models (GLMs) were used to investigate the relationship between these key variables and macroinvertebrate community structureand composition. Finally, we used canonical correspondence analysis (CCA) and partialCCA to illustrate the significant relationships from the GLMs across site categories.Macroinvertebrate community structure and composition was significantly driven byfour physico-chemical variables (dissolved oxygen, pH, total nitrogen, total dissolvedsolids), five hydrological variables (base flow index, low pulse duration, December low flow, small flood timing, extreme low flow duration) and interactions among these variables. Interactions of physico-chemical conditions and low flows demonstrated that pollution will aggravate impacts of hydrological alteration on biota. Partial CCAshowed that physico-chemical variables had the greatest influence on macro-invertebrate communities. Hence, physico-chemical conditions remain the most important in structuring macroinvertebrates in the Manyame Catchment. The study highlights the importance of accounting for multiple stress factors in in-stream ecology.