Abstract:Ecological restoration of freshwater ecosystems is now being implemented to mitigate anthropogenic disruption. Most emphasis is placed on assessing physico-chemical and hydromorphological properties to monitor restoration progress. However, less is known about the structural integrity and ecosystem health of aquatic ecosystems. In particular, little is known about how ecosystem function changes following river habitat restoration, especially in China. Leaf litter decomposition can be used as an indicator of stream ecosystem integrity. Therefore, the leaf breakdown rate was measured to assess the ecosystem function of restored rivers. By comparing leaf breakdown rates in urban rivers undergoing habitat restoration with that in degraded urban rivers and rivers in forested areas (i.e., reference conditions), we aimed to determine: (i) how habitat restoration affected leaf litter decomposition? (ii) the relationship between leaf litter decomposition to both environmental (habitat and physicochemical variables) and biological factors (benthic communities), and (iii) identify the factors that contribute most to the variance in leaf litter breakdown rates. The results demonstrated a significant increase in leaf breakdown rate (120%% in summer and 28%% in winter) in the restored rivers compared to the degraded rivers. All environmental and biotic factors evaluated contributed synergistically to the differences in leaf litter decomposition among the three river types. The role of macroinvertebrates, mainly shredders, appeared to be particularly important, contributing 52%% (summer) and 33%% (winter) to the variance in decomposition, followed by habitat characteristics (e.g. substrate diversity, water velocity; 17%% in summer, 29%% in winter), physico-chemical variables (e.g. nutrient and organic pollutants; 11%% in summer, 1%% in winter) and biofilm bacteria (0%% in summer, 15%% in winter). Habitat restoration positively affected the structure and function of the previously degraded streams. Knowledge on controlling variables and their attribution to changes of ecosystem functioning provides guidance to assist the future planning of ecological restoration strategies. (C) 2020 Elsevier B.V. All rights reserved.