Almost all natural ecosystems on Earth have been disturbed by human development. Billions of dollars are invested annually to restore degraded ecosystems, yet many countries continue to face a dilemma between the needs of economic development and ecosystem restoration. Therefore, adequate assessment of the efficiency of restoration projects in maintaining and restoring natural ecosystem services in line with the rapid urbanization is needed.

Recently, a research team led by Prof. XU Jun from the Institute of Hydrobiology (IHB) of the Chinese Academy of Sciences demonstrated that, despite the unstopped expansion of urbanization, the ecological damage caused by recent rapid economic development could potentially be mitigated by the combined effect of massive restoration investments synchronized across whole catchments, although these effects can be expected to be muted if urbanization continues apace at the same time. The study was published in Water Research.

Prof. XU’s team used a moving window approach that could compare the local spatial trends of each restoration project index with the local spatial trends of aquatic ecosystem indices and urbanization intensity. This was the first time the moving window approach was applied to assess the effectiveness of aquatic ecosystem restoration.

The researchers assessed the relationships between restoration projects and ecosystem recovery by using the generalized linear mixed models (GLMM). They demonstrated that investments in environmental restoration projects improved water quality and biodiversity despite urban growth, investments in source control had a stronger impact on water quality than investments in restoring sinks, and investments in sink water quality control improved nutrient levels, albeit not as strong as investments in source controls.

Besides, the researchers analyzed the changes in species composition between restored (2017) and degraded (2007) sites using the command beta.temp in the R package betapart. The partitioning of the Sørensen dissimilarity index was dominated by species turnover (β_SIM) (mean = 0.44, standard deviation = 0.36), implying that in any given site an average of 44% of the species were unique to the time, either 2007 or 2017 site assemblage.

Since permutational multivariate analysis of variance (PERMANOVA) has showed the composition of benthic macroinvertebrate communities differed significantly between degraded (2007) and restored (2017) time periods, they used similarity percentage (SIMPER) analysis to identify which taxa contributed the most to the average Bray-Curtis dissimilarity between the two-restoration phase. SIMPER analysis identified eight species cumulatively contributing > 70% to the dissimilarity between restored (2017) and degraded (2007) invertebrate communities. Some sensitive species to anthropogenic pressures with low tolerance values recolonized after restoration (2017).

This study provided new insight into the different response between abiotic and biotic indices after the restoration, and served as a reference to plan a better allocation of resources and money when restoring aquatic ecosystems.