Ecological Indicators (Aug 2022)
From simple to complex – Comparing four modelling tools for quantifying hydrologic ecosystem services
Abstract
The pursuit of good management of our waters poses permanent challenges to the whole society. Decision-makers often need to define appropriate and sustainable strategies on interdisciplinary topics, like water management issues. The rapidly evolving quantification and mapping of hydrologic ecosystem services (HES) is putting hydrologic and water management issues into an ecosystem services (ES) framework, which can be a step towards reconciling different aspects of land use and water management. Different tools can be used for modelling HES, with a wide range according to their basic properties, e.g., structure, methodology, computational needs, data requirements, reliability, controllability. As a result of that, the numeric values, spatial patterns, and reliability of HES assessments and the uncertainties in their results may differ significantly.In this paper, we covered almost the whole palette of HES mapping tools with regards to modelling approach: we used InVEST, SWAT and two novel rule-based matrix models for the same pilot area, the 1530 km2 hilly catchment of the Zala River (Hungary). We mapped three HES: flood control, erosion control and nutrient (total phosphorus) retention. Our aim was to examine the relevance of the differences between the HES mapping tools through analysing the spatial differences between the results obtained with the applied. We carried out spatial similarity tests and hotspot analysis at the computational unit level for the individual HES and in an aggregated way as well.As a result of the spatial pattern similarity tests, InVEST and the matrix models showed moderate to strong correlation (p < 0.001) for each HES. Due to that, the novel matrix models could be considered as robust HES mapping tools on a larger spatial scale (regional or larger). InVEST appeared to be the most efficient HES mapping tool considering computational demand, result reliability, and data- and expertise requirements. The results of our study draw attention to the importance and actual shortcomings of the land use and land cover (LULC) structure in the riparian zone. We pointed out that the studied HES in agricultural areas close to the watercourse are often disservices (negative HES were provided with the actual LULC scenario compared to a non-vegetated LULC scenario) due to the nutrient loads from fertilization. We found that parts of the best and worst HES provisioning areas (hotspots and coldspots) could be delineated without hydrologic modelling, because their unfavourable combination of environmental factors and LULC conditions themselves determine these areas to be hotspot or coldspot.
