Field data-based implementation of land management and terraces on the catchment scale for an eco-hydrological modelling approach in the Three Gorges Region, China

• In-field acquisition of land management and terrace data on the catchment level.
• Processing of field data towards SWAT input datasets.
• Step-wise testing of the implementation of land management and terraces in SWAT.
• Evaluation of the implementation towards water balance, streamflow and sediment.

In this study, an innovative method to generate spatially-distributed data sets on land management and terraces based on sparse field data for a steep-sloping catchment in the Three Gorges Region in China is introduced and tested using the eco-hydrological model SWAT. The generation of such data sets is necessary for the development and evaluation of Best Management Practices (BMP) towards a reduction of high inputs of sediment and nutrients in water bodies. It is hypothesized that the inclusion of land management as well as terraces in the eco-hydrological modelling approach are individually as well as combined able to increase the model efficiency regarding streamflow and sediment. The results of the study show that the field data sets on land management and terraces can be used to generate useful SWAT input data sets to represent management and conservation practices and the model results are plausible. The effect of land management and terraces on streamflow is identified to be rather small. At the same time a strong effect of the inclusion of the terrace dataset on sediment yields can be observed, which can be seen as an improvement of the process representation within the model. By introducing the new method the study contributes to an improved representation of land management and terraces in data-scarce study regions in eco-hydrological models. At the same time the study confirms the importance of the consideration of BMPs in eco-hydrological modelling, especially towards the representation of the dynamics of sediment and sediment-bound substances.