Harnessing Soil Moisture Data for Enhanced Eco-Hydrological Modeling Precision in Snow-Dominated Catchment in Finland

Eco-hydrological modeling in water resources management has a pivotal role in the assessment of physical processes at various spatial-temporal scales. However, modeling the hydrological processes intrinsically contains uncertainties. Such uncertainties need to be addressed to develop a reliable hydrological model. In this study, in-situ and remotely sensed soil moisture data are used to enhance the precision of hydrological modeling using the Soil and Water Assessment Tool (SWAT). The objectives of this study are, (i) to assess the uncertainty and their propagation in hydrological modeling using the conventional and multi-source data set, and (ii) to simulate the hydrologic parameters using soil moisture as an indicator to evaluate uncertainties in hydrological forecasting. This study is carried out in the Temmesjoki basin of northern Finland with a basin area of 1190 km². This region’s land cover is dominated by forest (61%), agricultural lands (18%), and shrubs (13%). The average annual rainfall and annual average temperature in this region are 406.21 mm, and 2.60°C respectively. The mean daily discharge ranges from 0.17 to 34.15 m³/s. The in-situ soil moisture data and Soil Water Index from the Copernicus Global Land Service are used to test the hypotheses. The Sequential Fitting Algorithm (SUFI-2) in R-SWAT was used for sensitivity and uncertainty analysis and calibration of the streamflow and ET. Two conceptual models are built to compare conventional data sources and multi-source data sets for the assessment of uncertainties in the simulation of the hydrological process. Preliminary analysis of hydrologic parameters of the basin reveals higher and non-uniform distribution of rainfall, ET, and discharge during summer months. Furthermore, the application of soil moisture data for the calibration of the SWAT model reveals higher fitness score, and, at the same time, the in-situ soil moisture data are found to reflect more accurately the soil moisture conditions in SWAT model, which results in the reduction of uncertainties. Consequently, the model conceptualized with the multi-source data sets provides a better water budget for the catchment.