Please login first
A comparative analysis of SMAP derived soil moisture modeling by optimized machine learning methods; a case study of Quebec province
1 , * 2
1  Universit√© Laval
2  Ottawa University


Many hydrological responses rely on the water content of the soil (WCS). As soil moisture rises, more runoff is created, resulting in increased sediment movement. This environmental element affects the soil's erosion resistance. Runoff, sediment, and erosion are crucial in hydraulic structure design and watershed studies. The variations in the WSC affect the agriculture section, and the sustainable management of agricultural water and land resources will depend on this factor. Many environmental parameters like soil and surface temperature, the amount of precipitation, and groundwater level influence this parameter. Hydrological extremes and climate variations intensely impact these parameters, which increases the importance of studying WCS under changing climate conditions. The constraints of measuring and expenditure limitation cause this parameter not to be accessible in high spatio-temporal resolutions everywhere, particularly in vast areas like Quebec, Canada. Therefore, a strategy should be researched to collect and model this useful parameter in data-scarce locations. This research will use SMAP products to model and forecast the SWC. Accordingly, Google Earth Engine cloud datasets will be used. Using this platform provides the possibility of obtaining curated datasets worldwide. The regression support vector machine (SVM) and extreme learning machine (ELM) models optimized by the teacher-learning algorithm will be used to model and forecast the datasets. These two methods are the most conventional artificial intelligence methods in non-linear phenomena modeling. The inherent intense seasonality and stochastic patterns in the WCS make these modeling techniques suitable for forecasting and extracting patterns in the datasets. The former model can generate explicit equations that can be handy for applying to other datasets and conditions, while the latter is considerably fast among other AI methods and can generate real-time results. In the end, the accuracy of outcomes and model efficiency will be evaluated concurrently.

Keywords: Water resources, Watershed, Soil, Moisture, Machine learning, AI