“…Despite its importance, understanding the spatiotemporal dynamics of soil moisture is an arduous task, mainly due to its high nonlinearity (Xiadong et al, 2016). To overcome this difficulty, various physically based models have been used to model soil moisture, such as the Sacramento Soil Moisture Accounting (SAC‐SMA) Model (Peck, 1976), Soil and Water Assessment Tool (SWAT) (Amorin, Viola, Junqueira, Oliveira, & Mello, 2020; Arnold, Srinivasan, Muttiah, & Williams, 1998; Srinivasan, Ranabarayanan, Arnold, & Bednarz, 1998), Variable Infiltration Capacity Model (VIC) (Liang, Lettenmaier, Wood, & Burges, 1994), Lavras Simulation of Hydrology (LASH) (Andrade et al, 2020), and Decision Support System for Agrotechnology Transfer (DSSAT) (Ritchie & Otter, 1985). A common problem associated with the application of these models is that they usually require various physical parameters (e.g., hydraulic conductivity, preferential pathways, porosity and air entry pressure) that can be difficult to obtain due to the complexity of the soil–vegetation–atmosphere continuum (Gill, Asefa, Kemblowski, & McKee, 2006; Meng & Quiring, 2008) and the deficiency of monitoring systems in developing countries such as Brazil.…”