The application of spatially distributed integrated hydrologic models (IHMs) to better manage water resources depends on reasonable parameter estimates. In general, it is difficult to estimate the parameters of IHMs, especially in areas with sparse monitoring networks; however, recent advances in remote sensing (RS) data acquisition herald a new era in IHMs calibration. The present study aims to estimate the values of different observation data sets for calibrating an IHM from the surface to the saturated zone. Also, these data sets will be evaluated for their effectiveness in reducing uncertainty related to changes in saturated zone storage. Data worth assessment was performed using linear analysis through singular value decomposition and the Bayesian method. The results show that remote sensing-based evapotranspiration (ET) and streamflow (R) data sets help infer parameters related to surface runoff, evapotranspiration, and capillary water in the surface and soil zones. Additionally, R observations provide information for estimating only a few parameters in the vadose zone, particularly those related to vertical hydraulic conductivity and soil pore size. The current resolution of ET and R data sets does not affect parameter estimation for stream and groundwater flow. The hydraulic head (H) data set is crucial for estimating parameters in the saturated zone. It also improves parameter identifiability in the surface, soil, and vadose zones. Further, uncertainty analysis indicates that the H data set is the most effective for reducing uncertainties in predicted changes to saturated zone storage. Nevertheless, ET and R observations can also help reduce uncertainty in the absence of the H observation data set. Overall, this study provides valuable insights into the observational data sets required for accurate parameter estimation in IHMs and the minimization of uncertainties in predictions of saturated zone storage in a semiarid region.

Keywords