Search

Shortwave solar radiation is an important component of the surface energy balance and provides the principal source of energy for terrestrial ecosystems. This paper presents a machine learning approach in the form of a random forest (RF) model for estimating daily downward solar radiation flux at the land surface over complex terrain using MODIS (MODerate Resolution Imaging Spectroradiometer) remote sensing data. The model-building technique makes use of a unique network of 16 solar flux measurements in the semi-arid Reynolds Creek Experimental Watershed and Critical Zone Observatory, in southwest Idaho, USA. Based on a composite RF model built on daily observations from all 16 sites in the watershed, the model simulation of downward solar radiation matches well with the observation data (r2 = 0.96). To evaluate model performance, RF models were built from 12 of 16 sites selected at random and validated against the observations at the remaining four sites. Overall root mean square errors (RMSE), bias, and mean absolute error (MAE) are small (range: 37.17 W/m2-81.27 W/m2, -48.31 W/m2-15.67 W/m2, and 26.56 W/m2-63.77 W/m2, respectively). When extrapolated to the entire watershed, spatiotemporal patterns of solar flux are largely consistent with expected trends in this watershed. We also explored significant predictors of downward solar flux in order to reveal important properties and processes controlling downward solar radiation. Based on the composite RF model built on all 16 sites, the three most important predictors to estimate downward solar radiation include the black sky albedo (BSA) near infrared band (0.858 μm), BSA visible band (0.3–0.7 μm), and clear day coverage. This study has important implications for improving the ability to derive downward solar radiation through a fusion of multiple remote sensing datasets and can potentially capture spatiotemporally varying trends in solar radiation that is useful for land surface hydrologic and terrestrial ecosystem modeling.

Accurate characterization of snow-covered area (SCA) and snow water equivalent (SWE) in complex terrain is needed to improve estimation of streamflow timing and volume, and is important for land surface modeling. Direct field observations of SWE, SCA and atmospheric forcing inputs for models of snow accumulation and ablation are typically sparsely sampled in space. Satellite imagery is, therefore, a critical tool for verification and confirmation of snow model estimates of SCA. The Landsat system provides snow-covered area estimates at a spatial resolution of 30 m with a 16-day return interval, while daily estimates of SCA and fractional SCA ( f SCA ) are available at 500 m from the Moderate Resolution Imaging Spectroradiometer (MODIS). This study describes and tests a linear model to downscale MODIS MOD10A1 f SCA (500 m) data to higher-resolution (30 m) spatially explicit binary SCA estimates. The algorithm operates on the assumption that two variables, potential insolation and elevation, control differential ablation of snow cover throughout spring melt at 30 m to 500 m scales. The model downscales daily 500 m f SCA estimates from MODIS to provide daily SCA estimates at a spatial resolution of 30 m, using limited Landsat SCA for calibration and independent Landsat SCA estimates for validation. Downscaled SCA estimates demonstrate statistically significant improvement from randomly generated model ensembles, indicating that insolation and elevation are dominant factors controlling the snow cover distribution in the semi-arid, mountainous region in southwestern Idaho, USA where this study is performed. Validation is performed with Landsat data not used for calibration, and is also performed using Landsat 500 m aggregate f SCA instead of MODIS f SCA as an ideal case. Downscaled estimates show reasonable accuracy (test metric outperforms random ensembles at p = 0.01 significance level for multiple ranges of snow cover) with only one calibrated parameter.