Accurate assessment of grassland ecosystem structural and functional indicators, such as percent canopy cover (CC) and aboveground biomass (AGB), are crucial for sustainable rangeland management. Conventional CC and AGB mapping studies lack extrapolation robustness and induce uncertainty as they upscale limited in situ samples to broad spatial extents. Therefore, we proposed and tested a spatial cross-scale approach. We integrated small unoccupied aerial systems (sUAS) imagery as a bridge to upscale in situ data to satellite-based estimates across Mongolia (MN) and Kazakhstan (KZ). We sampled in situ herbaceous CC (%) and AGB (g/m2) and conducted sUAS overflights in 84 sites (n = 252) across two provinces in KZ (North to South gradient in 2022) and dominant grassland steppe types of MN (East to West gradient in 2023). We employed a random forest (RF) regression model first to scale up quadrat-based field estimates to site scale (3.5 cm resolution) using spectraltextural-structural metrics from drone imagery and then to regional-scale (10 m resolution) using predictor variables from Sentinel-2 imagery and environmental covariates from various satellite sensors. We derived perpixel uncertainty estimates to quantify the reliability of CC and AGB predictions. Results showed superior RF model performance during the cross-scale approach with an estimated R2 (RMSE) of 0.83 (10.4%) and 0.83 (31.74 g/m2) for upscaling CC and AGB estimates, respectively, from quadrat-to-site and 0.85 (5.14%) and 0.69 (31.8 g/m2) from site-to-regional. Notably, the cross-scale approach improved predictions over the conventional approach, with 44.0% and 43.7% lower RMSEs for CC and AGB, respectively. We produced the first fineresolution (10 m) gridded herbaceous CC and AGB estimates for KZ and MN by leveraging extensive in situ sampling and robust upscaling approaches. This study demonstrates the sUAS capability for upscaling ground data to satellite imagery across large spatial scales and estimating wall-to-wall CC and AGB across diverse regions and ecosystems.