The current study investigates a method to provide land surface parameters [i.e., land surface temperature (LST) and normalized difference vegetation index (NDVI)] at a high spatial ( and 60 m) and temporal (daily and 8-day) resolution by combining advantages from Landsat and moderate-resolution imaging spectroradiometer (MODIS) satellites. We adopt a previously developed subtraction method that merges the spatial detail of higher-resolution imagery (Landsat) with the temporal change observed in coarser or moderate-resolution imagery (MODIS). Applying the temporal difference between MODIS images observed at two different dates to a higher-resolution Landsat image allows prediction of a combined or fused image () at a future date. Evaluation of the resultant merged products is undertaken within the Southeastern Arizona region where data is available from a range of flux tower sites. The fused products capture the raw Landsat values and also reflect the MODIS temporal variation. The predicted LST improves mean absolute error around 5°C at the more heterogeneous sites compared to the original satellite products. The fused NDVI product also shows good correlation to ground-based data and is relatively consistent except during the acute (monsoon) growing season. The sensitivity of the fused product relative to temporal gaps in Landsat data appears to be more affected by uncertainty associated with regional precipitation and green-up, than the length of the gap associated with Landsat viewing, suggesting the potential to use a minimal number of original Landsat images during relatively stable land surface and climate conditions. Our extensive validation yields insight on the ability of the proposed method to integrate multiscale platforms and the potential for reducing costs associated with high-resolution satellite systems (e.g., SPOT, QuickBird, IKONOS).