Remotely sensed data have often been used to assess and monitor vegetation state through the computation of the Normalized Difference Vegetation Index (NDVI). NDVI information can be critical for monitoring vegetation response to varying moisture conditions. Operational activities of this type are usually based on coarse resolution remotely sensed data that blend the signals from multiple vegetation types in regions of heterogeneous vegetation. This research utilizes multi-temporal SPOT data to detect vegetation response at high resolution to a range of moisture conditions for a selected region at the southeastern edge of the Black Hills in South Dakota. Three images pertaining to near normal (19 August 1991), dry (3 September 1989), and wet (31 August 1993) growing season conditions were acquired for this study. A subset of each image was extracted and statistical pattern recognition procedures implemented to classify them into 5 land cover categories. Vegetation response to the climate variation was determined by computing the NDVI for each subset and comparing the land cover category to its respective NDVI value. The largest variation from the normal NDVI values for cover types was detected for agricultural land (alfalfa) during wet and dry conditions. Mean and standard deviation results indicated that mixed land cover (i.e., forest/grassland mix) and grasslands were more affected by the differences in moisture conditions than homogeneous land cover (e.g., forest). In a broader context, the high-resolution examination of heterogeneous vegetation response to different moisture regimes is critical for the improved interpretation of coarse resolution vegetation response to climate variation and change.