Santa Cruz Island (SCI), California is an example of a sensitive landscape that is experiencing severe soil erosion and gullying and has undergone intensive grazing in its recent history. This research has tested the hypothesis that gully erosion and potential controlling parameters can be identified, documented and quantified using remote sensing and GIS analyses and field calibration. We examined gully development on Santa Cruz Island using three generations of air photographs: 1929, 1985, and 1998. Each photoset was orthorectified to remove distortions that occur from terrain and camera lens producing a digital, composite map for each set. The 1929, 1985, and 1998 orthoimages provided a means of quantifying changes in gully parameters through time as well as providing a timeline for studying long-term processes and effects of gully erosion. Using a GIS, a series of data layers has been assembled representing a broad range of possible controlling parameters. The analysis conducted compared upstream watershed area of the gully to slope, aspect, curvature, geology, vegetation, and elevation to try and establish control relationships between the 1929, 1985, and 1998 composite maps. Vegetation cover is a key variable in this study with respect to land-use. Collection of ground-truth data has been used to determine how accurately the digital maps represent the real-world position of the gullies. The impact of each of these parameters on gully erosion was tested using a multivariate analysis. A Normalized Difference Vegetation Index (NDVI) from the late 70’s and the late 90’s was used as a secondary comparison to study change in land use on the east side of the island versus the west side by studying changes in land cover. This research provides a comprehensive study of gully erosion that assesses the impact of land-use and other factors on a sensitive island ecosystem. The results are an indication of how successful the rehabilitation efforts to date have been and how quickly sensitive ecosystems can recover from severe degradation.