Habitat preserve systems have been established adjacent to the densely populated regions of southern California to support indigenous plant and animal species that are listed as rare, threatened, or endangered. Monitoring the condition of habitat across these broad preserves is necessary to ensure their long-term viability and may be effectively accomplished using remote sensing techniques with high spatial resolution visible and near-infrared (VNIR) multispectral imagery. The utility of 1 m spatial resolution VNIR imagery for detailed change detection and monitoring of Mediterranean-type ecosystems is assessed here. Image acquisition and preprocessing procedures were conducted to ensure that image-detected changes represented real changes and not artifacts. Change classification products with six spectral-based transition classes were generated using multiband image differencing (MID) for three change periods: 1998-1999, 1998-2001, and 1998-2005. Land cover changes relevant to habitat quality monitoring such as human-induced disturbance, fire, vegetation growth/recovery, and drought related vegetation stress were readily detected using the multitemporal VNIR imagery. Suggestions for operational habitat monitoring using image products and mobile geographic information system technologies are provided.
Read full abstract