Landscape Monitoring and Analysis

Land cover information provides important inputs to local, regional, and state land use planning. The importance of accurate and timely information describing the kind and extent of land resources is increasing. This is especially true in metropolitan areas such as the Twin Cities of Minneapolis-St. Paul, Minnesota which encompasses seven counties and more than 100 civil government units.

Classification of remote sensing data has been an important source of land use-land cover information. Research at the University of Minnesota has proven the potential for classifying land cover with Landsat TM data. Several projects on classification of Landsat TM of the Twin Cities Metropolitan Area (TCMA) have demonstrated that it is possible to achieve overall classification accuracies of 90% for general (Level-1) land cover classes (agricultural cropland, forests, wetland, water, and urban) classes, and approximately 80% for Level-2 classes (Sawaya, et al., 2001). We have also found a strong relationship in the Landsat data to impervious surface that can be used to map percent impervious surface area.