Remote Sensing-Based Geostatistical Modeling For Conifererous Forest Inventory and Characterization

Abstract

Resource managers of private and public forests are often faced with a host of questions on forest extent, condition, and change in the course of land management. With more than 700 million acres of land covered by forest in the United States, the task of mapping and inventorying forested lands is a challenging one. Detailed and accurate maps of forest condition and structure are a necessity for rigorous ecosystem management. Forest maps are a fundamental information source for fire behavior modeling, animal habitat management, prediction and mapping of forest insect infestations, and plant and animal biodiversity assessment. Digital images acquired by earth imaging satellites are being used to help forest managers provide this information. Satellite images, when analyzed using advanced geostatistical techniques, can produce information on forest condition and structure, information that can be used to help answer questions such as those posed above. Satellite imagery has been used for many years to map land cover in forested regions, but natural resource managers are also starting to use remotely sensed satellite imagery to calculate the age, density, species, and successional state of forests under their care. In May 1999, the Kansas Applied Remote Sensing (KARS) Program at the University of Kansas was selected by NASA Earth Science Enterprise Applications Division to develop methods that use remote-sensing data and advanced geostatistical methods to create maps of forest age and successional state, or "cover types," and of forest biophysical factors, including density, biomass, leaf area, basal area, and height. By calibrating remotely sensed multispectral data with a small number of ground measurements, characteristics of the forest measured at sample points can be extrapolated across a large geographic region. This has significant advantages for forest management, especially when forests are in remote or inaccessible locations. The goal of this research is to develop new methods for the analysis of forest canopy structure, secondary forest regrowth, and forest fire history that take advantage of both the spectral and spatial correlation of ground phenomena and remotely sensed information.