A neural network method for mixture estimation for vegetation mapping

While most forest maps identify only the dominant vegetation class in delin eated stands, individual stands are often better characterized by a mix of vegetation types. Many land management applications, including wildlife hab itat studies, can benefit from knowledge of mixes. This article examines va rious algorithms that use data from the Landsat Thematic Mapper (TM) satell ite to estimate mixtures of vegetation types within forest stands. Included in the study are maximum likelihood classification and linear mixture mode ls as well as a new methodology based on the ARTMAP neural network. Two par adigms are considered: classification methods, which describe stand-level v egetation mixtures as mosaics of pixels, each identified with its primary v egetation class; and mixture methods, which treat samples as blends of vege tation, even at the pixel level. Comparative analysis of these mixture esti mation methods, tested on data from the Plumas National Forest, yields the following conclusions: 1) Accurate estimates of proportions of hardwood and conifer cover within stands can be obtained, particularly when brush is no t present in the understory; 2) ARTMAP outperforms statistical methods and linear mixture models in both the classification and the mixture paradigms; 3) topographic correction fails to improve mapping accuracy; and 4) the ne w ARTMAP mixture system produces the most accurate overall results. The Plu mas data set has been made available to other researchers for further devel opment of new mapping methods and comparison with the quantitative studies presented here, which establish initial benchmark standards. (C)Elsevier Sc ience Inc., 1999.