A vegetation classification logic-based on remote-sensing for use in global biogeochemical models

A simple new classification logic for global vegetation is proposed. The critical features of this classification are that: it is based on simple, observable, unambiguous character- istics of vegetation structure that are important to ecosystem biogeochemistry and can be measured in the field for validation; the structural characteristics can be determined by remote sensing, so that repeatable and efficient global re-classifications of existing vegetation will be possible; and the defined vegetation classes directly translate into the biophysical parameters of interest by global climate and biogeochemical models. A first test of this logic for the continental United States is presented based on an existing 1 km Normalized Difference Vegetation Index database. Currently recognized global biome classes can easily be derived from this classification by adding climate descriptors and defining mixtures of these fundamental six vegetation classes.