Continuous fields of vegetation characteristics at the global scale at 1-km resolution

The geographic distribution of vegetation over the Earth's land surface is traditionally described using classification schemes with discrete numbers of vegetation types. When such land cover data sets are used as boundary co nditions in Earth system models, abrupt boundaries and unrealistic homogene ity are introduced into parameter estimates. This paper proposes an alterna tive approach to describe global land cover with continuous fields of veget ation characteristics. A linear mixture model is applied to l-km advanced v ery high resolution radiometer data to estimate proportional cover for thre e important vegetation characteristics: life form (percent woody vegetation , percent herbaceous vegetation, and percent bare ground), leaf type (perce nt needleleaf and percent broadleaf), and leaf duration (percent evergreen and percent deciduous). Linear discriminants for input into the mixture mod el are derived from 30 metrics representing the annual phenological cycle. Through comparison with training data derived from a global network of Land sat multispectral scanner scenes, we conclude that the linear assumption im plicit in the linear mixture model is not severely violated. The linear rel ationships between percent cover as determined from the training data and t he linear discriminants are used to estimate end-member values, and the mix ture model is applied to derive the seven layers of global continuous field s. The availability of Moderate Resolution Imaging Spectroradiometer data i n the future holds promise for refining the simple technique used in this p aper to derive improved global continuous fields.