Land cover and global productivity: A measurement strategy for the NASA programme

Abstract

NASA's Earth science programme is developing an improved understanding of terrestrial productivity and its relationship to global environmental change. Environmental change includes changes that are anthropogenic, caused for example by increasing population and resource use, as well as those that are natural, caused by interannual or decadal variability in climate and intrinsic vegetation dynamics. In response to current science and policy concerns, the Earth science programme has carbon and the major biogeochemical cycles as a primary focus but is broad enough to include related topics such as land-atmosphere interactions associated with the hydrological cycle and the chemical composition of the atmosphere. The research programme includes the study of ecosystems both as respondents to change and as mediators of feedback to the atmosphere. Underlying all the research elements are important questions of natural resources and sustainable land management. The land cover and land use change element of the programme is aimed specifically at studying the causes and effects of land transformation and changes in land use practices. The NASA Earth science programme has a primary focus on using satellite remote sensing systems but also recognizes the need for an integrated approach to achieving its science goals by combining satellite and in situ process measurements and numerical modelling. This paper outlines the programme strategy for addressing its major focus. The approach adopted provides a balance between long-term satellite measurements of the Earth's surface at moderate and high spatial resolutions that are needed to quantify change, and the new experimental satellite missions that are aimed at addressing specific process research questions and testing new sensing technology. In addition to satellite measurements, ground-based in situ measurements are needed to validate the satellite data products, to describe and quantify processes, and to parameterize and validate process models. Numerical models need to be enhanced to provide both the study of processes and a predictive capability for the study of global change.