Seeing the wood for the e-trees: ensuring comparability in forest-related data on the web.

摘要

Recent decades have seen a mushrooming of available data on the extent and depletion of forest cover, much of it generated by remote sensing, supplemented by statistics compiled by national governments and industry. This is extremely important in modeling climate change, for example, when coupled with historic data often originally published only in 'grey' literature or not at all. Much progress has been made in retrieving these data to allow the creation of more complete time series, through digitization programmes, etc. However, apparently comparable data from different countries have often been compiled to different standards and definitions; combining datasets to establish a global picture can therefore be misleading. Comprehensive metadata is required to enable identification of compatible data and appropriate conversion factors where relevant. Librarians are working alongside statisticians and forest scientists to chart a way through this minefield and flag up 'danger areas' for those using sources now publicly available on the web. This article examines the issues, their significance and possible solutions.