作者: Laure M. Montandon , Souleymane Fall , Roger A. Pielke , Dev Niyogi
DOI: 10.1175/2010EI371.1
关键词:
摘要: Abstract The Global Historical Climate Network version 2 (GHCNv.2) surface temperature dataset is widely used for reconstructions such as the global average (GAST) anomaly. Because land use and cover (LULC) affect temperatures, it important to examine spatial distribution LULC representation of GHCNv.2 stations. Here, nightlight imagery, two datasets, a population cropland historical reconstruction are estimate present worldwide occurrence types number stations within each. Results show that station locations biased toward urban (>50% versus 18.4% world’s land) past century reclaimed areas (35% 3.4% land). However, occurring open shrubland, bare, snow/ice, evergreen broadleaf forests underrepresented (14% 48.1% land), well nonurban have remained uncultivated in pa...