Mapping where ecologists work: biases in the global distribution of terrestrial ecological observations

© The Ecological Society of America www.frontiersinecology.org T geographical context of field study sites greatly influences the ecological patterns, processes, and dynamics observed in these locations. For this reason, the disciplines of ecology and conservation biology have been criticized for disproportionately conducting field studies in temperate zones (Schoener 1983; Platnick 1991; Collen et al. 2008), biodiversity hotspots (Metrick and Weitzman 1994; Kier et al. 2005), and unpopulated areas (Botkin 1992; Collins et al. 2000). And though ecologists increasingly recognize the importance of urban ecology and “novel ecosystems” (Botkin and Beveridge 1997; Hobbs et al. 2006), ecological studies of urban and suburban areas represent just 0.4–6.0% of the ecological literature (Collins et al. 2000; Miller and Hobbs 2002). In contrast, landscapes transformed by agriculture and human settlements cover roughly 75% of Earth’s ice-free land and incorporate nearly 90% of terrestrial net primary productivity (NPP; Ellis and Ramankutty 2008). Although past critiques of the geographical distribution of field sites have been based on detailed disciplinary knowledge, few have been supported by quantitative assessments. There are three reasons why such quantification matters. First, because ecological field studies are costly in time and resources, they will always be in limited supply. The geographical distribution of this relatively small set of studies can therefore substantially influence conclusions reached by ecological theorists. Quantifying that distribution would enable those working to synthesize ecological knowledge to account for uneven sampling across study sites. Second, ecological knowledge is often used to prioritize conservation projects; it is therefore critical to know which biomes, regions, and landscapes remain understudied and undervalued. For example, the indicator framework of the Convention on Biological Diversity was recently criticized for incorporating a disproportionate amount of data from Europe and North America (Butchart et al. 2010; Pereira et al. 2010). There is also a complex relationship between “conservation attention” and the accumulation of ecological knowledge; better funded or longer protected sites are often more intensively studied, leaving open the question of whether protection follows study or vice versa (Ahrends et al. 2011). Third, the geographical distribution of study sites says much about the disciplinary norms of ecology; ecologists’ selections of field sites are influenced by a wide array of physical, financial, and institutional constraints, as well as by the discipline’s philosophical underpinnings, values, and history (Evans and Foster 2011). With these three considerations in mind, we set out to analyze the global distribution and environmental context of terREVIEWS REVIEWS REVIEWS