Diversity Delivering a Global, Terrestrial, Biodiversity Observation System through Remote Sensing GRAEME M. BUCHANAN, ∗ ‡ ANDREW NELSON,† PHILIPPE MAYAUX,† ANDREW HARTLEY,† AND PAUL F. DONALD ∗ ∗ Royal Society for the Protection of Birds, The Lodge, Sandy, Bedfordshire SG19 2DL, United Kingdom †Institute for Environment and Sustainability, Joint Research Centre of the European Commission, Ispra, Italy Introduction Land-cover change is of major concern to conservation- ists because of its generally negative impact on biodi- versity (Brooks et al. 2002). There is a clear need to track these changes, and such information could make a major contribution to a global biodiversity observation system. Monitoring of biodiversity is an essential com- ponent of conservation because it allows problems to be identified, priorities to be set, solutions to be de- veloped, and resources to be targeted (Balmford et al. 2003). Monitoring also allows assessments of progress toward targets and indicators in unilateral and interna- tional conservation-policy instruments (e.g., Convention on Biological Diversity [CDB]), of the impacts of inter- national conservation policy (Donald et al. 2007), and of other policy sectors (Donald et al. 2001). Nevertheless, a paucity of information has led to a poor understanding of the cost-effectiveness of conservation policies (Ferraro & Pattanayak 2006), exposing them to criticism (Stokstad 2005). The overwhelming majority of species and ecosystems receive no systematic monitoring, and there is a conspicu- ous mismatch between the distribution of monitoring ef- fort and the distribution of terrestrial biodiversity at a global scale (Green et al. 2005). The need to improve monitoring is widely recognized (Balmford et al. 2003; Pereira and Cooper 2006), and although some system- atic monitoring of terrestrial biodiversity for conserva- tion is undertaken locally in the developing world (e.g., Danielsen et al. 2008), there is no protocol to tackle the issue at a global scale. Traditionally, monitoring of popu- ‡email graeme.buchanan@rspb.org.uk Paper submitted June 4, 2008; revised manuscript accepted July 24, 2008. lations or habitats has involved field-based observations, an approach that is time consuming, requires specialist skills and resources, and is difficult in remote or politi- cally insecure areas. Although field-based monitoring needs to continue, a global biodiversity-observation system, based on changes in land cover assessed from data collected by earth ob- servation (also known as EO or remote sensing), would address the current shortfall in monitoring. Recent an- nouncements on free higher-resolution satellite-image data (Group on Earth Observation 2007; U.S. Geologi- cal Survey 2008) begin to make an EO-based monitoring system a real possibility because until now the high cost of appropriate remotely sensed images has been a ma- jor limitation for conservation monitoring. Although EO data may not be appropriate for monitoring all threats to biodiversity directly (e.g., poaching), it is appropriate for monitoring of land-cover change. Land-cover change is of particular concern because, for example, the most prevalent threats to a network of sites of high conser- vation importance entail land-cover changes that could be monitored remotely (Buchanan et al. 2008). Although the potential of EO data in biodiversity conservation and monitoring has been recognized (Turner et al. 2003; Pet- torelli et al. 2005) and utilized (e.g., Buchanan et al. 2008), there is still no systematic protocol for its use in terrestrial biodiversity monitoring for conservation at continental, let alone global, scales. We argue that developing an EO- based biodiversity observation protocol would require knowledge of the distribution of biodiversity, an ability to monitor land-cover change, and an understanding of the impacts of land-cover change on populations. We 499 Conservation Biology, Volume 23, No. 2, 499–502 C 2008 Society for Conservation Biology DOI: 10.1111/j.1523-1739.2008.01083.x