Climate and land use often interact in ways that influence biodiversity, implying that these factors cannot be considered in isolation. For example, land use ma
22 December 2009 in Proceedings of the National Academy of Sciences
Protected areas are crucial for biodiversity conservation because they provide safe havens for species threatened by land-use change and resulting habitat loss. However, protected areas are only effective when they stop habitat loss within their boundaries, and are connected via corridors to other wild areas. The effectiveness of protected areas is threatened by development; however, the extent of this threat is unknown. We compiled spatially-detailed housing growth data from 1940 to 2030, and quantified growth for each wilderness area, national park, and national forest in the conterminous United States. Our findings show that housing development in the United States may severely limit the ability of protected areas to function as a modern “Noah’s Ark.” Between 1940 and 2000, 28 million housing units were built within 50 km of protected areas, and 940,000 were built within national forests. Housing growth rates during the 1990s within 1 km of protected areas (20% per decade) outpaced the national average (13%). If long-term trends continue, another 17 million housing units will be built within 50 km of protected areas by 2030 (1 million within 1 km), greatly diminishing their conservation value. US protected areas are increasingly isolated, housing development in their surroundings is decreasing their effective size, and national forests are even threatened by habitat loss within their administrative boundaries. Protected areas in the United States are thus threatened similarly to those in developing countries. However, housing growth poses the main threat to protected areas in the United States whereas deforestation is the main threat in developing countries.
As one samples species from a particular assemblage, the initial rapid rate with which new species are encountered declines with increasing effort. Nine candidate models to characterize species–accumulation functions were compared in a search for a model that consistently fit geographically extensive avian survey data from a wide range of environmental conditions. Landscape‐specific species–accumulation curves generated under a bootstrap resampling plan were best described by a generalized Weibull cumulative distribution function. Traditional species–area models of cumulative species richness as a function of accumulated sample had notable functional bias. The Weibull model fitted species–accumulation data equally well among sixty‐six forested landscapes in the eastern U.S. Landscapes with a greater proportion of agricultural and urban land uses accumulated species more slowly than landscapes which retained a greater proportion of natural habitats (r=−0.64, P<0.001). This finding supports predictions of ecosystem behavior under human land use. There was no evidence that intermediate levels of land use intensity maximized accumulation rates. The approach reviewed in this paper makes no assumptions about the form of the species–abundance distribution or how species are distributed in space, thereby offering some advantages over more conventional diversity indices for characterizing how species assemblages respond to anthropogenic disturbance. Investigation of how species accumulation varies over time in a given geographic area is needed to evaluate fully the potential application of this approach to regional land use planning.