The agricultural expansion and intensification required to meet growing food and agri-based product demand present important challenges to future levels and management of biodiversity and ecosystem services. Influential actors such as corporations, governments, and multilateral organizations have made commitments to meeting future agricultural demand sustainably and preserving critical ecosystems. Current approaches to predicting the impacts of agricultural expansion involve calculation of total land conversion and assessment of the impacts on biodiversity or ecosystem services on a per-area basis, generally assuming a linear relationship between impact and land area. However, the impacts of continuing land development are often not linear and can vary considerably with spatial configuration. We demonstrate what could be gained by spatially explicit analysis of agricultural expansion at a large scale compared with the simple measure of total area converted, with a focus on the impacts on biodiversity and carbon storage. Using simple modeling approaches for two regions of Brazil, we find that for the same amount of land conversion, the declines in biodiversity and carbon storage can vary two-to fourfold depending on the spatial pattern of conversion. Impacts increase most rapidly in the earliest stages of agricultural expansion and are more pronounced in scenarios where conversion occurs in forest interiors compared with expansion into forests from their edges. This study reveals the importance of spatially explicit information in the assessment of land-use change impacts and for future land management and conservation.ecosystem services | deforestation | agricultural expansion | fragmentation | edge effects

Carbon stock estimates based on land cover type are critical for informing climate change assessment and landscape management, but field and theoretical evidence indicates that forest fragmentation reduces the amount of carbon stored at forest edges. Here, using remotely sensed pantropical biomass and land cover data sets, we estimate that biomass within the first 500 m of the forest edge is on average 25% lower than in forest interiors and that reductions of 10% extend to 1.5 km from the forest edge. These findings suggest that IPCC Tier 1 methods overestimate carbon stocks in tropical forests by nearly 10%. Proper accounting for degradation at forest edges will inform better landscape and forest management and policies, as well as the assessment of carbon stocks at landscape and national levels.

No abstract

A new methodology is proposed for clustering datasets in the presence of scattered observations. Scattered observations are defined as unlike any other, so traditional approaches that force them into groups can lead to erroneous conclusions. Our suggested approach is a scheme which, under assumption of homogeneous spherical clusters, iteratively builds cores around their centers and groups points within each core while identifying points outside as scatter. In the absence of scatter, the algorithm reduces to k-means. We also provide methodology to initialize the algorithm and to estimate the number of clusters in the dataset. Results in experimental situations show excellent performance, especially when clusters are elliptically symmetric. The methodology is applied to the analysis of the United States Environmental Protection Agency's Toxic Release Inventory reports on industrial releases of mercury for the year 2000. KeywordsBayes Information Criterion, biweight estimator, exact-c-separation, k-clips, MCLUST, methylmercury, tight clustering Disciplines Statistics and Probability CommentsThis is the peer reviewed version of the following article: Maitra,R., and Ramler, I. Clustering in the presence of scatter. A new methodology is proposed for clustering datasets in the presence of scattered observations. Scattered observations are defined as unlike any other, so traditional approaches that force them into groups can lead to erroneous conclusions. Our suggested approach is a scheme which, under assumption of homogeneous spherical clusters, iteratively builds cores around their centers and groups points within each core while identifying points outside as scatter. In the absence of scatter, the algorithm reduces to k-means. We also provide methodology to initialize the algorithm and to estimate the number of clusters in the dataset. Results in experimental situations show excellent performance, especially when clusters are elliptically symmetric. The methodology is applied to the analysis of the United States Environmental Protection Agency (EPA)'s Toxic Release Inventory (TRI) reports on industrial releases of mercury for the year 2000.