Assessing sustainability indicators for tropical forests: Spatio-temporal heterogeneity, logging intensity, and dung beetle communities

Sustainable management of tropical forests has been identified as one of the main objectives for conservation of global biodiversity and management carbon stocks. To achieve this goal, managers need tools to assess the sustainability of current management practices. Several international initiatives have undertaken the development of sets of criteria and indicators to help managers move towards sustainability. Among the indicators considered, the structure and composition of dung beetle communities have been identified as excellent indicators of ecological sustainability. However, as occurs with most indicators of the ecological sustainability of forest management, dung beetle surveys require intensive field work making their application over large areas expensive, time consuming, and logistically challenging. A need for prioritization is evident. This work presents a novel approach to the assessment of the Center for International Forestry Research (CIFOR) ecological sustainability indicator 1.2.1.2: The change in diversity of habitats as a result of human interventions is maintained within critical limits as defined by natural variation and/or regional conservation objectives. Using variography of vegetation index data derived from remotely sensed imagery, we show (1) how the differences in forest structural heterogeneity observed between forest management units and natural areas can be used to identify priority areas for field survey of ecological sustainability indicators (hereafter priority-for-survey) and (2) how these priorities correspond to dung beetle community structure and composition. Links between temporal change in forest structural heterogeneity, logging intensity, and dung beetle community structure and composition were established by means of correlation analysis and matrix regression modeling. We found that areas ranked as low priority-for-survey based on image analysis showed no significant difference in dung beetle species richness or diversity from natural reference areas. Further, we found significantly higher dung beetle species richness and diversity estimates in areas ranked as moderate or moderate-low priori ty-for-survey over the low and reference areas. Finally, the dung beetle community composition in the high pfiofity-for-survey category was significantly less rich and less diverse than any other category. We identified a logging intensity threshold of four trees per hectare as a transition to significant differences in forest structural heterogeneity and the richness and diversity of associated dung beetle communities. (c) 2007 Elsevier B.V. All rights reserved.