Where do people vulnerable to deforestation live? Triaging forest conservation interventions for sustainable non-timber forest products

Rural households in developing countries often depend on non-timber forest products (NTFPs), including foods, traditional medicines, fuelwood, fodder, and construction and craft materials such as bamboo and rattan, for their livelihoods, but the forests that supply NTFPs are subject to competition from other land uses. NTFP collectors who are unable to sustain their livelihoods because of changes caused in the NTFP collection environment by deforestation and degradation may even convert surrounding forests to agricultural land to compensate for their reduced incomes. This feedback loop of deforestation and degradation can create further conflicts among natural resource users and undermine long-term development goals. Here, we aimed to advance the discussion of forest conservation interventions in protected areas in developing countries—particularly the discussion of patrolling strategies in and around existing wildlife sanctuaries—by presenting a new approach that uses the case of NTFP collection and a human geography perspective. We used a structured questionnaire composed of closed questions to measure the awareness of village households of the impacts of deforestation and/or illegal extraction of trees on NTFP collection over a short period of time (2014–2016) near the Prey Long Wildlife Sanctuary in the eastern part of the Kampong Thom Province, Cambodia. Our survey was based on the conceptual framework of Bohensky and Lynam (Ecol. Soc. 10, 11; 2005) that people in complex adaptive systems change their behaviour after becoming aware of the impact or consequences of a change. The probability of a village having an affected household (i.e., a household that was aware of the impact) was predicted by using generalized linear mixed models. By using identified geospatial indicators of environmental and socioeconomic aspects, including distance between the village and the wildlife sanctuary, size of deforested area per capita in 2014–2016 within a 10-km radius of the village and proportion of households in the village that collect NTFPs for cash income, we predicted the spatial distribution of villages with affected NTFP-collecting households. Priority areas for NTFP collection were then mapped with reference to the identified indicators. By using these results, we present an approach to breaking the deforestation feedback loop by identifying priority areas for patrolling and other forest conservation interventions in the wildlife sanctuary.