Tropical forest disturbance contributes to global climate change from increased carbon emissions, and causes loss of biodiversity. Thus, identifying its direct causes and underlying drivers are necessary for effective land use, climate change control and conservation strategies. We integrated remote sensing forest cover data from 2000 to 2018 with georeferenced national socioeconomic and field-collected household data to determine underlying drivers behind forest cover transitions (e.g., deforestation, degradation, and recovery) in the Selva Maya (‘Mayan Forest’) of southeast Mexico. Spatial and statistical models (multinomial logistic regression, log-linear regression, and analysis of variance) and social science methods (household surveys and qualitative comparative analysis) were applied to evaluate and identify socioeconomic, institutional, and environmental drivers intervening at landscape and community scales. Forest cover transitions varied geographically, and associated drivers differed by scale of analysis. Using multiple methods improved the understanding of drivers. Population growth, poverty, and roads are major drivers influencing forest cover transitions (e.g., deforestation, degradation, and recovery) in the landscape. Community scale analysis identified more drivers and offered greater detail of causal relationships. Besides population and poverty, less off-farm employment, agriculture and cattle production, immigrant population, and private property were related to deforestation and degradation. Indigenous populations, forest dependence, off-farm employment, and common property were associated with forest conservation. Sustainable rural development should include poverty alleviation through diversification of economic activities and increased off-farm employment opportunities. Conservation measures should pursue the enhancement of forest value for local subsistence and economic benefits by strengthening community forest management and enterprises.

中文翻译：

---

墨西哥塞尔瓦玛雅森林覆盖转变的驱动因素：整合区域和社区尺度进行景观评估

热带森林干扰因碳排放增加而导致全球气候变化，并导致生物多样性丧失。因此，确定其直接原因和潜在驱动因素对于有效的土地利用、气候变化控制和保护战略是必要的。我们整合远离2000传感森林覆盖数据，2018地理参照国家的社会经济和现场采集的家庭数据，以确定在森林覆盖的转变（例如，砍伐森林，退化和恢复）背后潜在驱动塞尔瓦玛雅（“玛雅森林”）墨西哥东南部。应用空间和统计模型（多项逻辑回归、对数线性回归和方差分析）和社会科学方法（家庭调查和定性比较分析）来评估和识别干预景观和社区规模的社会经济、制度和环境驱动因素. 森林覆盖转变因地域而异，相关驱动因素因分析规模而异。使用多种方法提高了对驱动程序的理解。人口增长、贫困和道路是影响景观中森林覆盖转变（例如，森林砍伐、退化和恢复）的主要驱动因素。社区规模分析确定了更多的驱动因素，并提供了更详细的因果关系。除了人口和贫困，减少非农就业，农业和牲畜生产、移民人口和私有财产与森林砍伐和退化有关。土著人口、森林依赖、非农就业和公共财产与森林保护有关。可持续农村发展应包括通过经济活动多样化和增加非农就业机会来减轻贫困。保护措施应通过加强社区森林管理和企业来提高森林价值，以实现当地的生存和经济利益。可持续农村发展应包括通过经济活动多样化和增加非农就业机会来减轻贫困。保护措施应通过加强社区森林管理和企业来提高森林价值，以实现当地的生存和经济利益。可持续农村发展应包括通过经济活动多样化和增加非农就业机会来减轻贫困。保护措施应通过加强社区森林管理和企业来提高森林价值，以实现当地的生存和经济利益。