Modeling population-environment dynamics in the Ecuadorian Amazon

描述(申请人提供)：该项目的主要目标是研究热带雨林背景下人口和环境动态之间的一些复杂的多维关系，热带雨林由于农业前沿的发展正在经历重大变化。更好地了解这些关系对于森林保护和可持续发展至关重要，特别是在厄瓜多尔北部亚马逊地区，这是一个拥有非凡生物多样性的地区，森林砍伐速度在世界上排名第九。虽然砍伐森林的直接原因各不相同，但农村-农村流动农民已被确定为主要的直接动因。我们建议基于1990年和1999年收集的土地地块面板数据在国家能源部进行研究，构建一个土地地块和社区的分层框架，使用多水平统计模型来检验假设。为了审查人口、社会经济、制度和生态因素，提出了三个相互关联的目标：(I)确定与家庭家庭加班的空间分布相关的因素；(Ii)确定与家庭生育率加班变化相关的因素；(Iii)将土地利用和土地覆盖随时间的变化分解为与人口增长、位置/社区/市场接近、农场建立年份、自然资源和财富有关的因素。将根据人口-环境理论的综合方法，结合人口学理论、查亚诺夫农业家庭模型、冯·图宁经济土地使用模型以及利用体制和背景变量的最新方法，开发空间显式模型。数据来源于1990年至2000年间收集的地理编码纵向家庭和社区数据，再加上1972年至今的一系列遥感图像。来自家庭和社区的纵向地理编码数据的可获得性，以及多层次模型估计的最近改进，为评估许多因果因素的相关性提供了独特的机会。该项目将有助于增进对人口-环境关系的了解，推动我们完善热带变化理论，并为有兴趣实现森林保护和可持续发展的政策制定者提供重要信息。

DESCRIPTION (provided by applicant): The primary objective for this project is to examine some of the complex multidimensional relationships between population and environment dynamics in a tropical rainforest context that is undergoing substantial change due to the advance of the agricultural frontier. A better understanding of these relationships is vital for forest conservation and sustainable development, particularly in the Northern Ecuadorian Amazon (NEA), an area of extraordinary biodiversity that has the 9th fastest deforestation rate in the world. While the proximate causes of deforestation vary, rural-rural migrant farmers have been identified as the primary direct agents. We propose to undertake research in the NEA based upon panel data collected in 1990 and 1999 from land plots, constructing a hierarchical framework of land plots and communities to test hypotheses using multilevel statistical models. Three interrelated Aims are proposed to examine demographic, socioeconomic, institutional, and ecological factors: (i) Determine factors associated with the spatial distribution of family households overtime; (ii) Determine factors associated with changing household fertility overtime; (iii) Decompose changes in land use and land cover over time into factors related to population growth, location / proximity to communities / markets, year of farm establishment, natural resources, and wealth. Spatially-explicit models will be developed based upon an integrated approach to population- environment theory that focuses on proximate and underlying causes of change, combining theories from demography, Chayanovian agricultural household models, Von Thunen economic land use models, and recent approaches utilizing institutional and contextual variables. Data are derived from geo-coded longitudinal household and community data collected between 1990 and 2000, augmented by a time series of remotely sensed images from 1972 to the present. The availability of longitudinal geo-coded data from households and communities along with recent improvements in multilevel model estimation provide a unique opportunity to assess the relevance of many causal factors. This project will help advance knowledge of population-environment relationships, moving us towards refining theories of change in the tropics, and proffering vital information to policy-makers interested in achieving forest conservation and sustainable development.