Doctoral Dissertation Research: Multilevel Modeling of Household and Accessibility-Zone Drivers of Land Change in the Northeastern Peruvian Amazon

博士论文研究：秘鲁亚马逊东北部土地变化的家庭和可达区驱动因素的多层次建模

• 批准号: 0623229
• 负责人: Kelley Crews
• 金额: --
• 依托单位: University of Texas at Austin
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-08-01 至 2008-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0623229&HistoricalAwards=false
• 关键词: Doctoral; Dissertation; Research; Multilevel; Modeling

This study will examine differential land-use patterns associated with peasant households in the urban-rural gradient around Iquitos, the largest city in the Peruvian Amazon. Access to urban markets and services in the Iquitos region is hypothesized to affect land-use decision-making among peasant households in neighboring rural settlements. To explicitly address the spatial variability of transportation and market accessibility, multilevel models will be constructed. Multilevel modeling allows testing the significance of geographic contextual factors that might variably affect local factors among different localities. Four generalized land-use types (mature forest, commercial and subsistence agriculture, and fallow) will be modeled through multilevel regression considering demographic, political-economic, socioeconomic, and environmental factors as predictor variables collected through geocoded structured interviewing at the household level, in approximately 625 peasant farms. A total of 25 accessibility zones will be delineated based on proximity to urban markets and transportation routes, and differentiated among types of roads and rivers employing cartographic and remotely sensed data. In these models a clear distinction between commercial and subsistence farm activities will be drawn in order that the determinants of peasant land use can be examined in terms of access to and use of urban markets. Further, the analysis will also model forest extent to facilitate exploring deforestation drivers associated with peasant livelihood strategies and fallow/forest regrowth dynamics, which will provide insights about land-use intensity and the long-term sustainability of peasant agricultural practices. The analysis of peasant land-use patterns has not been addressed regionally in the urban-rural gradient centered on Iquitos. The study area presents the unique opportunity to test a multilevel statistical procedure that could better account for contextual drivers of landscape change in regions with significant environmental and socioeconomic heterogeneity, as opposed to more common single-level regression models. Results will indicate whether such multilevel conceptualization of the rural-urban landscape is valuable for the analysis of peasant land use in the Peruvian Amazon, so positing its applicability and usefulness in other regions. This research will contribute theoretically and materially to land-use and land-cover change and Amazonian studies generally while specifically addressing the importance of transportation accessibility and multilevel impacts on landscape transformation. In addition to contributing to academic studies, this work will also be presented in written and oral form in both English and Spanish to current project collaborator Peruvian Institute of Amazonian Studies (IIAP) as a tool for management and decision-making. Further, students or recent graduates from the National University of the Peruvian Amazon in Iquitos will be hired as field assistants and trained for household survey, including survey coding and collection of geographically referenced data.

这项研究将考察秘鲁亚马逊最大城市伊基托斯周围城乡梯度上与农民家庭相关的不同土地利用模式。假设伊基托斯地区进入城市市场和服务的机会会影响邻近农村住区的农户的土地使用决策。为了明确解决交通运输和市场可达性的空间变异性，将构建多水平模型。多水平建模允许测试地理背景因素的重要性，这些因素可能在不同地点之间可变地影响当地因素。四种普遍的土地利用类型(成熟森林、商业和自给农业以及休耕)将通过多水平回归进行建模，考虑人口、政治经济、社会经济和环境因素作为预测变量，通过家庭层面的地理编码结构化访谈收集到约625个农民农场。将根据靠近城市市场和交通路线的程度划定总共25个无障碍区，并利用制图和遥感数据区分不同类型的道路和河流。在这些模式中，将明确区分商业性和自给自足的农场活动，以便从进入和使用城市市场的角度考察农民土地使用的决定因素。此外，分析还将建立森林范围模型，以便于探索与农民生计战略和休耕/森林再生动态有关的毁林驱动因素，这将为土地利用强度和农民农业做法的长期可持续性提供见解。在以伊基托斯为中心的城乡梯度中，对农民土地利用模式的分析没有按区域处理。与更常见的单级回归模型相比，该研究领域提供了测试多水平统计程序的独特机会，该程序可以更好地解释环境和社会经济异质性显著的地区景观变化的背景驱动因素。结果将表明，这种多层次的城乡景观概念化是否对秘鲁亚马逊地区的农民土地利用分析有价值，从而证明其在其他地区的适用性和有用性。这项研究将在理论上和物质上对土地利用和土地复盖变化以及亚马逊研究作出贡献，同时具体探讨交通便利程度和对景观变化的多层次影响的重要性。除了对学术研究作出贡献外，这项工作还将以英语和西班牙语的书面和口头形式介绍给目前的项目合作者秘鲁亚马逊研究所(IIAP)，作为管理和决策的工具。此外，伊基托斯的秘鲁亚马逊国立大学的学生或最近毕业的学生将被聘为外地助理，并接受家庭调查方面的培训，包括调查编码和收集地理参考数据。