Doctoral Dissertation Research: Effects of Management Regime on Forest Structure and Disturbance Patterns in the Usambara Mountains, Tanzania

热带森林在人类主导的景观中的命运取决于跨学科研究，以更好地了解生态系统过程和生态功能，同时确定如何管理这种森林，以满足依赖资源的人民的需求。当地森林使用者对热带森林管理的参与继续获得法律和体制基础，但在这类政策下的特定管理行动和资源使用模式的生态影响仍未得到充分研究。本研究将热带森林生态学和制度分析相结合，考察了坦桑尼亚七个森林及其邻近村庄在当地森林管理制度安排、人为森林干扰模式以及随之而来的森林结构和组成变化方面的差异。将对目前在三种不同管理战略下的森林进行横向比较，根据地方参与程度的不同进行比较，并对国家森林政策修订前后每种森林进行纵向比较，以促进社区参与。这项研究的目的是(1)比较联合森林管理、社区森林管理和中央政府森林管理战略在森林状况、人类利用森林的空间格局和强度方面的变化，以及(2)确定与相对改善的森林状况相关的地方机构战略。森林状况将通过从覆盖地块和半球形照片样带测量的组成和结构属性来指示。森林转换和树冠空隙形成的当前和历史模式是伐木强度的结构性指标，将通过利用12年陆地卫星图像时间序列进行光谱混合分析和多时相变化检测来量化。对依赖森林的村民、当地森林管理委员会和林业部工作人员的半结构化访谈将确定森林准入规则、规则执行以及这些行为者之间的关系随着国家森林政策的变化而发生的变化。配对对照林将被用来解释管理变化以外的因素，这些因素可能会导致研究地点观察到的森林状况变化。研究还包括最小干扰的研究林，作为比较森林结构和组成的生态学参考，并确定在没有重大人类干扰的情况下树冠缺口形成的基准速率。研究结果将确定不同森林经营战略中依赖森林的人民在制度结构和当地参与方面的差异如何不同地影响研究森林中的人类干扰模式，从而影响森林状况和生态动态。研究结果可能有助于改进参与式环境下当地森林管理的功能设计，并增加目前关于不同的人为森林利用模式和强度如何影响热带森林结构的知识。这些信息对于建立一个更强有力的森林管理框架是必要的，以促进世界各地以人类为主的景观的社会和生态可持续性。

The fate of tropical forests in human-dominated landscapes relies on interdisciplinary research to better understand ecosystem processes and ecological functioning, while also determining how such forests may be managed to meet the needs of resource-dependent people. Participation by local forest users in tropical forest management continues to gain legal and institutional grounding, but the ecological effects of particular management actions and patterns of resource use under such policies remain understudied. This dissertation research combines tropical forest ecology and institutional analyses to examine differences in local institutional arrangements for forest management, anthropogenic forest disturbance patterns, and ensuing changes in forest structure and composition, among seven Tanzanian forests and forest-adjacent villages. Comparisons will be made horizontally among forests currently under three different management strategies varying by degree of local participation, and longitudinally in each forest before and after national forest policy was revised to facilitate community participation. The objectives of the study are (1) to compare changes in forest condition, spatial patterns and intensity of human forest use, between Joint Forest Management, Community Forest Management and central government forest management strategies and (2) to identify local institutional strategies correlated with comparatively improved forest conditions. Forest condition will be indicated by composition and structural attributes measured from overstory plots and hemispherical photo transects. Current and historical patterns of forest conversion and canopy gap formation, a structural indicator of logging intensity, will be quantified via spectral mixture analysis and multi-temporal change detection using a 12-yr Landsat imagery chronosequence. Semi-structured interviews with forest-dependent villagers, local forest management committees, and Forest Department staff will identify changes in forest access rules, rule enforcement and relationships between these actors in response to national forest policy change. Paired control forests will be used to account for factors other than management change that may drive observed changes in forest condition at the study sites. A minimally disturbed research forest is also included in the study to serve as an ecological reference for comparison of forest structure and composition, and to determine a benchmark rate of canopy gap formation in the absence of significant human disturbance.Research results will determine how differences in the institutional structure and local participation of forest-dependent peoples among different forest management strategies disparately affect human disturbance patterns in the study forests, hence forest condition and ecological dynamics. Study results may enable an improved design for functional local forest management in participatory settings, and increase the currently small body of knowledge about how different patterns and intensities of anthropogenic forest use affect tropical forest structure. Such information is necessary for building a stronger framework for forest management that promotes social and ecological sustainability in human-dominated landscapes worldwide.