UC Berkeley

This dissertation is placed at the center of debates over adding human dimension to reserve selection in Madagascar. For many years, we knew that forested lands are cleared at an alarming rate. And for many years, decision makers in conservation planning were aware of the necessity to address human needs have at the same time saving habitat for species if we were to be successful in conservation program. However, No effort has been undertaken in Madagascar to include the human dimension. Instead conservation program continues to conflict with local communities dependant on natural resources for their subsistence. I looked at the National scale to address the issue of selecting reserve network while considering different socio-economic costs. The results show that inclusion of cost in conservation planning did not drastically differ to current design. At regional scale, I looked at the design of Community based natural resources management established around Makira Protected Area. The result suggests that current involvement of community in conservation activities is not likely to stop deforestation.

In chapter II, I modeled rice field expansion in Madagascar, analyzed the different parameters that influence land use suitability for rice, and predicted the location of changes under different future scenarios. The specific objectives are to map existing rice fields and produce a model of suitable land for rice cultivation under current climate and conditions, understand the parameters influencing the expansion or constraints on rice cultivation, and predict the spatial location of future rice cultivation under assumptions of increasing population and future climate change. Analyzing and interpreting the change in land suitability based on circumstances that drive the changes provide essential information to decision makers and enable them to respond adequately to development and conservation issues. I found that land suitability value decreases with increasing slope, the model is improved if I use geology, a proxy for soil variables, to stratify the data, a significant portion of currently cultivated rice fields will experience drier and warmer conditions in the future, and large shifts to the northern and western side were observed under future climate scenarios and as much as 36% of current lands may become unsuitable.

In chapter III, I re-examines the effectiveness of the reserve network proposed by Kremen et al. (2008), by looking at the possible conflict in the existing protected areas given the integration of various costs into the process of network reserve selection. After looking at the possible changes needed in the design of current conservation areas in Madagascar when introducing cost to conservation planning, I also investigate what changes would be needed to take into account the effect of future suitable agricultural land (under future climate change scenarios) in planning the reserve network, and provide recommendations for the expansion and priority setting of new PA priority sites. My results show that at the national level, inclusion of costs in systematic conservation planning did not drastically change the design of the current reserve network. The effect of including costs may be more pronounced at the regional scale. My results were inconclusive with regards to taking into account shifting costs resulting from future climate change. I conclude by giving recommendations regarding new reserve areas regarding the government priority for setting up additional conservation areas.

In chapterIV, I explore the current status of the Makira Protected Area, and analyze the relationship between land uses to a community management strategy. I first examined how the forest management contracts were set up and administered, and then assessed the efficacy of these contracts with respect to institutional effectiveness (Ostrom, 1990) and reduction of deforestation, the key driver of biodiversity endangerment in Madagascar (Harper, Steininger, Tucker, Juhn, & Hawkins, 2008; Kremen et al., 2008). In this study, I first present a qualitative narrative of the processes of establishing management transfer. Second, I evaluate the forest management contracts in Makira Protected Area relative to the 8 design principles of Ostrom (1990) for management of common property resources. Third, I present data from household surveys showing the prevalence of deforestation in forest management contract areas.