Addressing new challenges in dynamic conservation planning

The available resources for biodiversity conservation are so chronically limited that conservation agency must choose what to protect immediately and what to leave for the future, understanding that some of what they leave remains at risk. This implies not only the choice of areas or species requiring protection, but also what to protect first, that is, designing a conservation schedule. Scheduling is the coordination of actions over time and space depending on the urgency for intervention, the spatial options for protecting features, the availability of funds, and other factors. Because most of these factors are poorly known, several simplifications are commonly made when setting priorities. These simplifications, or assumptions, concern both the socio-economic and the ecological aspects of biodiversity conservation. The uncritical use of assumptions reduces the effectiveness of conservation actions and prevents further progresses towards best practice in conservation. In this regard, I set four objectives for my thesis the achievement of which would may foster the progress in the field of systematic conservation planning: 1. Providing a framework to identify influential assumptions in dynamic conservation planning and testing their effects on the spatial pattern of conservation priorities and on the effectiveness of a proposed plan; 2. Explicitly incorporate biodiversity processes (habitat fragmentation) and variable site costs in dynamic conservation planning; 3. Identifying the planning contexts in which habitat vulnerability needs to be accounted for in dynamic conservation planning; 4. Assessing the utility of future global change scenarios, for conservation planning by applying them to predict the future conservation status of terrestrial mammals. The first, introductory chapter, provides the context for the thesis by retracing the history of systematic conservation planning with a focus on dynamic conservation planning. The second chapter (first research chapter) investigates the influence of assumptions on conservation-decision making with a focus on proactive and reactive approaches to priority setting. In this chapter I propose that the process of making assumptions explicit and testing them with scenarios and sensitivity analyses can help in reconciling apparently contrasting approaches to prioritization, and find an informed and more effective balance between proactive and reactive conservation. I find that assumptions regarding the irreplaceability of areas under threat, the viability of species in habitat fragments, and the relationship between costs and threats can significantly affect the spatial allocation of conservation efforts by shifting priorities in different ways with respect to the South American deforestation frontier. These assumptions should be critically evaluated before choosing an approach to priority setting, from the local to the global scale. In the third chapter I test the implications for conservation effectiveness (persistence of three forest-dwelling species) of two assumptions when scheduling conservation actions: 1) that populations are viable in small fragments; and 2) that costs are homogeneous in the study area. I do this by incorporating realistic estimates of acquisition cost of private land and simulating the species-specific effect of fragmentation on the distribution of three forest-dwelling species in coastal New South Wales (Australia). I tested two approaches to scheduling Maximizing short term gain of biodiversity in reserves (MazGain) and Minimizing short term loss of biodiversity in the planning region (MinLoss). This is a local scale application respectively of a proactive and reactive approach to conservation. I find that, although accounting for fragmentation and variable site cost reduces the cost-effectiveness of MinLoss, this approach remains the most effective with respect to MaxGain. In the fourth chapter I identify the conditions in which it is useful to incorporate information about habitat vulnerability (probability of development) in prioritization algorithms. I use computer simulations to investigate the role of various parameters in determining the best choice between three options: 1) using available vulnerability estimates when setting priorities; 2) investing in data collection to improve the accuracy of vulnerability estimates before selecting priority areas; or 3) discarding vulnerability altogether and selecting priority areas based only on biodiversity value and cost. I find that uncertainty in vulnerability estimate can alter the balance between proactive and reactive investments. In fact, when vulnerability estimates are on average wrong by at least 50%, discarding vulnerability and taking a proactive approach is best. Improving vulnerability estimates is the most effective choice when vulnerability variance is high and uncertainty is between 20% and 50%, otherwise, using existing vulnerability estimates with a reactive approach is best. In the fifth chapter I explore the implications for the conservation of terrestrial mammals in four global scenarios of human development from the Millennium Ecosystem Assessment (Millennium Ecosystem Assessment 2005b) and identify the future hotspots of global mammal conservation. I find that, across all scenarios, Mexico, most of South America and Sub-Saharan Africa will experience large losses of suitable habitat for many terrestrial mammals. Projected agricultural expansion both for food and energy crops will pose a severe threat to mammals in the coming decades. This thesis has contributed to the theoretical advancement of the field by shedding light on the relative performance of different conservation planning approaches and testing their sensitivity to approximations in data and models. My research has also provided an empirical framework to evaluate the role of assumptions in conservation planning. The first step is to frame a conservation planning problem using decision-theory to identify conservation objectives, actions, benefits, costs, and constraints. Then a series of questions related to the validity of the data and models available to solve the problems is used to identify the parameters and processes (ecological and socio-economic) for which more uncertainty exists. Different assumptions and different approaches to solve the conservation planning problem are then tested with different scenarios, each reflecting a different assumption about parameters and models. Combining the structured approach to problem-solving of decision-theory with the flexibility and foresight of scenario planning allows planners to make and learn from false assumptions without paying the consequences of real world mistakes. This, in turn, can provide insights on the approach to conservation that is most robust to future uncertainties. This thesis might also contribute to advancing conservation practice. The rules of thumb and recommendations on optimal priority setting for conservation practitioners stemming from my research can have important practical consequences for conservation decisions at local to global scales