Incorporating socio-economic considerations into systematic conservation planning

A primary focus of conservation efforts has been the establishment of protected areas, both by governments and conservation non-government organizations, to meet conservation commitments. While most research to date has neglected considerations other than biodiversity, there is a growing recognition that the socio-economic context in which conservation actions are ultimately applied influences the success of implementation. This is reflected by increasing prominence in the literature of considerations of conservation costs. However, despite the increased focus on cost-efficiency in conservation, there is still a lack of critical guidance on techniques for estimating costs and the implications for conservation outcomes of using untested surrogates for actual costs. The inclusion of appropriate cost measures in conservation planning is not only of academic interest but is relevant to ensuring that conservation plans accurately reflect the social context and provide relevant policy advice to government and conservation practitioners. The current knowledge gaps include appropriate methodologies for cost estimation, ways of costing multiple conservation actions, and consideration of costs that vary between groups of stakeholders. These gaps must be filled to ensure that conservation plans translate into effective on-ground actions that are adequately funded and supported by local communities.\ud \ud The primary objective of my thesis is to identify the appropriate costs of conservation given particular socio-economic goals and the likely associated conservation actions, such as acquisition of land for national parks or engaging stakeholders in private land conservation agreements. My thesis research explores these issues for three different types of conservation planning exercises and then develops and applies rigorous economic methods for estimating the costs of conservation. The three kinds of conservation planning exercises are:\ud \ud 1. Acquiring land in a well defined land market for extension of an existing protected area system, when acquisition costs are a key consideration;\ud \ud 2. Protecting areas in the absence of a well defined market, when opportunity costs are a better measure of the costs of conservation;\ud \ud 3. Conserving land on private property through changes in land management that involve incremental changes in management costs.\ud \ud The second objective of my thesis is to incorporate costs to multiple stakeholders and account for multiple conservation strategies and stakeholder preferences. Most conservation plans will involve engaging with a range of stakeholders and implementing different conservation strategies, ranging from acquisition to stewardship programs with private landholders. To accurately reflect the true costs associated with conservation, the variable costs to different stakeholders associated with different strategies must be accounted for explicitly.\ud \ud My third objective is to incorporate uncertainties in my cost estimates into robust budgeting and decision making. Studies to date have not addressed how uncertainties associated with cost assumptions can drive budget estimates and shape conservation outcomes. Uncertainties arise from many factors which remain unexplored in the literature. In addition, given an understanding of the uncertainties associated with cost estimates, there is no guidance in the literature on how to make robust decisions, that is, decisions that still give an acceptable outcome despite uncertainty in parameters. Robust decisions require better accounting for uncertainty in the decision making process.\ud \ud In Chapters 2 and 3, I examine the most common application of conservation planning in which areas are selected for full protection in a fully developed land market and therefore the cost of conservation is the acquisition cost of land at market value. I use the expansion of the protected area system in Queensland as a case study for this planning context. The government has committed to adding ~12 million ha to the reserve system, ~4 million ha of which will be in national parks, bringing the total area to 20 million ha by 2020. But there have been no details released on explicit conservation objectives, financial allocations, or formal budgeting. I estimate the costs associated with acquisition of land for national parks by first estimating sales value for all properties with a geographically weighted regression model. I then use this model, and varying values for several key factors, to incorporate uncertainties in my cost estimates into robust budgeting and decision making. I use two methods to explore uncertainty: sensitivity analysis and information-gap modelling.\ud \ud In Chapter 2, I explore the challenge of budgeting a priori for protected area expansion in the face of uncertainty, specifically considering the future expansion of protected areas in Queensland, Australia. I use Marxan to estimate the costs of potential reserve designs with data on actual land value, market value (including land and improvements), transaction costs, and tenure status. With scenarios, I explore three sources of budget variability: size of biodiversity objectives; subdivision of properties; and legal acquisition routes varying with tenure. Depending on the assumptions made, the estimated budget varies from $214 million to $2.9 billion. I demonstrate how a rigorous analysis can inform discussions about the expansion of systems of protected areas, including the identification of factors that drive budget variability over time.\ud \ud In Chapter 3, I again use the Queensland protected area expansion as a case study and apply an info-gap model as a second method for addressing uncertainty. In contrast to Chapter 2, in which I use Marxan to select comprehensive sets of properties, this chapter applies a common prioritization strategy by calculating benefit-cost ratios and targeting those properties with the highest ratio value. In recent applications of BCRs, surrogate measures of the full benefits and costs of conservation projects, such as measuring area of vegetation types or using land value as a surrogate for acquisition cost, have been used. In this chapter I demonstrate the potential limitations of applying BCRs with surrogate measures. I compare the performance of BCRs based on two surrogate measures of both benefits and costs with the performance of prioritization based only on benefits. In addition, I explore the robustness of my BCR allocation, defined as the proportion of properties misallocated for investment due to uncertainties about the benefits and costs of projects, using an information-gap model (info-gap). I find that BCR allocations are more efficient than benefit allocations, and protect more regional ecosystems at a minimum level of 10%, and are generally more robust than benefit-only allocations.\ud \ud The second conservation planning context that I examine is the case of selecting areas for full protection in regions where markets are not fully developed and therefore opportunity costs are a better measure of the full costs of conservation. In Chapters 4 and 5, I consider opportunity costs in marine and terrestrial contexts. In Chapter 4, I consider the planning region of Kubulau district, Vanua Levu, Fiji Islands, where the community is considering extending or reconfiguring the marine protected area (MPA) network. I present a novel method for calculating the opportunity costs to fishers from their displacement by MPAs. I model opportunity costs as a function of food fish abundance and probability of catch, based on gear type and market value of species. I also develop a profit model to investigate the effect of restricted access to transport on costs to fishers. I then include the opportunity cost model in Marxan to examine potential MPA configurations for Kubulau District, Fiji Islands, to minimize costs to local communities. I identify optimum areas for protection in Kubulau with: (a) the current MPA network locked in place; and (b) a clean-slate approach.\ud \ud In Chapter 5, I use the Mbaracayu Forest Biosphere Reserve in Paraguay as a case study. The biosphere reserve has been identified as a high priority area for conservation interventions compared to other similar regions in Paraguay. I chose the biosphere reserve as my study region because of the availability of high-resolution data on costs and vegetation types. In this chapter I partition land costs into three distinct opportunity costs to smallholder agriculture, soybean agriculture, and ranching. I demonstrate that opportunity costs to single stakeholder groups can be inaccurate measures of true opportunity costs and can inadvertently and disproportionately displace conservation costs to groups of stakeholders who have not been considered. Additionally, I examine how spatial correlations between costs as well as size of conservation objectives affect the performance of opportunity costs to single stakeholder groups as surrogate measures of true opportunity costs. I conclude that planning with opportunity costs to single stakeholder groups can result in cost burdens to other groups that could undermine the long-term success of conservation. Thus, an understanding of the spatial distributions of opportunity costs that are disaggregated to groups of stakeholders is necessary to make informed decisions about priority conservation areas.\ud \ud In Chapters 6 and 7, I consider the third conservation planning context: the case of selecting areas, not for full protection, but for participation in incentive programs that might only partially reduce production or require only marginal changes to management regimes. In Chapter 6, I estimate the potential costs of a stewardship program in the Daly River catchment, Northern Territory, which would underwrite the cost difference between routine land management and the additional requirements of conservation management. Based on survey responses from landholders, I first assess the current costs of land management in the catchment and use regression to identify key drivers of spatial variation in both routine land management costs and total land management costs (defined as land management costs to meet objectives for both routine property management and conservation). I then estimate the additional costs of conservation management over and above routine land management at an average of $1.99 per ha. Much of the cost-effectiveness of stewardship payments would come from their ability to leverage the costs of routine land management already met by landholders.\ud \ud In Chapter 7, I apply the cost estimates developed in Chapter 6 but also look at additional design criteria for a stewardship program in the Daly Catchment including the probability of landholders participating in the program. I use choice modelling to estimate the probability of participation in two legal arrangements - conservation covenants and management agreements - based on payment level and proportion of property required for management under the agreement. I then spatially predict landholders' probability of participating at the resolution of individual properties and incorporate these spatial predictions into conservation planning software to examine the potential for the stewardship program to meet conservation objectives. I find that there is a tension between planning for a cost-effective program and planning for a program that targets properties with the highest probability of participation.\ud \ud Overall, my thesis highlights the importance of stating explicit socio-economic objectives. Explicit objectives are at the heart of systematic conservation planning; however, objectives associated with social and economic values are often neglected. The recent rush to include socio-economic costs in conservation planning has brought to the forefront the importance of considering the social context for conservation, and the lack of established methods for accounting effectively for costs underlies much of the current debate in the literature. Of the studies that have included cost metrics, relatively few link them to explicit objectives and most have not been based on rigorous economic methods. My thesis provides techniques for rigorously estimating costs in three broad types of situations that have to be addressed in conservation planning, while also accounting for uncertainties in cost estimates