This dissertation comprises four chapters. The unifying theme is the evaluation of environmental programs. Specifically, each chapter examines some facet of the impacts of protected areas. The first chapter examines the heterogeneous environmental and economic impacts of protected areas in Costa Rica. Previous studies suggest that Costa Rica's protected area system induced both reduced deforestation and alleviated poverty. We demonstrate that these environmental and social impacts were spatially heterogeneous. Importantly, the characteristics associated with the most avoided deforestation are the characteristics associated with the least poverty alleviation. In other words, the same characteristics that have limited the conservation effectiveness of protected areas may have improved the social welfare impacts of these areas. These results suggest that `win-win' efforts to protect ecosystems and alleviate poverty may be possible when policymakers are satisfied with low levels of each outcome, but tradeoffs exist when more of either outcome is desired. The second chapter explores in more detail the heterogeneous impacts of protected areas in Costa Rica and Thailand. In particular we investigate the potential for protected areas to act as a mechanism for poverty traps and use semiparametric models to identify the spatial congruence of environmental and economic outcomes. We find no evidence that protected areas trap historically poorer areas in poverty. In fact, we find that poorer areas at baseline appear to have the greatest levels of poverty reduction as a result of protection. However, we do find that the spatial characteristics associated with the most poverty alleviation are not necessarily the characteristics associated with the most avoided deforestation. We demonstrate how an understanding of these spatially heterogeneous responses to protection can be used to generate suitability maps that identify locations in which both environmental and poverty alleviation goals are most likely to be achieved. In the third chapter we address the mechanisms through which protected areas affect economic outcomes. Using recently developed quasi-experimental methods and rich biophysical and demographic data, we quantify the causal post-treatment mechanism impacts of tourism, infrastructure development and ecosystem services on poverty, due to the establishment of protected areas in Costa Rica prior to 1980. We find that nearly 50% of the poverty reduction estimated in a previous study can be attributed to tourism. In addition, although the mechanism estimates for the infrastructure and ecosystem services proxies are negligible, we argue that the results provide evidence that enhanced ecosystem services from the establishment of protected areas has likely helped to reduce poverty. The results provide additional information to policy makers that wish to enhance the future establishment of protected areas with complementary policy. The final chapter studies the economic impacts of protected areas in Bolivia. We find that municipalities with at least 10% of their area occupied by a protected area between 1992 and 2000 exhibited differentially greater levels of poverty reduction between 1992 and 2001 compared to similar municipalities unaffected by protected areas. We find that the results are robust to a number of econometric specifications, spillover analyses and a placebo study. Although the overarching results that Bolivia's protected areas were associated with poverty reduction are similar to previous studies , the underlying results are subtly, but significantly, different. In previous studies it was found that controlling for key observable covariates lead to fundamentally antithetical results compared to naive estimates. Conversely, these results indicate that naive estimates lead to an over-estimation of the poverty reducing impacts of protected areas. The results expose the heterogeneity of protected area impacts across countries and, therefore, underscore the importance of country-level impact evaluations in order to build the global knowledge base regarding the socioeconomic impacts of protected areas.

Remote ecosystems in Canadian Rocky Mountain parks and protected areas are being pressured by indirect impacts of human activities across the landscape. Ecological impacts can result from a variety of stressors such as climate change, fire suppression and prescription, visitor use, invasive species, and surrounding land-uses. With intensified challenges relating to logistics and moral issues inherent in remote ecosystem management (Higgs and Hobbs, 2010; Higgs and Roush, 2011), managers of these landscapes continue to struggle with questions of "what do we do here?" For ecological restoration and conservation management, historical landscape changes (predominantly following years of fire suppression and rapid climate change) are complicating decisions and our understanding of ecological processes. While intervention may become increasingly necessary for maintaining biodiversity and ecosystem services under conditions of rapid change (Hobbs et al., 2011), direct action may not be the most appropriate tactic, especially when lacking adequate information and foresight (Harris et al., 2006; Higgs and Roush, 2011). This study investigates observable historical remote landscape change in two protected areas in the Canadian Rockies, and examines the resulting implications for management and restoration. Present conditions in Willmore Wilderness Park and Mount Robson Provincial Park were evaluated in a comparative case-study approach. The influence of factors such as jurisdiction, climate, socioeconomics and settlement history were seen to strongly shape how environmental changes impacted management efforts and decisions. Long-term landscape changes were observed through repeat photography with the Mountain Legacy Project. Through focus groups using photo-elicitation with park managers, repeat photo pairs guided discussion. I pursued this research question: "how do long-term landscape changes influence conservation and restoration objectives in remote mountain ecosystems?" This also incorporated subsidiary questions: "what are the inherent challenges in managing (and restoring) remote ecosystems?” and “how could managers of remote ecosystems best approach these issues in the face of rapid ecological change?” Significant landscape changes are observed in both parks and include glacier retreat, forest stand aging, valley infill (encroachment) and upward movement of the treeline ecotone. While ecological changes are seen as significant and as threatening to various park values and public safety, efforts to better understand these changes or address them are limited. The majority of remote ecosystem management efforts in both parks are indirect (passive), with the exception of fire management. This is largely due to capacity and resource constraints, and agency recreation mandates and visitor needs monopolizing manager focus. Suggested restoration efforts would assist climate adaptation and reduce indirect impacts without placing notable pressure on remote ecosystems. The use of repeat photography for monitoring of ecological change is a strong possibility for parks management, particularly if public engagement through citizen science was implemented to minimize dependence on management resources. / Graduate / 0368 / 0768 / jlfalk@uvic.ca

Earlier studies have found that endangered species recovery is only weakly associated with the tools enabled by the U.S. Endangered Species Act (ESA). With habitat loss often cited as a leading cause of species declines, we tested whether the recovery of ESA-listed species is instead associated with the protection of critical habitat (CH) by protected areas. We tested the relationship for 299 species using recovery indices derived from the biennial status reports to Congress (1990-2010), as well as NatureServe and IUCN population status data. We found no overall relationship between recovery and the extent to which CH is protected. However, restricting the analysis to recovering species, listed species with larger areas of protected (R2 = 0.158) or strictly protected (R2 = 0.194) CH fared better than species with less protected or strictly protected CH areas. Declining species (199 of 273 species studied) fared no better with more protected habitat. We conclude that the abatement of habitat loss alone does not necessarily facilitate recoveries for the majority of ESA-listed species. We also note that the weak relationships we observed in this study may be reflective of poor recovery status estimates.

In Canada, little is known about the governance of private protected areas (PPAs) and their contribution to the public interest. This dissertation evaluates the extent to which PPAs in Canada adhere to generally advanced principles of good governance - Direction, Legitimacy, Fairness, Performance and Accountability - and it assesses their contribution to the public interest. Findings from descriptive analysis of six case studies of Individual, Corporate, Non-Government and First Nation authorities indicate that principles are adhered to with relative vigor and visibility and that the public interest is generally advanced. The governance pattern is not strongly differentiated and it is generally uniform in strengths and limitations. The cases studied were exceptionally strong in fulfilling the principles of Direction and Legitimacy. Management documents not only employed long-term, ecosystem-based approaches, the private areas were shaped by the collaboration and support of a diverse array of interested parties. Public participatory processes remain relatively ad hoc in nature. Issues management occurred in each of the private areas studied. There was no evidence, however, of intractable disputes, suggesting that the principle of Fairness was fulfilled. The principle of Performance was determined to be the weakest across the sector. Generally, appropriate standards are not well defined, monitoring is highly fragmented and there is an absence of reporting on performance outcomes, which creates uncertainty about the standing of protection and conservation values. Accountability is multi-dimensional: accountability to different stakeholders for various goals and activities through different mechanisms is salient. Independent oversight is employed infrequently. Findings also reveal subscription to a broad spectrum of protected area values that are achieved through moral. fiduciary and legal means. The desire to pass along a natural legacy to future generations is a leading feature. Formally expressed values that are widely shared include protection of ecosystems and biological diversity. Public access is generally constrained. Protection in perpetuity is most often guaranteed through instruments that impose obligations upon future land owners. Support for strengthened governance processes and structures would further the complementary role and contribution that this sector can make to the important mission that protected areas seek to accomplish.

Protected areas, such as national parks, can provide nearby communities with a wide range of environmental, social, and economic benefits, such as ecosystem services and tourism development. However, protected areas can also subject communities to a number of costs, such as displacement and an increase in negative human-wildlife interactions. This study investigates how the communities of Tofino and Ucluelet, British Columbia, Canada perceive they have been impacted by Pacific Rim National Park Reserve (PRNPR). Specific areas of focus include respondents’ attitude and level of support for PRNPR and their perceptions of concerns and benefits related to the park. Data for this study was collected using a focus group and questionnaire administered using the “drop-off” method. The results of this study suggest that residents of Tofino and Ucluelet generally have a positive attitude towards the park and perceive PRNPR subjects their communities to a number of concerns and benefits. It was found that attitudes were linked to perceived concerns and benefits, perceived changes in the community, and level of involvement with PRNPR. With regards to concerns and benefits, it was found that respondents were most concerned with financial costs related to PRNPR and most valued benefits related to conservation. / Graduate

This research examined the potential for protected areas with camas (including tall camas, Camassia leichtlinii (Baker) Wats., and common camas, C. quamash (Pursh) Greene; Liliaceae (Agavaceae)) habitat to support the renewal of Central Coast Salish camas traditions while at the same time maintaining and even expanding their ecological restoration and conservation goals. For many generations Central Coast Salish Peoples of northwestern North America have cultivated camas plants and harvested, processed, and consumed their edible bulbs in large quantities. Today, after camas use has almost completely disappeared from their lives, some Indigenous peoples are working to restore camas habitats and cultivation practices on southern Vancouver Island and neighbouring areas. Tall camas and common camas can still be found growing in many Garry oak ecosystems, which, due to the decreased range and the large proportion of rare species found within them, are frequently the focus of ecological restoration and conservation efforts. I interviewed people from the resource management and First Nations communities to gain an understanding of the current interests, opportunities, challenges, and potential approaches for incorporating traditionally based camas harvesting and management into protected areas today. Protected areas were identified as important areas for teaching traditional plant cultivation techniques to younger generations, and as bulb and seed banks for ethnoecological restoration projects. Overall, managers of protected areas and First Nations participants were receptive to collaborating on management of camas populations. Anticipated or existing challenges or concerns included ecological uncertainties of harvesting disturbance, ensuring safety, finding funding, and gaining trust. I conducted one season of experimental camas harvesting in a Garry oak savannah near Duncan, BC within an ecological preserve and monitored the effects of this harvesting on the extant camas populations, on surrounding plant communities, and on soil porosity. Harvesting of, primarily tall, camas bulbs, at both low and medium intensity, did not affect the weight or abundance of camas bulbs or the quantity, stem height or flowering/fruiting potential of the camas populations in the following year. Harvesting significantly reduced the abundance of Kentucky bluegrass (Poa pratensis) and common snowberry (Symphoricarpos albus, but significantly increased the abundance of Scotch broom (Cytisus scoparius) common cleavers (Galium aparine), hairy cat’s ear (Hypochaeris radicata), and nipplewort (Lapsana communis). Harvesting significantly reduced the level of soil compaction. Using the insights gained from the interviews and experimental harvesting I have proposed an “Ethnoecological Restoration Support Model”. This model explains how protected areas can support cultural restoration both within and outside of protected areas while maintaining and even expanding upon current conservation and restoration goals. / Graduate / 0329 / 0740 / 0471 / kproctor@uvic.ca

Remote ecosystems in Canadian Rocky Mountain parks and protected areas are being pressured by indirect impacts of human activities across the landscape. Ecological impacts can result from a variety of stressors such as climate change, fire suppression and prescription, visitor use, invasive species, and surrounding land-uses. With intensified challenges relating to logistics and moral issues inherent in remote ecosystem management (Higgs and Hobbs, 2010; Higgs and Roush, 2011), managers of these landscapes continue to struggle with questions of "what do we do here?" For ecological restoration and conservation management, historical landscape changes (predominantly following years of fire suppression and rapid climate change) are complicating decisions and our understanding of ecological processes. While intervention may become increasingly necessary for maintaining biodiversity and ecosystem services under conditions of rapid change (Hobbs et al., 2011), direct action may not be the most appropriate tactic, especially when lacking adequate information and foresight (Harris et al., 2006; Higgs and Roush, 2011). This study investigates observable historical remote landscape change in two protected areas in the Canadian Rockies, and examines the resulting implications for management and restoration. Present conditions in Willmore Wilderness Park and Mount Robson Provincial Park were evaluated in a comparative case-study approach. The influence of factors such as jurisdiction, climate, socioeconomics and settlement history were seen to strongly shape how environmental changes impacted management efforts and decisions. Long-term landscape changes were observed through repeat photography with the Mountain Legacy Project. Through focus groups using photo-elicitation with park managers, repeat photo pairs guided discussion. I pursued this research question: "how do long-term landscape changes influence conservation and restoration objectives in remote mountain ecosystems?" This also incorporated subsidiary questions: "what are the inherent challenges in managing (and restoring) remote ecosystems?” and “how could managers of remote ecosystems best approach these issues in the face of rapid ecological change?” Significant landscape changes are observed in both parks and include glacier retreat, forest stand aging, valley infill (encroachment) and upward movement of the treeline ecotone. While ecological changes are seen as significant and as threatening to various park values and public safety, efforts to better understand these changes or address them are limited. The majority of remote ecosystem management efforts in both parks are indirect (passive), with the exception of fire management. This is largely due to capacity and resource constraints, and agency recreation mandates and visitor needs monopolizing manager focus. Suggested restoration efforts would assist climate adaptation and reduce indirect impacts without placing notable pressure on remote ecosystems. The use of repeat photography for monitoring of ecological change is a strong possibility for parks management, particularly if public engagement through citizen science was implemented to minimize dependence on management resources. / Graduate / 0368 / 0768 / jlfalk@uvic.ca

An analysis of the impact of incomplete species distribution information on our capacity to identify minimum size protected area networks was performed. Using bird distribution data for southern Quebec it was found that efficient protected area networks could be established using incomplete distribution information, as little as 30%. The results also revealed that reserve networks encompassing the majority of the biodiversity could be established using as little as 10% of the information. The key to producing efficient networks is the ability of the selection algorithm to identify complementary sets of sites in terms of species composition. The results reveal that the algorithm can best accomplish this when the data set used is produced using a geographically intensive site-based survey approach. The results were consistent over three data sets, the original and two modified, indicating that they are transferable to geographic areas and species groups with differing species distribution characteristics.

There are concerns that many national parks worldwide are ineffective at conserving biological diversity and ecosystem processes, are socially unjust in their relations with Indigenous communities, or both. This dissertation asks: can national parks protect ecological integrity and concurrently address social equity issues? It presents empirical results of a systematic evaluation of six case study national parks in Canada and South Africa. Purposive sampling was used to select the six case study national parks. Data sources included State of the Park Reports; park ecological monitoring data; archival data; and semi-structured interviews with park biologists, managers, and Indigenous members of park co-management boards. Status and trend assessments and effectiveness evaluations of park ecological monitoring data were used to evaluate how effectively the parks addressed three ecological integrity criteria. Results show that all six parks effectively addressed the priority indicators for which they had monitoring data. However, the effectiveness ratings of each park decreased when all indicators, including those identified as priorities but lacking monitoring data, were analysed. This indicates that the parks had generally identified more priority indicators than they were actually able to address (for reasons including lack of budget or trained staff, managerial challenges). Thematic coding of semi-structured interview and archival data, and the assignation of numerical ratings to these data, were used to evaluate how effectively the parks addressed three equity criteria. Results show that all but one of the case study parks were equitable, parks with more comprehensive co-management and support from neighbouring Indigenous groups were more equitable than parks with lower levels of co-management, the parks with settled land claims were not necessarily more equitable overall, and a few parks were found to be co-managed in name only. The overall results of this evaluation demonstrate that parks effective at protecting ecological integrity can also successfully address social equity, but that further efforts to integrate these two realms are both possible and necessary. A logical starting point would be to build upon those existing integrative processes already institutionalised in many parks and protected areas: the co-management and integrated conservation and development efforts.

This thesis provides the first direct comparison between – and integration of – community-based and science-based approaches to the establishment of marine protected areas (MPAs). MPAs are one potentially effective conservation tool, but are being established very slowly. My research shows that community involvement in placing MPAs can help meet many ecological goals, although biophysical data improve the conservation value of sitings. To assess the need for MPAs in British Columbia (BC), Canada, I mapped stressors resulting from human activities. This produced a powerful rationale for MPAs: very little of the ocean, and almost none of the continental shelf of BC, lies beyond the reach of human stressors. My work helps reconcile differing perspectives about the efficacy of community-based vs. science-based MPA selection. I explored and analyzed these approaches, separately and together, in two areas in BC. First, I generated a community-based plan for MPA placement through partnerships with two First Nations (indigenous peoples) in BC. They offered strong support for spatial protection measures, and individuals nominated overlapping areas. Second, I applied a decision support tool (Marxan) to determine MPA placement under scientific precepts. Conservation planning usually lacks detailed ecological information but the Marxan approach was robust to some missing data; in such cases, it was best to use available abiotic and biotic data to ensure that both habitats and species were represented. Third, I integrated community-based and science-based approaches, to find that they verified and complemented each other. Indeed, an integration of the two was preferred by participants and also achieved all conservation objectives. Finally, I took a novel and pragmatic approach to ocean zoning. I used spatial data for thirteen commercial fisheries on Canada’s west coast to select areas where fishing should be permitted, rather than prohibiting fishing under a MPA paradigm. The results revealed that small reductions in fisheries yields, if judiciously selected, could allow creation of large unfished areas that embraced diverse biophysical regions and habitat types. Such a pragmatic approach could achieve remarkable conservation gains.