A plea for inserting evidence‐based management into conservation practice

Worldwide, the area of land designated for conservation has doubled since the 1980s: it now exceeds 15 000 000 km2 and represents over 10% of the earth's land surface (Bertzky et al., 2012). At the same time, investment in conservation activities, both inside and outside protected areas, is increasing. Yet, broad-scale biodiversity indicators continue to decline internationally (e.g. Butchart et al., 2010), even in protected areas (Woinarski, Milne & Wanganeen, 2001; Gaston et al., 2006), raising the question of whether our protected areas are being effectively and/or efficiently managed. This apparent gap between conservation investment and ecological outcomes has led to a burgeoning literature on, and application of, protected area evaluation (e.g. see review in Leverington et al., 2010). Recognition of the importance of protected area evaluation is now mainstream: for example, in 2004 the Convention of Biological Diversity's adopted Programme of Work on Protected Areas included targets for countries to develop systems to assess management effectiveness (http://www.cbd.int/protected/pow/learnmore/intro/). There is now a plethora of protected area management assessment methods; for example, Leverington et al. (2010) describe over 50 different assessment methods applied to over 6000 sites. Most assessment methods can be nested within the IUCN-WCPA framework (Hockings et al., 2006) that sets out six elements of management effectiveness evaluation (context, planning, inputs, process, outputs, outcomes). Of these six elements, relatively few assessments address the last element – ecological outcome monitoring (Leverington et al., 2010). Where outcome monitoring and reporting does exist, it is almost always qualitative (Hockings, 2003; Gaston et al., 2006; Hockings et al., 2009). Although that can be useful, it is vulnerable to error (Cook, Hockings & Carter, 2009) and critically, not embedded in a question-based framework. Its use in decision-making and adaptive management is therefore limited (Pullin & Knight, 2005; Margoluis et al., 2009; Timko & Innes, 2009; Lindenmayer & Likens, 2010). In a review of the need for, and value of, evidence-based conservation management, Sutherland et al. (2004) reported that only 2% of the information used to support management decisions at a high-value conservation site in the UK was based on scientific evidence, while 77% was based on anecdote or experience. At many sites, no data were recorded at all. These global patterns are also apparent in Australia: the total area of land in the National Reserve System has almost doubled to 1.19 million km2 (15.5% land area) between 1997 and 2012 (www.environment.gov.au/topics/land/nrs/science-maps-and-data); government spending on the environment has also increased (competitive conservation grants more than doubled between 2001 and 2007; ANAO, 2007; Hajkowicz, 2009) and yet broad-scale biodiversity indicators continue to decline (Sattler et al., 2002; SoE-Committee, 2011; Szabo et al., 2012). Audits of government spending on the environment by the Australian National Audit Office consistently comment on the absence of outcome measurements and the absence of links between expenditure and outcomes (i.e. no accountability; summarized in Hajkowicz, 2009). While the National Reserve System Directions Statement says that protected area agencies need to assess and report on their management effectiveness (Hockings et al., 2009), the federal and state government conservation agencies continue to rely mainly on anecdote and experience, rather than quantitative data, including for ecological outcome monitoring and reporting (Pullin & Knight, 2005; Parks-Victoria, 2007; Hockings et al., 2009; Lindenmayer et al., 2012a; Varcoe, 2012). This means that there are few time series data, limited information on threatened taxa and ecosystems, and no feedback loops to management (Buckley et al., 2008; Lindenmayer & Gibbons, 2012). A detailed review of the conservation management of over 1000 protected areas in Australia found that almost one-third of park rangers were unable to assess the condition of biodiversity and threats on their parks. Only a quarter of rangers could make assessments about the ecological condition and the extent of threats on their reserve, and this was mostly based on experience rather than field data (Cook et al., 2009). In other words, while our investment in conservation is growing, and our efforts to measure and report on conservation ‘activity’ are increasing, our ability to report the ecological outcomes is not (Salafsky et al., 2002; Saterson et al., 2004; Regan et al., 2008). Very few protected area agencies or natural resource management programmes can quantify their conservation outcomes or modify their management reactively (Hockings et al., 2006). [For interesting exceptions, see Parks Canada's Ecological Integrity Monitoring (Timko & Innes, 2009; Woodley, 2010); the US National Parks Service's Vital Signs Monitoring (Fancy, Gross & Carter, 2009); the Kruger National Park's Monitoring Program (Du Toit, Rogers & Biggs, 2003)]. This outcome monitoring vacuum leads to ineffective targeting of expenditure, poor quality reporting, poor accountability, no connection to management and a continuance of biodiversity declines. In Australia, the non-government sector is arguably making the most concerted effort towards addressing the effective monitoring deficit on conservation-managed lands. Practical conservation management delivered by non-profit non-government organizations has a long history in some countries (especially the US) but is a relatively recent phenomenon in Australia. Perhaps the task of wedging out a slice of the philanthropic pie has encouraged these groups to have a stronger focus on ecological outcome monitoring, not only to inform management and improve conservation outcomes, but also to promote the efficient use of resources and to establish robust reporting and accountability – issues of critical importance to discriminating funders (Legge & Fleming, 2012; Radford et al., 2012). The Australian Wildlife Conservancy (AWC) is Australia's largest non-government conservation organization, managing 3 million hectares at 23 sanctuaries across Australia. It directs an unusually high proportion of its income to on-ground programmes: over 80% of its staff are based at sanctuaries, and about half of these field staff are scientists. This staffing model encourages meaningful collaborative planning, implementation and reporting by scientists and managers. The design of the conservation management programmes at each sanctuary is led by science staff; thus the programmes are science based, and management can be designed, where sensible, around key questions. This partnership between science and land managers is a critical ingredient for designing good monitoring programmes. AWC has also developed a system for measuring on ecological outcomes that is applied with site-specific detail but has a generic structure that allows aggregation up to higher regional and national levels for reporting and comparative analysis. The system is question based, including measurement of counterfactual outcomes where possible (i.e. what would have happened in the absence of management; Lindenmayer et al., 2012b), and both management and monitoring are iterated adaptively (Ferraro & Pattanayak, 2006). Many methods for measuring ecological health have been proposed (Landres, Verner & Thomas, 1988; Noss, 1990; Lindenmayer & Likens, 2010); so perhaps the most notable aspect to AWC's system is that it is actively being implemented and is underpinning management decisions (e.g. Legge et al., 2011a,b; Kennedy et al., 2012; Legge & Fleming, 2012), improving the efficiency and effectiveness of conservation outcomes, and providing a high level of accountability to AWC supporters. Given that numerous science-based frameworks for outcome monitoring exist, and given the need for, and benefits from, quantitative outcome-based monitoring, why don't more organizations carry it out? Outcome-based monitoring requires a relatively high degree of technical capacity and a commitment to the employment of science-trained staff (or collaboration with science organizations) and financial resources in the long term. It also requires structural reform to elevate the role of science in conservation management planning and reporting. Organizations need to have clear outcome-focused objectives to report against and be committed to transparency and accountability (Parrish, Braun & Unnasch, 2003; Ferraro & Pattanayak, 2006; Lindenmayer & Likens, 2010). These requirements are challenging, but they are surmountable and without facing the task, managers, policy makers and funders will continue to make decisions in the absence of information, and biodiversity will continue to decline. Funders – taxpayers, grantors, philanthropic individuals and foundations – can help drive the required changes by becoming more selective about where their conservation funding is directed by asking the question: what ecological outcomes will be generated by this support? Thanks to my colleagues at the Australian Wildlife Conservancy for much discussion on this issue over the years, and to David Lindenmayer for his review and comments. Thanks also to Trish Blann and Fay Lewis, whose support allowed me to write this paper.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.