LETTER Climate and Fire Scenario Uncertainty Dominate the Evaluation of Options for Conserving the Great Desert Skink Natasha C. R. Cadenhead1, Michael R. Kearney2, Danae Moore3,4,SteveMcAlpin5, & Brendan A. Wintle1 1 Quantitative and Applied Ecology Lab, School of Biosciences, The University of Melbourne, Parkville, VIC 3010, Australia 2 School of Biosciences, The University of Melbourne, Parkville, VIC 3010, Australia 3 Department of Biological Sciences, Macquarie University, North Ryde, NSW 2113, Australia 4 Australian Wildlife Conservancy, Newhaven via Alice Springs, NT 0872, Australia 5 School of Environmental and Rural Science, The University of New England, Armidale, NSW 2351, Australia Keywords Abstract Arid, Australia; decision-making; dynamics, fire; fire management; landscape ecology; lizard; Fire regimes are predicted to change under climate change, with associated population modelling; species distribution impacts on species and ecosystems. However, the magnitude and direction of modelling. regime changes are uncertain, as will be species’ responses. For many species, how they respond will determine their medium-long-term viability. We prop- Correspondence agate fire regime and species’ response uncertainties through a 50-year via- Natasha Cadenhead, Quantitative and Applied bility analysis of the great desert skink, Liopholis kintorei, in central Australia, Ecology Lab, School of Biosciences, The University of Melbourne, Melbourne, VIC 3010, characterizing fire regime change under three scenarios. Species’ response un- Australia. certainty was characterized with three competing models based on fire and Tel: +61 3 9035 6164; fax: +61 3 9348 1620. habitat variables, fitted to 11 years of occupancy data. We evaluate fire man- E-mail: [email protected] agement options for conserving the species, based on their robustness to un- certainty about fire and species’ response. Efforts to minimize the frequency Received and size of fires provides the most consistent improvements to species’ persis- 25 March 2015 tence. We show that disentangling important from unimportant uncertainties Accepted 7 August 2015 enables conservation managers to make more efficient, defensible decisions. Editor Dr. Matt Hayward [Correction added on 1 April 2016, after first publication: addition of acknowledgement of support from APSF.] doi: 10.1111/conl.12202 Introduction 2009; Bird et al. 2013). Conservation managers can therefore use fire and fire suppression as management The management of threatened species should be based tools for achieving conservation outcomes (Edwards et al. on an understanding of habitat requirements, and the dy- 2008). However, wildfire is irreducibly stochastic in its namics of that habitat through space and time. However, ignition and propagation, and our understanding of fire there are typically large uncertainties in such knowledge, dynamics is often uncertain. Severe uncertainty about especially in the context of a changing climate that fire is compounded by projected changes in temperature can dramatically influence management effectiveness and rainfall, and by an incomplete understanding of (Hulme 2005; Kujala et al. 2013). Fire is a fundamental species’ fire ecology at both local and regional scales. driver of habitat change and uncertainty across many Conservation managers should aim to incorporate continents (Bond & Keeley 2005), and endemic species such uncertainties into the development of conservation have evolved to tolerate or depend on certain broad pat- strategies for two reasons. First, an explicit characteriza- terns of fire that occur in their landscapes (Van Wilgen tion of uncertainty allows managers to anticipate a range Conservation Letters, May/June 2016, 9(3), 181–190 Copyright and Photocopying: C 2015 The Authors. Conservation Letters published by Wiley Periodicals, Inc. 181 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Resolving uncertainty in fire management N.C.R. Cadenhead et al. of possible outcomes, avoiding nasty surprises. Second, a Materials and methods quantitative description of uncertainty allows managers to choose actions that will maximize beneficial outcomes. Case study However, considering all uncertainties is overwhelm- L. kintorei is a large, nocturnal skink found in restricted ing and inefficient because some are more important than pockets of arid and semiarid habitat across the central and others. In the context of management decisions, an un- western deserts of Australia (McAlpin 2001). L. kintorei certainty is only important if its resolution would lead to live in family groups, unusually for reptiles, and build ex- a change in management. Ignoring unimportant uncer- tensive burrow systems (McAlpin et al. 2011) which pro- tainties can be liberating for decision-makers, increasing vide some protection from fire and feral predators (Moore clarity by reducing apparent complexity. Ignoring impor- et al. 2015a). tant uncertainties, however, generally leads to decisions The Australian Wildlife Conservancy’s (AWC) that are fragile (Ben-Haim 2006), and it is not a priori Newhaven Wildlife Sanctuary in central Australia is obvious which uncertainties are important. Explicit un- an important stronghold for L. kintorei. Securing L. kin- certainty analyses are not common in the conservation torei in the sanctuary is a key AWC objective. Newhaven decision-making literature, and even fewer deal with re- covers 262,000 ha of the Great Sandy Desert Bioregion, alistic levels of uncertainty for conservation management 300 km northwest of Alice Springs. Widespread and problems. intense wildfire is a common and integral feature of Here we consider how to quantify and incorporate un- spinifex (Triodia spp.) dominated landscapes (Paltridge certainties into fire management decisions for the great 2005) during the extremes of summer, particularly desert skink (Liopholis kintorei; Stirling & Zeitz 1893), a following above average rainfall. AWC have a number of listed threatened species (IUCN 1996) that occurs in arid fire-related management objectives, focused on conserv- central Australia (McAlpin 2001). There is substantial ing biodiversity through a variety of habitats at various uncertainty about the drivers of apparent population stages of postfire recovery. declines, and about how the species will respond to changing climates and fire regimes. L. kintorei is expected to experience accelerated declines as increasing fire Habitat models frequency expands recently burnt habitat that is thought to be unsuitable (Moore et al. 2015a). L. kintorei occurrence records consisted of burrow pres- Interactions between fire and feral predators may also ence/absence data collected over 11 years (2003–2013) play a role, as for other small reptiles and mammals in the through a mixture of survey techniques described in the region (McGregor et al. 2014; Moore et al. 2015b). This Supporting Information (S1). Environmental variables uncertainty is compounded by the dynamics of future of vegetation, geology, and fire history on Newhaven fire regimes in central Australia (Watterson et al. 2015), were obtained from AWC along with radiometric data which are driven by poorly-understood relationships be- on soil types and geological formations (Minty et al. tween temperature, rainfall, plant growth, fuel load, and 2009). Annual fire records were used to calculate time- ignition. since-last-fire and number-of-fires over a 44-year period Here, we demonstrate the value of understanding the (1970-2013). Modelling was restricted to soft spinifex- differential value of multiple sources of uncertainty— dominated (Triodia pungens Brown 1810) sandplain both model and scenario uncertainty—when managing habitat, as no L. kintorei burrows were found outside threatened species in fire-prone landscapes. The aim this habitat type, despite 11 years of search effort. These of this study is to illustrate how metapopulation mod- data support local knowledge about the specificity of L. elling, combined with a spatially-explicit fire model, can kintorei to soft spinifex habitats within the study area (D. describe the likely fate of a species under a range of Moore, personal communication, 2013). future fire scenarios, and can identify fire management Competing generalized linear models (GLM: McCul- strategies most likely to secure its persistence. We predict lagh & Nelder 1989) were fitted and evaluated in R (R the medium-term (50 years) impacts of three different Core Team 2015), describing L. kintorei presence/absence fire regimes on the persistence of L. kintorei metapopula- as a function of environmental variables. Candidate vari- tion on Newhaven Wildlife Sanctuary, central Australia, ables were chosen based on ecological relevance (sensu under three competing models of species response to Austin 2002). Generalized additive models (Hastie & Tib- fire. Our analysis supports the choice of fire management shirani 1990) and boosted regression trees (Elith et al. practices for conserving L. kintorei within fire-prone 2008) were fitted to investigate the effect of mod- spinifex (Triodia spp.) habitat. We provide general elling method. Models were evaluated using: (1) AIC insights into the use of metapopulation models for eval- (Akaike 1973); (2) deviance explained; and (3) predic- uating management options in the face of uncertainty. tive discrimination (ROC curve: Hanley & McNeil 1982). 182 Conservation Letters,