Formulating conservation targets for a gap analysis of endemic lizards in a biodiversity hotspot Verônica de Novaes e Silva a , Robert L. Pressey b , Ricardo B. Machado a , Jeremy VanDerWal c , Helga C. Wiederhecker a , Fernanda P. Werneck d , Guarino R. Colli a,⇑ a Departamento de Zoologia, Universidade de Brasília, 70910-900 Brasília, DF, Brazil b Australian Research Council Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, QLD 4811, Australia c School of Marine and Tropical Biology and Division of Research and Innovation eResearch Centre, James Cook University, Townsville, QLD 4811, Australia d Programa de Coleções e Acervos Científicos, Instituto Nacional de Pesquisas da Amazônia, Av. André Araújo 2936, 69060-000 Manaus, AM, Brazil article info Article history: Received 30 May 2014 Received in revised form 28 August 2014 Accepted 1 September 2014 Keywords: Conservation Savanna Fragmentation Reptiles Neotropics Reserves abstract Species gap analyses that adopt conservation targets based on individual species attributes recognize that some biodiversity features need more protection than others and should lead to better outcomes than uniform conservation targets. In the Brazilian Cerrado hotspot, 4 of the 30 endemic lizard species are included in the IUCN or Brazilian red lists of threatened species. For 18 species with more than 5 occur- rence records, we produced distribution models using Maxent and for 12 species with less than 5 occur- rence records we used a 5 km radius around the records to indicate distributions. For all species, we estimated habitat loss after discounting cleared areas from indicated distributions. Non-modeled species were considered as truly restricted-range endemics and had conservation targets set a priori as 100%. We formulated conservation targets for 18 modeled species based on three characteristics: natural rarity, vulnerability, and life-history. We estimated vulnerability from a model of future habitat loss across the Cerrado, derived with Maxent. We then performed a gap analysis considering strictly protected con- servation areas. We applied percentage targets (between 12% and 23%) to estimated species distributions prior to habitat loss and evaluated the targets against the presence of the species within strictly protected conservation areas. Disturbingly, only one species is adequately protected by the current system of pro- tected areas. We also found that one species is a minor conservation gap, whereas the remaining 28 spe- cies are either major (13) or total (5) conservation gaps. Habitat loss has erased a significant fraction of the original distribution of Cerrado endemic lizards and the existent network of protected areas is wholly inadequate to ensure their conservation. The use of conservation targets based on natural rarity, vulner- ability, and life-story will support more defensible conservation guidelines than commonly used uniform targets for this threatened Neotropical savanna biome. Ó 2014 Elsevier Ltd. All rights reserved. 1. Introduction The protection of natural areas is a historical concern of human- kind. However, the effectiveness of many protected areas (PAs) around the world is questionable, because they have often been established for reasons unrelated to biodiversity conservation, such as their scenic value or lack of competing interests (Andelman and Willig, 2003; Pressey and Tully, 1994; Rodrigues et al., 2004b; Rouget et al., 2003a; Scott et al., 2001). To be effec- tive, a PA system should be composed of reserves that complement each other in their biodiversity attributes, minimizing redundancy across space (Margules and Pressey, 2000; Pressey and Nicholls, 1989). They should also be representative, containing samples of these attributes at adequate levels to ensure their long-term per- manence and viability (Pressey and Nicholls, 1989); otherwise, gaps in regional biodiversity conservation will occur (Jennings, 2000; Scott et al., 1993). Among the several tools to make PA sys- tems more representative (Margules and Pressey, 2000), gap anal- ysis has been successfully used in conservation planning (Catullo et al., 2008; De Klerka et al., 2004; Oldfield et al., 2004; Paglia et al., 2004). To detect whether a PA system adequately protects a given spe- cies or taxonomic group of interest, a gap analysis requires: (1) an estimate of its distribution in the region, (2) the identification of protected sites (Scott et al., 1993), and (3) the definition of explicit http://dx.doi.org/10.1016/j.biocon.2014.09.016 0006-3207/Ó 2014 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. Tel.: +55 61 3107 3013; fax: +55 61 3202 2039. E-mail address: grcolli@unb.br (G.R. Colli). Biological Conservation 180 (2014) 1–10 Contents lists available at ScienceDirect Biological Conservation journal homepage: www.elsevier.com/locate/biocon