TECHNICAL ADVANCE Tipping points in tropical tree cover: linking theory to data EGBERT H. VAN NES*, MARINA HIROTA* † , MILENA HOLMGREN ‡ andMARTEN SCHEFFER* *Department of Aquatic Ecology and Water Quality Management, Wageningen University, P.O. Box 47, Wageningen NL-6700 AA, The Netherlands, †Centre for Monitoring and Warnings of Natural Disasters, Cachoeira Paulista 12630-000, SP Brazil, ‡Resource Ecology Group, Wageningen University, P.O. Box 47, Wageningen NL-6700 AA, The Netherlands Abstract It has recently been found that the frequency distribution of remotely sensed tree cover in the tropics has three dis- tinct modes, which seem to correspond to forest, savanna, and treeless states. This pattern has been suggested to imply that these states represent alternative attractors, and that the response of these systems to climate change would be characterized by critical transitions and hysteresis. Here, we show how this inference is contingent upon mechanisms at play. We present a simple dynamical model that can generate three alternative tree cover states (for- est, savanna, and a treeless state), based on known mechanisms, and use this model to simulate patterns of tree cover under different scenarios. We use these synthetic data to show that the hysteresis inferred from remotely sensed tree cover patterns will be inflated by spatial heterogeneity of environmental conditions. On the other hand, we show that the hysteresis inferred from satellite data may actually underestimate real hysteresis in response to climate change if there exists a positive feedback between regional tree cover and precipitation. Our results also indicate that such posi- tive feedback between vegetation and climate should cause direct shifts between forest and a treeless state (rather than through an intermediate savanna state) to become more likely. Finally, we show how directionality of historical change in conditions may bias the observed relationship between tree cover and environmental conditions. Keywords: alternative stable states, climate vegetation feedback, remote sensing, savanna, simple model, tropical forests Received 27 May 2013; revised version received 29 August 2013 and accepted 11 September 2013 Introduction Remotely sensed tree cover distributions in the tropics show three distinct modes: forest, savanna, and a tree- less state, which have been suggested to represent alternative stable states (Hirota et al., 2011; Staver et al., 2011). Although there is no coherent theory to explain the complex relation among three alternative tree cover states and precipitation, much work has been done on the transitions between savanna and forest, and between savanna and treeless states. Bistability between savanna and forest is thought to be mainly dri- ven by fire (Cochrane et al., 1999; Bond, 2008; Staver et al., 2011; Hoffmann et al., 2012; Staver & Levin, 2012). Wildfires can keep the savannas open but once a critical tree cover has been reached, the suppression of grass fuels reduces fire occurrence and the extent of burnt areas (Liedloff & Cook, 2007; Archibald et al., 2009; Warman & Moles, 2009). In addition to tree density, also precipitation plays a role in the occurrence of fires. Under very wet conditions, fires are rare and even when triggered they are not sufficiently intense to dam- age trees; on the other hand, under arid to semiarid conditions, tree cover is mostly limited by water sup- plies and less affected by fires (Bond et al., 2005; Higgins et al., 2007; Archibald et al., 2009). Transitions to entirely treeless conditions have been described mostly in dry climates where a treeless state can be difficult to be (re)colonized by trees either because seedlings require the protection by adult nurse plants to survive the first life stages (Holmgren et al., 1997; Rietkerk & Van De Koppel, 1997) or because seedlings recruit only during rainy periods when they grow suffi- ciently fast and in large numbers to successfully escape from herbivores, fire, and drought (Holmgren et al., 2006; Scheffer et al., 2008). This may create an elevated mortality at low densities (which is called inverse density dependence or the Allee effect, see e.g., Courchamp et al., 1999). If this effect is strong enough it could cause the system to remain trapped in a stable treeless state (Rietkerk & Van De Koppel, 1997). The availability of detailed satellite data of world- wide tree cover may offer unique possibilities to quan- tify the stability of these alternative stables states (Hirota et al., 2011). However, interpretation of such Correspondence: Egbert H. van Nes, tel. +31 317482733, fax +31 317484411, e-mail: egbert.vannes@wur.nl © 2013 John Wiley & Sons Ltd 1016 Global Change Biology (2014) 20, 1016–1021, doi: 10.1111/gcb.12398