Contents lists available at ScienceDirect Global Environmental Change journal homepage: www.elsevier.com/locate/gloenvcha Diverging forest land use dynamics induced by armed conflict across the tropics David M. Landholm a,b, ⁎ , Prajal Pradhan a , Juergen P. Kropp a,c a Potsdam Institute for Climate Impact Research (PIK), Member of the Leibniz Association, P.O. Box 60 12 03, D-14473 Potsdam, Germany b Resource Economics Group, Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany c Institute for Environmental Science and Geography, University of Potsdam, 14476 Potsdam, Germany ARTICLE INFO Keywords: Armed conflict Deforestation Internally displaced persons Migration Land use change drivers Colombian conflict ABSTRACT Armed conflicts trigger region-specific mechanisms that affect land use change. Deforestation is presented as one of the most common negative environmental impacts resulting from armed conflicts, with relevant consequences in terms of greenhouse gas emissions and loss of ecosystem services. However, the impact of armed conflict on forests is complex and may simultaneously lead to positive and negative environmental outcomes, i.e. forest regrowth and deforestation, in different regions even within a country. We investigate the impact that armed conflict exerted over forest dynamics at different spatial scales in Colombia and for the global tropics during the period 1992–2015. Through the analysis of its internally displaced population (departures) our results suggest that, albeit finding forest regrowth in some municipalities, the Colombian conflict predominantly exerted a negative impact on its forests. A further examination of georeferenced fighting locations in Colombia and across the globe shows that conflict areas were 8 and 4 times more likely to undergo deforestation, respectively, in the following years in relation to average deforestation rates. This study represents a municipality level, long-term spatial analysis of the diverging effects the Colombian conflict exerted over its forest dynamics over two distinct periods of increasing and decreasing conflict intensity. Moreover, it presents the first quantified estimate of conflict's negative impact on forest ecosystems across the globe. The relationship between armed conflict and land use change is of global relevance given the recent increase of armed conflicts across the world and the importance of a possible exacerbation of armed conflicts and migration as climate change impacts increase. 1. Introduction A growing body of literature has attempted to improve our under- standing of the existing bi-directional relationships between armed conflicts (including warfare) and the environment (e.g. land use and climate). There is evidence relating past changes in temperature and precipitation to the onset of armed conflict through region-specific mechanisms (Scheffran et al., 2012; Hsiang et al., 2013; Gleick, 2014; Schleussner et al., 2016; Werrell and Femia, 2017; Sterzel et al., 2014), suggesting that if future populations respond similarly to past popula- tions, then anthropogenic climate change has the potential to sub- stantially increase conflict around the world, relative to a world without climate change. On the other hand, armed conflicts constitute the most drastic shocks that can impact societies and, in turn, affect land use change (LUC). Although several types of LUC are possible (Baumann and Kuemmerle, 2016), those affecting forest land use (LU) are of particular interest, given their crucial role in carbon storage and provisioning of ecosystem services (Foley, 2005; Baccini et al., 2012). Deforestation represents the most negative environmental impact arising from armed conflict (Baumann and Kuemmerle, 2016; Suarez et al., 2017) and forests are reported to have suffered during times of violence in the Democratic Republic of Congo (Butsic et al., 2015; Nackoney et al., 2014), Rwanda (Ordway, 2015), Vietnam (Van et al., 2015) and in regions of Central Africa (Basnet and Vodacek, 2015). However, the impact of conflict on forests is not unidirectional, and reported bene- ficial impacts also exist for Sierra Leone (Burgess et al., 2015; Wilson and Wilson, 2012), South Sudan (Gorsevski et al., 2012), El Salvador (Hecht et al., 2006), Cambodia (Le Billon, 2000) and Nicaragua (Berhe, 2007). Albeit forest ecosystems can suffer directly in some extreme cases [e.g. defoliation with Agent Orange during the Vietnam War (Van et al., 2015)], armed conflicts’ impact on forests are more likely to occur through an increase or decrease of existing regional drivers of LUC by https://doi.org/10.1016/j.gloenvcha.2019.03.006 Received 2 February 2018; Received in revised form 8 October 2018; Accepted 27 March 2019 ⁎ Corresponding author at: Telegraphenberg A56, room 214, D-14473 Potsdam, Germany. E-mail address: landholm@pik-potsdam.de (D.M. Landholm). Global Environmental Change 56 (2019) 86–94 0959-3780/ © 2019 Elsevier Ltd. All rights reserved. T