Droughts and deforestation: Does seasonality matter?

PLOS ONE RESEARCH ARTICLE Droughts and deforestation: Does seasonality matter? Giuliaz Vaglietti ID1,2*, Philippe Delacote1,2, Antoine Leblois ID2,3 1 Université de Lorraine, AgroParisTech-INRAE, BETA, Nancy Cedex, France, 2 Climate Economics Chair, Palais Brongniart, Paris, France, 3 CEE-M Univ Montpellier, CNRS INRAE Institut Agro, Montpellier, France * a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Vaglietti G, Delacote P, Leblois A (2022) Droughts and deforestation: Does seasonality matter? PLoS ONE 17(10): e0276667. https://doi. org/10.1371/journal.pone.0276667 Editor: Alison Parker, Cranfield University, UNITED KINGDOM Received: June 2, 2022 Abstract Extreme weather events, particularly droughts, have strong impacts on the livelihoods of populations in rural areas. In a context of low access to insurance and credit markets, households respond to such shocks by implementing different risk-management strategies, which in turn are likely to have an impact on the environment, in particular through land-use changes and deforestation. This paper contributes to the emerging literature on the links between droughts and deforestation: (1) distinguishing responses to previously experienced droughts versus current droughts, and (2) disentangling the time of the agricultural season at which droughts occur. We show that deforestation declines whenever a drought occurs during the growing season, while it increases whenever a drought occurs during the harvesting season. These impacts are mitigated within protected areas and are exacerbated in more accessible locations, i.e., areas within 4 hours of travel time of main/major cities. By contrast, deforestation outcomes following droughts that occur during the planting season depend on whether the crop considered is maize or cassava. Accepted: October 12, 2022 Published: October 27, 2022 Copyright: © 2022 Vaglietti et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: It was created an open ICPSR data Repository (Vaglietti G, Delacote P, Leblois A. Data Repository - Supplementary Materials to Droughts and deforestation: Does seasonality matter? Vaglietti et al. 2022. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor]. 2022; doi: https://doi. org/10.3886/E178361V1.), reporting: a data paper describing the databases in use and how those have been processed, a.csv file containing the final data sample, a geopackage matrix to associate each unit of observation to a specific geographic location, and an R code to replicate the econometric model. Additionally, the raster files 1 Introduction In a context of climate change, shocks linked to meteorological hazards threaten a large number of actors, particularly in countries relying on subsistence rainfed agriculture. In these areas, which are characterized by poor access to credit and insurance markets, a large number of risk-management strategies may be implemented by rural households. Some of these strategies are likely to be related to forest cover change, as they may engender agricultural expansion, forest loss and degradation or, conversely, abandonment of agricultural lands. Such outcomes emphasize the potentially decisive role of meteorological hazards on land use changes. The literature has recently begun to address the link between droughts and forest loss. Our contribution is two-fold. First, risk-management strategies to face extreme weather events may be implemented either ex-ante or ex-post. Agricultural households adopt ex-ante adaptation strategies based upon their experience of past droughts in an attempt to anticipate future ones. On the other hand, these same households may implement coping strategies when a drought actually occurs (ex-post), as a short-term response to the shock. Our first contribution is to PLOS ONE | https://doi.org/10.1371/journal.pone.0276667 October 27, 2022 1 / 21 PLOS ONE underlying the results presented in the study are fully and freely available at: GADM (https://gadm. org/) for borders, the Global Forest Change Database (https://earthenginepartners.appspot. com/science-2013-global-forest/download_v1.7. html) for forest cover and forest losses, CHIRPS2.0 (https://www.chc.ucsb.edu/data/chirps) for Africa monthly precipitations, the Crop Calendar Dataset (https://sage.nelson.wisc.edu/data-andmodels/datasets/crop-calendar-dataset/) for crops’ agricultural seasonality, the Global Accessibility Map (https://forobs.jrc.ec.europa.eu/products/ gam/download.php) for travel time to major cities and at the World Database on Protected Areas (https://www.protectedplanet.net/en/thematicareas/wdpa?tab=WDPA) to assess protected areas location. References to the DOI are added as follows in the manuscript: “Final data, information on how these were retrieved and processed, as well as codes for the econometric analysis are available in the data repository supporting this publication [Vaglietti et al. 2022]” (P.4, lines 145147) Additionally, the underlaying databases are freely available under a Creative Commons Attribution 4.0 International License at the following links: GADM (https://gadm.org/), Global Forest Change Database (https:// earthenginepartners.appspot.com/science-2013global-forest/download_v1.7.html), CHIRPS-2.0 (https://www.chc.ucsb.edu/data/chirps), Crop Calendar Dataset (https://sage.nelson.wisc.edu/ data-and-models/datasets/crop-calendar-dataset/), Global Accessibility Map (https://forobs.jrc.ec. europa.eu/products/gam/download.php), World Database on Protected Areas (https://www. protectedplanet.net/en/thematic-areas/wdpa?tab= WDPA). Funding: The author(s) received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. Droughts impacts on DRC forest cover deepen the analysis of household risk-management strategies by comparing how the experience of past and current droughts affect forest loss. Despite only the shock-response strategies having an impact on land-use change are addressed here, the more general terms adaptation and coping are used instead of risk-management. It allows to highlight the differences in the time-response of the strategies: adaptation is used for response induced by past shocks while coping is used for responses induced by current shocks. Second, the literature addressing deforestation usually deals with annual droughts without looking at the timing of shocks across the agricultural season. However, the timing of rainfall is particularly important. In this article, the effects of experienced and current droughts are broken down according to seasonality, i.e., the time of the agricultural cycle at which a drought occurs. Our second contribution is to distingu (...truncated)