Bio-cultural fire regions of Guinea-Bissau: Analysis combining social research and satellite remote sensing

Highlights

• There is a need to know more about why and how people use fire as a local landmanagement tool.

• Burning practices are diverse, reflecting cultural, socioeconomic, demographic and agro-ecological heterogeneity.

• The recent development of new farming systems, notably cashew plantations, has induced changes in traditional uses of fire.

• New fire practices in the agricultural frontier lead to a decrease in uncontrolled burning.

• Public policies need to engage with resource users, their traditional knowledge and ability to adapt to new circumstance.

Abstract

Ever since colonial times, the rural inhabitants of Guinea-Bissau have been blamed for lighting uncontrolled fires all over the country. Based on in-depth ethnographic research in two regions, a country-level rapid rural appraisal, and analysis of satellite active fire data, this essay shows how burning practices are, however, diversified according to cultural, socio-economic, demographic and agro-ecological conditions, and how they have been changing recently as a locally-developed adaptation to new farming systems. Many new bush-fire uses correspond to current best practices (e.g., use of firebreaks, backfiring, burning at cooler hours of the day), but under a scenario of changing climate and land uses there is room for improving fire management to reduce negative environmental impacts, while preserving cultural, economic and biodiversity benefits generated by some traditional burning practices. The design of public policies for reducing material damages, biodiversity loss, and carbon emissions from fires requires an understanding of their local drivers. Following from that, the development of ecological management practices must engage with farmers’ knowledge and networks/institutions and meet their priorities and needs.

Introduction

Vegetation burning in Africa was long considered an archaic, disorderly, and environmentally damaging practice, responsible for forest loss, degradation of savanna areas, and desertification, which ought to be eradicated for the sake of the social progress of traditional societies. However, many African ecosystems are fire-dependent for the provision of ecological services and biodiversity conservation (Bowman et al., 2011) and various authors pointed out the importance of early dry season fires in preventing late season wildfires (Kull, 2002b; Kull & Laris, 2009; Russel-Smith et al., 2013).

Archibald (2016) underlined that a challenge of contemporary fire management is to balance the needs of large rural populations against national and global views on the acceptability of various fire practices, while Kull and Laris (2009) remarked that after many decades of hostility, traditional fire uses are becoming better understood and accepted. In reality, most fire use in African savannas and croplands is a rational, regular activity developed for a variety of purposes, including animal husbandry, to facilitate hunting and food gathering, and to reduce the likelihood of wildfires (Caillault, Ballouche, & Delahaye, 2015). In his article “Why burn the bush?” Hough (1993) identified the underlying logics of savanna seasonal burning in West Africa. In the case of Ghana, Appiah, Damnyag, Blay, and Pappinen (2010, p. 557) concluded that shifting cultivation was the main driver of fires, followed by hunting practices and by revenge actions related to conflicts. Mbow, Nielson, and Rasmussen (2000) illustrated that in East-Central Senegal, agricultural fires were of minor importance in relation to those used by pastoralists to promote the growth of herbaceous vegetation. According to this study, fire regimes and local practices and perceptions of the pastoralist Pular were diversified, reflecting a well-adapted fodder production strategy (e.g., Mbow et al., 2000).

Indigenous wildfire mitigation practices in Africa are still an understudied topic. Dugast (2006), Erikson (2007), Bird, Codding, Parker, and Jones (2008) and Walters (2012), who described how fire practices were ritually controlled among certain groups, highlighted the importance of religion. Appiah et al. (2010, pp. 557–8) also argued that indigenous mitigation practices in Ghana are mostly based on prohibitions and taboos and the establishment of fire belts around agricultural fields. For the case of the African continent, Pereira, Oom, Pereira, Turkman, and Turkman (2015) revealed how religion modulates weekly cycles of fire activity, contrasting Christian regions (with fire activity minima on Sundays) and Muslim regions (with fire activity minima on Fridays). Consequently, we may speak of the existence of “fire cultures” – a concept introduced by Wardell, Nielsen, Rasmussen, and Mbow (2004, p. 368) −, as human pyrogenic behaviour is socially constructed and subject to changes across space and time. Building on this concept, and combining it with the concept of “bio-cultural diversity” (e.g., Maffi, 2007), we propose the concept of “biocultural fire regions”, defined as specific landscapes emerging across time and space from the interplay between biocultural diversity and culturally determined fire practices.

Population density can also be an important factor shaping fire regimes, although its impact may vary according to locally specific circumstances and the way factors interact among them (Kull, 2003, p. 157; Archibald, Roy, Van Wilgenz, & Scholes, 2009, pp. 614–627). For Southern Africa, Archibald et al. (2009, p. 627) established that population densities below 10 people/km² were associated with fewer fires, but no relation was found with the burnt area; however, at higher population densities there was a decrease in the area being burnt. Kull's (2003, pp. 157) study of long-term fire practices in Madagascar concluded that the use of fire has decreased in areas where there was an increase in population density and/or land use intensification. The same applies to some regions of Senegal and Burkina Faso, but in the case of some sparsely populated grasslands of Madagascar fires have a moderate to low intensity (see Kull & Laris, 2009, pp. 185–186). Other important factors shaping fire regimes are land use/cover changes, namely the development of agricultural frontiers in Africa, received little attention (Amanor & Paby, 2007, pp. 59–61). Nevertheless, Grégoire et al. (2012) showed that cropland expansion has generated a significant declining trend in area burned, at continental scale.

Fire governance is also a topic of major importance for the development of sustainable practices. The impact of state forest policies in Africa (state ownership and control of the allegedly vacant forested land; exclusion or restriction of access to forest reserves by local communities; use of sanctions) is perhaps the subject that most attracted the attention (and critiques) of fire researchers (e.g., Fairhead & Leach, 1996; Huffman, 2013; Laris & Wardell, 2006; Mbow et al., 2000; Nielsen, 2000; Preece, 2007; Wardell et al., 2004). Similarly, some scholars focused on the impact of protected areas in preventing wildfires (Adeney, Christensen, & Pimm, 2009; Erikson, 2007; Hough, 1993; Jeffery et al., 2014; Walters, 2015, p. 281). Recently, some studies emphasized the need to adopt participatory fire management approaches relying on local knowledge, strategies and networks (e.g., Mbow et al., 2000; Juvélius, 2003 and, 2004; Erikson, 2007; Erikson & Gill, 2010; Appiah et al., 2010; Walters, 2015; Fowley & Welch, 2018). Laris and Wardell (2006, p. 285) have argued that fire management issues ought to be addressed at a local level and framed according to the needs and epistemologies of local resource management objectives. However, while most colonial and post-colonial fire control measures have long proved inefficient (e.g., Kull, 2002a, 2003; Laris & Wardell, 2006; Walters, 2015), community participation in integrated fire management is still being tested and has presented mixed results (e.g., Appiah et al., 2010; Bird et al., 2008; Goldammer, Frost, Jurvélius, Kamminga, & Kruger, 2004, p. 282; Kull & Laris, 2009; Russell-Smith et al., 2013). In the case of Zambia, Erikson (2007) illustrates that although changes in the farming system would allow for more sustainable practices, late dry season burning continues due to local power structures and a lack of cooperation between traditional and state authorities. Thus, how to create a sustainable fire-governance framework is still a topic needing further research.

Wardell et al. (2004, p. 357), consider that West Africa constitutes a special case to study fire regimes, because of the role played by the pronounced north-south precipitation gradient, which not only influences environmental and agricultural conditions, but also local perceptions and attitudes toward uses of fire. Additionally, political, demographic and socio-cultural factors also affect the use of fire (e.g., Bowman et al., 2011) in a region inhabited by multiple ethnic groups. Among the studies that have analyzed changes in fire practices across time and space through local agency or external intervention in West Africa are Mbow et al. (2000), Wardell et al. (2004), Laris and Wardell (2006), Erikson (2007), Bowman et al. (2011), and Huffman (2013). Relying on this body of research, we argue that in order to create a sustainable fire governance framework, not only there is a need to know more about why and how people use fire and the drivers of change of fire practices across time and space, but also public policies ought to incorporate the knowledge and priorities of local resource users, engage with their networks/institutions, and recognize their ability to adapt to new circumstances.

The literature review presented above sets the stage for the subsequent description of the highly diversified fire practices across Guinea-Bissau, their complex drivers, and the way these practices and the perceptions about fire are changing due to the creation of a cashew plantation agricultural frontier in a path-differentiated way. These changes, combined with the relative scarcity of studies of changes in local fire management knowledge and practices in Africa (and a total absence of fire studies in Guinea-Bissau), provide the main motivations for the present research. Mixed methods research ─ combining remotely sensed and other geographical information with social science research techniques, which have been broadly called people and pixels (e.g., Coughlan & Petty, 2012) ─ are highly suitable to the study of coupled human-environment systems. They have been used to analyze relationships between fire and land cover change, and to understand cultural processes underlying observed fire activity patterns (Carmenta, Parry, Blackburn, Vermeylen, & Barlow, 2011; Sorrensen, 2000; Coughlan & Petty, 2012), but if not properly applied they can misrepresent reality (for a critique see Vayda, 2006). In this article we used a mixed methods approach to study fire use in Guinea-Bissau, including assessment of the existence of “fire cultures”, “bio-cultural fire-regions” and grassroots-induced changes in fire practices.

Given the unavailability of systematic satellite active fire data prior to 2001 (and that the drastic changes in LUCC started in the 1990s), and of detailed land cover maps, and also considering the ongoing process of cashew tree plantations, the “from-the-sky” magnitude and regional temporal frames of the alleged change in fire regimes could not be fully confirmed. Nonetheless, we present some broad conclusions and discuss the entry-points that policy and development interventions can explore to improve local practices and institutions for preventing wildfires and reduce carbon emissions in the West African countries where a similar agricultural frontier is developing.