A global expert elicitation on present-day human-fire interactions

Abstract

Human fire use contributes to fire regimes and benefits societies worldwide yet is poorly understood at the global scale. We present the Global Fire Use Survey (GFUS), an effort to elicit and systematize knowledge about fire use from experts, including academics and practitioners. The GFUS data cover the stakeholders using fire, reasons for and seasonality of burning, recent trends in anthropogenic ignitions and burned area and the presence/absence and effectiveness of different policy interventions targeting fire use. The survey garnered 311 responses for regions covering over 50% of the Earth's ice-free land, improving on the coverage of literature syntheses on fire use. Here, we analyse the data on the distribution of fire use and policy interventions. The survey suggests that the most widespread fire users are Indigenous and local people burning to meet objectives associated with small-scale livelihoods and cultural priorities, whereas burning by commercial land users, state agencies and non-governmental organizations is less widespread. Regulatory restrictions are the most common policy interventions targeting fire use but are ineffective in achieving their aims in regions with higher burned area. While community-led governance of burning is rarer, it was deemed more effective than restrictive policy interventions, particularly in regions with higher burned area.This article is part of the theme issue 'Novel fire regimes under climate changes and human influences: impacts, ecosystem responses and feedbacks'.

Keywords: contested knowledges; expert elicitation; fire policies; human fire use; indigenous fire ecologies.

Figure 1.

Global distribution of fire use, grouped by fire user type. Human fire is widespread and primarily used by small-scale livelihood and/or cultural fire users. State/NGO fire use is the most spatially heterogeneous, with apparently widespread use in Australia and parts of southern Africa but with seemingly less uptake in South America. A value of 0 denotes that all respondents agreed the fire user types to be absent, while 1 indicates agreement between respondents that the fire user type is present. Values between 0 and 1 indicate that survey respondents disagreed on whether the fire user type is present (i.e. 0.5 indicates half believed it was present, half absent).

Figure 2.

Distribution of selected policy types, demonstrating contrasting approaches to fire policy—outright bans on fire use (option a in table 1), economic incentives to reduce fire use (option e) and community-led governance of fire (option g). A value of 0 denotes that all respondents agreed the policy intervention not to be present, while 1 indicates agreement between respondents that the policy intervention is present. Values between 0 and 1 indicate disagreement between survey respondents, perhaps indicating that policy only applies in restricted areas or has limited impact or enforcement.

Figure 3.

Distribution of policy intervention types and perceptions of their effectiveness. (A) Policy effectiveness given the presence/absence of fire use by different fire user types; (B) Survey respondents’ perceptions of effectiveness of policy intervention types across quartiles of GFED5 burned area and (C) respondents’ perceptions of effectiveness of policy intervention types based on their sectors. GFED5 = 5th Global Fire Emissions Database. Fire use restricted is the mean of values for options b–d in table 1, which cover different forms of fire-use restriction.