Diverging fire trends in Northern Ghana’s savanna landscape: Insights from remote sensing and institutional perspectives

Diverging fire trends in Northern Ghana’s savanna landscape: Insights from remote sensing and institutional perspectives | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Diverging fire trends in Northern Ghana’s savanna landscape: Insights from remote sensing and institutional perspectives Rahinatu Sidiki Alare, Emma Tebbs, Kate Schreckenberg This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6441082/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Background Globally, there are growing demands for evidence-based policies to manage wildfires. Currently, fire management in Northern Ghana relies on policies and projects developed without a comprehensive understanding of fire trends and drivers. This study analysed spatio-temporal trends in burned areas (500m MODIS and 30m Landsat products), active fires and fire seasonality using linear regression analysis to investigate shifts in fire regimes between 2000 and 2022. Rainfall and land cover changes during this period and institutional perspectives of the observed trends were also examined. Results When averaged across Northern Ghana, MODIS burned area data revealed a significant decreasing trend, while Landsat burned area, and active fires showed no significant trend. When disaggregated by region, MODIS burned area showed significant decreasing trends for the Savannah Region, Upper East and Upper West Regions. Conversely, Landsat burned area showed no trend in all regions of Northern Ghana. Active fires increased significantly in the Northern Region. Active fire data also revealed a significant shift in fire seasonality in Northern Ghana towards more mid-dry season fires. Institutional perspectives attributed the decline in large fires (identified by MODIS) to the success of interventions designed to reduce uncontrolled burning (locally referred to as bushfires). Conversely, increasing small fires (Landsat burned area and MODIS active fires) were perceived as being associated with smallholder livelihoods dependent on burning, which aligned with the increase in cropland extent observed in land cover data. Conclusion Our results highlight that the scale and resolution of fire datasets are crucial considerations for analysing fire trends. Aggregating data across the whole of Northern Ghana obscured contrasting trends in individual regions. The different trends observed between MODIS and Landsat burned areas suggest a shift from large fires towards smaller ones in Northern Ghana. Institutions expressed a shared objective of suppressing all fires; however, when designing fire management policies, it is important to consider the type of fire, since fire trends, drivers and impacts can vary depending on the size and timing of burning, and the associated land use. Northern Ghana savanna fire regimes institutions MODIS burned area MODIS active fires Landsat burned area Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Introduction African savannas are known for their high flammability, and fire plays a significant role in shaping the composition and structure of these ecosystems (Bowman et al. 2009 , Laris et al. 2016 ). It is estimated that approximately 250 million hectares (Mha) of land are burned in Africa each year, accounting for an average of 70% of the global annual burned area (Giglio et al. 2009 ). These fires also contribute to land use modifications and greenhouse gas emissions (Hantson et al. 2015 , Van Der Werf et al. 2017 ). In the past, studies on fires were spatially limited to regional scales, but satellite remote sensing (SRS) now allows for extensive detection of burned and active fires, enabling analysis of conditions and monitoring of landscape changes (Chuvieco et al. 2019 , Szpakowski and Jensen 2019 , Wooster et al. 2021 ). This study focuses on two key applications of SRS data to understand shifting fire regimes in Northern Ghana’s savanna landscape: active fire and burned area. Fire regimes in this context refer to the spatiotemporal trends in active fire, burned area and fire seasonality. Active fire involves real-time monitoring of fire through a technique that monitors the thermal contrast between burnings and the background, while burned area is a post-fire analysis that relies on the changes in the spectral signatures of fire-affected vegetation and soil in parts of the electromagnetic spectrum (Chuvieco et al. 2019 , Wooster et al. 2021 ). Currently, there is an increase in the number of global active fires and burned area products from different satellite sensors, advancing studies on fire regimes, fire risk, climate change, and air quality studies (Szpakowski and Jensen 2019 ). Despite this progress, existing global fire products have been based on coarse spatial resolution sensors (e.g., MODIS with a resolution of 250 m to 1 km) that contain a ‘low-resolution bias’ where larger fires are detected while the smaller ones go undetected (Boschetti et al. 2004 , Laris 2005 , Laris et al. 2018 ). Yet, these small fires account for 24% (0.78M km 2 /year) of African burned areas (Randerson et al. 2012 ). They are often used to achieve a range of livelihood activities such as agricultural land preparation, hunting, sprouting fresh grasses for grazing, clearing bushes and deterring wild reptiles from entering homes (Laris 2002 , Amoako and Gambiza 2020). Sometimes, these fires can become uncontrollable and expand into large, destructive ones (Amoako and Gambiza 2020). Hence, both large and small fires are present in this landscape. However, the sole reliance on coarse resolution sensors by existing studies to analyse fire regimes and their drivers in Africa, including Ghana(Kugbe et al. 2012 , Andela et al. 2017 , Dwomoh and Wimberly 2017 , Dahan et al. 2023 ) poses challenges in developing policies for effective fire management in these landscapes.Mistry and Berardi (2016) contend that such studies and the prevailing antifire policies undermine traditional fire knowledge and marginalise local resource users who depend on fire for their livelihoods. Additionally, proper data is crucial for creating tailored policies and interventions and assessing their effectiveness (Jerven 2016 ). Meanwhile, interdisciplinary researchers have highlighted the limitations of SRS of fires in explaining the drivers of burning in African savanna landscapes (Harwell 2000 , Dennis et al. 2005 , Laris 2011 ). This perspective is valid because while biophysical factors account for fire ignition, fire management is inherently social (McCaffrey et al. 2013 ). For instance, the actual start of fires is influenced by biophysical factors such as weather conditions, fuel load, and ignition sources. However, managing and controlling fires is significantly influenced by human activities, including socio-economic factors, and policies. An important area of fire management research is the influence of institutions (McCaffrey et al. 2013 , Steen-Adams et al. 2017). Institutions play a key role in influencing fire regimes through decision-making, governing land access, and implementing policies and interventions (Van Wilgen et al. 2004 , Kull and Laris 2009, Carmenta et al. 2019, Moura et al. 2019 ). These interventions include actions taken by state and non-state actors to achieve specific aims, including regulations, economic incentives, and interventions that elicit voluntary responses, such as communication campaigns (Smith et al. 2024 ). Given the complexities associated with establishing a direct correlation between the drivers and the observed spatiotemporal trends of fire regimes in African savanna landscapes using SRS (Laris 2005 ), integrating institutional perspectives can enhance our understanding of the drivers in this area. Expert interviews from formal institutions provide professional knowledge and interpretation (Kleemann et al. 2017 , Von Soest 2023 ) of fire regime drivers and spatiotemporal trends. Therefore, this study addresses these gaps by employing a mixed methods approach to understanding the types of fires and their drivers in Northern Ghana’s savanna landscape. Northern Ghana's savanna landscape provides an important context for this study, as 90% of the total land cover experiences annual bush burning during the dry season (MLFM 2006 ), leading to the implementation of various policies and interventions to manage fires in these areas. The study focuses on analysing the spatiotemporal trends in fire regimes and institutional perspectives for a more comprehensive understanding of the drivers, given that the reasoning behind these regimes is often under-studied (Laris 2011 ). Specifically, we asked: i. What are the spatiotemporal trends in burned areas, active fires, and fire seasonality over the last 21 years (2000–2022) in Northern Ghana as assessed by MODIS and Landsat satellite imagery? ii. How do institutional perspectives on shifting fire regimes relate to trends identified with remote sensing? iii. What are the drivers of these shifting fire regimes, including landcover, precipitation and social factors? Methods Study Site This study focused on the five regions of Northern Ghana (Fig. 1 ), where a greater proportion of the landscape (90%) experiences annual vegetation burning (MLFM 2006 ). Before 2017, Northern Ghana consisted of Northern, Upper East and Upper West Regions. In 2018, two new regions, Savannah and North-East were created by carving them out of the Northern Region. This administrative change led to the five regions in Northern Ghana. Northern Ghana is situated within the Guinea and Sudan savanna ecological zones. The region’s vegetation is characterised by fire and drought-resistant woody species including Shea ( Vitelleria paradoxa), bushwillow (Combretum spp), wild syringa (Burkea africana) and ‘abogo’ ( Isoberlinia doka) with common grasses such as Andropogon, Heteropogon and Hyparrhenia spp (Amoako and Gambiza 2019). It covers a total landmass of about 97,250 km 2 , with thin soils overlaid with an iron pan. An iron pan is a hard soil layer cemented by iron oxides and usually consists of sand or sand gravel. The area is bounded by Burkina Faso to the North and North-west, Togo to the East and Côte D’Ivoire to the West. The southern edge of the zone is bordered by the Oti and Bono Regions. Northern Ghana has a tropical climate with unimodal rainfall patterns influenced by two air masses: the North-east trade winds and the South-westerly. However, because of its proximity to the Sahel, Northern Ghana experiences a dry climate caused by the north-easterly winds (harmattan) between November and April, facilitating vegetative burning. Fire suppression policies have a long history in Northern Ghana’s savanna landscapes. These policies were initially driven by the colonial rulers’ interest in forest resources and their concerns about the effects of fire on vegetation, as well as its impacts on local climate, hydrology and downstream flow of major rivers like the Niger and Volta (Wardell et al. 2004 , Wardell and Lund 2006). Realising that complete fire suppression in this area was undesirable and impracticable, scientific research by colonial botanists embraced an early burning regime as it affected trees slightly less than late fires (Laris et al. 2017 ). Thus, early burning was recognised as a necessary tool in savanna land management towards the end of the colonial rule (Laris and Wardell 2006). However, post-independence, the country witnessed a return to restrictive fire policies because of the government's interest in protecting its important projects like the hydroelectric dam, irrigated agriculture, and industrial forest plantations (Wardell et al. 2004 ). Moreover, the drought events in the Sahel region in the 1980s triggered wildfires, destroying lives and properties in the country (Ampadu-Agyei 1988 ). This reignited discussion about bushfires and led to key fire policies like the Provisional National Defence Council (PNDC) Law 46 in the 1980s, PNDC Law 229 in the 1990s, the 2006 Wildfire Policy and other related sectoral policies. The PNDC Laws 46 and 229 had several stipulations and restrictions, with punishment for non-compliance, but never achieved their goal because local communities resisted and set more fires (Apusigah 2007 ). The 2006 wildfire policy is thought to be more inclusive and recognises traditional fire use, aiming to reduce a perceived bushfire menace in the country (Apusigah 2007 , Doe 2008 ). However, the various terms used to describe fires (e.g., bushfire, wildfire, traditional burning) in policy documents and academic papers, along with the prevailing narratives about fires, have led to a misunderstanding of the types of fires and the ones requiring management (Alare, Tebbs, and Schreckenberg, n.d.). Biophysical data collection and analyses We employed the Moderate Resolution Imaging Spectroradiometer’s (MODIS) burned area dataset (MCD64A1, 500 m spatial resolution) to analyse spatiotemporal burned area trends. Additionally, we used MODIS Active fire dataset (MOD14A1 V6, 1 km spatial resolution) to investigate temporal trends in active fires and fire seasonality (Giglio et al. 2021 ) between 2001 and 2022. It is worth noting that MODIS active fire products are more sensitive to smaller fires (≥ 10 ha) than MODIS-burned area products (≥ 169 ha) under pristine conditions (Fusco et al. 2019 ). We also used the Global Annual Burned Area Mapping (GABAM) dataset, derived from 30 m Landsat 8 Surface Reflectance imagery as described by Long et al. ( 2019 ). This enabled us to identify the types of fires and to compare and assess the spatiotemporal burned area trends with the MODIS burned area data, considering the limitations of MODIS satellite sensors in detecting small fires (Laris 2005 , Roteta et al. 2019 ). The trend analysis was limited to a 22-year duration because there was no satellite data available beyond this point. Although Landsat has a fine spatial resolution (30 m), it is constrained by low temporal resolutions (with observations only once every 16 days) and for areas like Northern Ghana where vegetation recovery is very quick after a fire, long temporal gaps usually result in higher omission errors (Bowman et al. 2003). Additionally, Long et al. ( 2019 ) assert that one of the drawbacks of the Landsat burned area data is the uncertainty associated with agricultural lands. In these areas, harvested or ploughed surfaces can be mistaken for burned areas due to similarities in spectral and temporal characteristics. While Sentinel-2 would have been ideal for this study due to its 10 m spatial resolution and high revisit frequency coverage, the short temporal coverage (data available since 2015) makes it less suitable for long-term fire regime analysis. The connection between cropland expansion and the decline of the global and regional burned areas (Andela et al. 2017 , Dwomoh and Wimberly 2017 ), requires an analysis of the spatiotemporal trends of land use land cover (LULC) in Northern Ghana. Like remotely sensed burned area products, LULC maps for Africa must provide accurate spatial and temporal resolution of land cover types and their extent for effective policy development (Silva et al. 2005 , Kerner et al. 2024 ). Therefore, we chose the Global Land and Aridity Discovery (GLAD) LULC dataset because of its fine spatial resolution (Landsat satellite image, 30 m) and long-term temporal coverage (data available for the years 2000 and 2020) (Potapov et al. 2020 ). Moreover, it is one of the most accurate cropland products (Kerner et al. 2024 ). Rainfall data was used as a predictor variable to analyse the mean annual rainfall and its influence on fire seasonality and fuel load (Dwomoh and Wimberly 2017 ). Due to rain gauge data sparsity in Ghana, we relied on Climate Hazards Group Infrared Precipitation with Station Data (CHIRPS) data for rainfall analysis from 2001 to 2022. CHIRPS gridded daily rainfall product (UCSB-CHG/CHIRPS/DAILY) combines satellite infrared data and in-situ precipitation station data to produce accurate, detailed and extended periods of rainfall estimates (Funk et al. 2015 ). Observed climate data (rainfall) from Tamale (Northern Region), Yendi (Northern Region), Bole (Savannah Region), Navrongo (Upper East Region) and Wa (Upper West Region) were also obtained from Ghana Meteorological Agency rain gauges to validate findings of the CHIRPS Rainfall analysis. Following the data collection, we used Google Earth Engine (GEE) to analyse all remotely sensed images. Each region's time series was exported as a CSV file for further statistical analysis and visualisation in R studio. For the statistical analysis in R studio, we performed a Shapiro-Wilk normality test to determine if the data distribution of burned area, active fires and rainfall required a parametric statistical analysis. We considered the Shapiro-Wilk test of normality due to its effectiveness in analysing small datasets (n < 50) and being less impacted by outliers (Mendes and Pala 2003, Mohd Razali and Bee Wah 2011). Since the results revealed normally distributed data for the burned area (W = 0.97, p = 0.73) and active fires (W = 0.93, p = 0.11) at a 95% confidence level, we employed linear regression analysis to evaluate the significance of the observed trends in the time series data. Concerning rainfall, the Shapiro-Wilk test showed that the data was not normally distributed (W = 0.9, p = 1.70 e − 12 ). Therefore, a non-parametric test such as the Mann-Kendall test was used to assess whether there was a monotonic trend in the time series data. Landcover data was presented as histogram charts. Institutional interviews To acquire expert knowledge on the perceived trends of burned areas, active fires, and cropland areas, as well as the drivers of these spatial and temporal trends in Northern Ghana’s savanna landscapes, we conducted semi-structured interviews with local government institutions (n = 6), and non-governmental organisations (NGOs) (n = 4) (Table 1 ). The study employed a combination of different sampling approaches to select interview participants. Purposive sampling was used in the selection of Government institutions given their formal responsibility in fire management as outlined in the 2006 Ghana Wildfire Policy. However, NGOs involved in forest and biodiversity conservation-related projects were identified based on snowball sampling while participants’ availability and willingness to participate determined the selection process through convenience sampling (Frey et al. 2000 ). The interviews were conducted between November 1 and December 7, 2022, and took place both virtually and face-to-face. The interviews focused on institutional perspectives of traditional burning practices and perceived changes in fire regimes. Additionally, a stakeholder engagement workshop was conducted on 16–17 October 2024 in Damongo in the Savannah Region to discuss pathways for promoting equitable fire management in Northern Ghana. During the workshop, stakeholders identified additional social drivers that influenced the contrasting trends observed in the MODIS and Landsat satellite data. This helped to improve the reliability and consensus of the responses. Table 1 Institutional perceptions about changing fire regimes in Northern Ghana Institution (government, unless otherwise stated) Location Perception of trends in burned areas and active fires Explanation for the observed trend Forestry Commission Upper East Region Decreasing in project communities Introduction of multilateral projects such as Sustainable Land and Water Management Projects (SLWMP), Ghana Landscape Restoration and Small-Scale Mining Project, (GLRSSMP) and Ghana Shea Landscape Emission Reduction Project (GSLERP) (2021–2028) have and are: sensitising communities on the effects of bush burning, forming community fire volunteers, restoring degraded lands through tree planting, providing improved crop varieties to farmers and strengthening local resource governance with chiefs playing active roles. Ghana National Fire Service Upper West Region Fluctuating This is influenced by a range of factors including livelihoods driven, low awareness of the impacts of bush burning. Savannah Integrated Biodiversity and Conservation Initiative (NGO) Savannah Region Decreasing Introduction of technologies such as tractors for ploughing and herbicides Ghana National Fire Service (West Gonja Municipal Office) Savannah Region Remained the same or increased for West Gonja District over the last four years Behavioural (not ‘rational’ and just part of their lifestyle). West Gonja Municipal Assembly Savannah Region Remained the same Behavioural (not ‘rational’ and just part of their lifestyle) Ministry of Agriculture Northern Region Decreasing in project sites Sensitizations, introduction of community fire volunteers and training on how to combat bushfires, presence of more elite chiefs who are environmental enthusiasts National Disaster Management Organisation Upper West Region Fires have remained the same in the region Livelihood-driven World Vision International (NGO) Upper East Region Decreasing in project communities and districts Community awareness, radio discussions, collaboration with GNFS to train communities on how to combat fires and the establishment of community fire volunteers. A Rocha Ghana (NGO) Northern Ghana Decreasing in project communities Influence of high-profile projects such GSLERP, GLRSSMP and Landscape and Environmental Agility across the Nation (LEAN) projects focused on land restoration and fire management. Nandom Deanery Integrated Rural Development Project (NGO) Upper West Region Decreasing in project communities Community awareness programmes, Ghana National Fire Service Northern Region Fluctuating Livelihood-driven, lack of awareness We transcribed the virtual interviews using Zoom and Microsoft Teams’ transcription software, while face-to-face interviews were recorded and transcribed using Otter AI software. However, these transcription tools could not accurately transcribe the interviews due to accent-related issues, necessitating additional manual editing. This involved listening to the recording and editing sentences and words that were not transcribed well. The transcripts were then coded into themes based on the principles of grounded theory (Glaser and Strauss 1967 ). This approach aims to develop theories that emanate from the knowledge and perceptions shared by interviewees concerning the subject matter. The NVivo 14 software aided the analysis of codes, memos, and quotations. Additionally, data from the stakeholder engagement workshop were used to explain the observed spatiotemporal trends. Results Spatio-temporal patterns of burned area, fire seasonality, rainfall and land use land cover Figure 2 presents trends in burned areas, active fires, and rainfall averaged across the whole of Northern Ghana from 2001 to 2022. MODIS 500 m burned area (Fig. 2 A) declined from approximately 3750 km 2 in 2001 to 3305 km 2 in 2022. The trend was significant (p = 0.076) at a 90% confidence level, with notable peaks in 2005 (4729 km 2 ), 2011 (4279 km 2 ) and 2017 (3859 km 2 ). Conversely, Landsat 30 m burned area data (Fig. 2 B) showed an increase from 1914 km 2 in 2001 to 2250 km 2 in 2021, but the trend was not statistically significant (p = 0.54) at a 90% confidence level. The largest mean burned area of about 3400 km 2 occurred in 2002, whereas 2004 recorded the smallest mean burned area of about 200 km 2 . MODIS burned area disaggregated by regions of Northern Ghana revealed contrasting temporal trends (Fig. 3 ). Notably, there were statistically significant decreasing trends at a 95% confidence level for the Savannah Region (p = 0.01), the Upper East Region (p = 0.02), and the Upper West Region (p = 0.04). The North-East Region also showed a decrease in the burned area, but this trend was not statistically significant (p = 0.12) at a 90% confidence level. In contrast, the Northern Region had increased burned area, albeit insignificant (p = 0.26). For Landsat (Fig. 4 ), no significant trend was observed for the North-East (p = 0.19), Northern (p = 0.44), Savannah (p = 0.51), Upper East (p = 0.69), and the Upper West Region (p = 0.97). Regarding active fires, the average number of MODIS active fires increased from 123 in 2001 to 128 in 2022 (Fig. 2 C). Active fires peaked in 2005 (254), 2011 (274), 2017 (251), and 2020 (244). However, no significant trend (p = 0.73) was observed at a 90% confidence level. When analysing the trends within the individual regions of Northern Ghana (Fig. 5 ), the Northern Region exhibited a significant upward trend (p = 0.031) at a 95% confidence level, while the trends in other regions were statistically insignificant. The fire season spanned from October to April with peak fires mostly occurring in December, however, the seasonal pattern varied between years (Fig. 6 A). Further analysis revealed a shift in the timing of the fire season, with a significant declining trend in active fires for November (p = 0.038) (Fig. 7 A), no significant trend for December (p = 0.67) (Fig. 7 B), and a significant increasing trend for January (p = 0.0018) (Fig. 7 C) and February (p = 0.028) (Fig. 7 D) at a 95% confidence level. Rainfall in Northern Ghana between 2001 and 2022 showed considerable variability between years, with no significant trend (p = 0.34) (Fig. 2 D). Notable peaks in mean annual rainfall occurred in 2003 (11575 mm), 2010 (11771 mm), and 2018 (11475 mm), while significant dips occurred in 2013 (7972 mm) and 2017 (8544 mm). The onset of rainfall typically starts in April and ends in October with shifts in the peak rainfall months from year to year (Fig. 6 B). However, further analysis of the mean rainfall trends for all months (Fig. 1 in the supplementary document) revealed no significant trends in rainfall for any month. Rainfall in Northern Ghana showed no significant correlation with Landsat Burned Areas (r = -0.06, p = 0.80) (Fig. 2 in the supplementary document). However, a statistically significant but weak negative correlation was observed between rainfall and MODIS burned areas (r = -0.44) at a 95% confidence level (Fig. 3 . in the supplementary document). LULC data for 2000 and 2020 revealed a cropland increase from about 13,000 km 2 to 19,000 km 2 (Fig. 8 ). Similarly, built-up areas experienced expansion from 789 km 2 in 2000 to 1,999 km 2 in 2020. Conversely, there was a decrease in area for dense short vegetation and tree cover over the same period. Institutional perspectives on changing fire regimes Initial responses from interviews revealed varying perceptions about the changing fire regimes in Northern Ghana. However, in a subsequent stakeholder engagement workshop, participants perceived a decline in large fires (locally referred to as bushfires) in the area. They attributed this decline to several interventions, including intensified public sensitisations and government flagship programmes such as the Sustainable Land and Water Management Project (SLWMP), Planting for Food and Jobs (PFJ), Ghana Shea Landscape Emission Reduction Project (GSLERP), Ghana Land Restoration and Small-Scale Mining Project (GLRSSMP). Additionally, the establishment of fire posts at the various Metropolitan, Municipal and District Assemblies, as well as the provision of alternative mechanisms for land preparations using tractors and herbicides have contributed to managing fires. These perceptions aligned with the observed MODIS burned area trends for Northern Ghana. For instance, an interview with a project manager at a conservation NGO highlighted that: Generally, bush burning increased in the landscape within the period and peaked during the logging and charcoal production activities. Burning is used to make the landscape more open for undertaking these activities. Also, with the increasing and expanding agricultural activities, more fires were being used to clear and prepare lands. However, this has decreased a bit within the last five years due to national and local bans on logging (2012–2019) and charcoal burning (2017–2021), respectively. Additionally, within this period, the presence of large-scale multinational programmes such as the SLWMP in 2011 and the Landscape and Environmental Agility Across Nation (LEAN) in 2020 has provided financial support for fire management and restoration efforts in Northern Ghana. (Project Manager, A Rocha Northern Ghana Sector, 2023) In terms of region-specific trends for MODIS burned areas, participants perceived the observed increasing trend in the Northern Region as due to lower population density coupled with increased hunting activities and the lack of wildlife-protected areas or community resource management areas (CREMAs). For example, the Environmental Specialist for the Savannah Agriculture Value Chain Development Project (SADEP) noted that: During the dry season in the Northern Region, you will find a lot of young men in about 3–5 trucks going hunting. They target a large area for burning to hunt for game. (Environmental Specialist- SADEP, 2024) Conversely, participants attributed the lack of significant trends in smaller fires (Landsat burned area) in Northern Ghana to livelihood-related activities such as hunting, charcoal burning and using fires to clear new farmlands. These activities are particularly prevalent in the North-East, Northern, Savannah and Upper West Regions. The low levels of burned areas in the Upper East Region were attributed to low fuel availability. This scarcity is largely attributed to the region’s geology, which is predominantly rocky. Additionally, the Upper East Region has seen an increase in built-up areas and a decrease in tree cover due to charcoal production. In this region, many farmers also clear their fields by hoes (rather than using fire) so they can use the resulting crop residues as fodder for their livestock. For instance, a Lecturer at the University for Development Studies emphasised that: People are increasingly turning to charcoal production as an alternative livelihood, as farming has become less lucrative due to climate change. About 20 years ago, there was only one community along the route from Tamale to Accra where you could buy charcoal. Currently, numerous charcoal-selling spots exist, contributing to the prevalence of small fires. Moreover, existing studies indicate that increasing populations and built-up areas are limiting the areas available for large fires. As communities expand, they convert lands for settlements, which restricts the burning activities to small plots used for farming, ensuring that the fires do not spread uncontrollably. In the Upper East Region, for example, the land area is smaller and has a high population density. Therefore, there are not enough farms for burning activities, and there is less tree cover for charcoal production. (Lecturer, UDS) Discussion Spatio-temporal disparities in burned area and active fire trends in Northern Ghana’s savanna landscape Our results highlight the importance of scale and spatial resolution in fire regime analysis. MODIS and Landsat burned areas in Northern Ghana over the last 22 years revealed contrasting spatial and temporal trends. The MODIS dataset showed a decreasing trend in burned area, while the Landsat dataset showed no significant trend. Aggregating to the whole of Northern Ghana masked the contrasting trends occurring in different regions. The decreasing MODIS burned area trend (e.g., for Savannah, Upper East and Upper West Regions) aligns with previous research – also using MODIS – which shows both West Africa and global decreases in burned areas (Andela et al. 2017 , Dwomoh and Wimberly 2017 ). The coarse resolution (500 m) of MODIS hampers its ability to detect small fires (≤ 25 ha) in heterogeneous landscapes(Laris 2005 , Roteta et al. 2019 ) which are commonly found in Northern Ghana, suggesting that the decreasing trend refers to larger fires. Conversely, the lack of a significant trend in Landsat burned areas for Northern Ghana and the different regions suggests a more complex picture. Taking both datasets into account, this indicates that large-scale fires may have been decreasing across Northern Ghana as a whole, while small fires showed no such decline. However, it is important to acknowledge a key limitation. The trends we described are based on 22-year satellite data, constrained by MODIS and Landsat data availability. As such, the analysis captures only a relatively short window in the region's long-term trajectory of fire regimes. This makes it difficult to assess whether the observed pattern—particularly the decrease in large fires—represents sustained shifts or is part of a broader, more cyclical trend. However, historical records discussed in the study area section reveal significant shifts in fire governance—from colonial-era suppression policies to more adaptive approaches that recognised early burning practices. In post-independence, the country witnessed a return to restrictive fire policies because of the government's interest in protecting its important projects and the Sahelian drought in the 1980s that triggered bushfire events in the country (Wardell et al. 2004 ). These governance changes likely influenced fire use patterns before satellite data became available. While we cannot quantify these historical dynamics using SRS, they provide important context for interpreting the observed trends and suggest that contemporary patterns may be shaped by deeper, policy-driven legacies of fire management (Alare et al. n.d.). To investigate the factors influencing the observed year-to-year variability in the burned areas of Northern Ghana for both MODIS and Landsat, we conducted a correlation analysis to examine the relationship between rainfall and the burned areas. The results revealed a weak but significant negative correlation between rainfall and MODIS burned area, suggesting that rainfall inhibited burning. This is consistent with the findings ofN’Datchoh et al. ( 2015 ) who revealed that a wet rainfall season leads to high humidity, suppressing fire spread. As a result, no fires were observed during the rainy season (April-October), but in the dry season (November to March). Conversely, there was no significant relationship between rainfall and the Landsat burned area. This suggests that anthropogenic activities, primarily livelihood-driven, mostly control the year-to-year variability of small fires. For instance, participants from the stakeholder engagement workshop attributed the increasing number of small fires in Northern Ghana to hunting, farm preparations and charcoal production activities. Hunting and charcoal production activities provide alternate livelihoods for communities, helping to diversify income sources amid growing food insecurity caused by poverty and vulnerability to climate change impacts (Yaro and Hesselberg 2010, Bonye et al. 2023 ). This aligns with the findings ofArchibald et al. ( 2010 ) andN’Datchoh et al. ( 2015 ) who contend that human influence on fire regimes in African savannas is more influential than the effect of climate in driving variation between years. However, when these activities coincide with extreme weather conditions like the hot and strong windy conditions in the dry season where fuels are sufficiently dried with no fire breaks or belts, the latter can exacerbate large fire risk in these areas. Participants also emphasised that recent national bans on charcoal production and logging have reduced the occurrence of large fires. However, communities tend to revert to these practices when enforcement decreases. Social drivers of region-specific shifting fire regimes In regions (North-East, Savannah, Upper East and West Regions) where large fires were decreasing based on MODIS burned area and active fire data, the workshop participants credited the decrease to interventions such as intensified public sensitisations, and government and multilateral flagship programmes such as the SLWMP, PFJ, GSLERP, LEAN and GLRSSMP. These multilateral projects have implemented tree planting and protection projects in selected Districts and communities in Northern Ghana. A key aspect of these initiatives is the incorporation of fire management plans to suppress burning that could potentially destroy investments in these trees. The fire management plans typically include bushfire sensitisation, the establishment of community fire volunteer groups, and the provision of incentives to discourage landscape burning. Despite the win-win discourses that dominate tree planting initiatives, they can potentially promote land grabbing, constraining local people’s access to resources and other resource-dependent livelihoods (Benjaminsen et al. 2011 , Leach and Scoones 2015, Kandel et al. 2021 ). It was, therefore, not surprising to learn from institutional interviews that some tree planting projects in the Savannah Region of Northern Ghana were destroyed by arson. This occurred particularly in communities that lacked sufficient lands for farming. Moreover, extensive tree planting projects along with fire suppression policies in these areas can also promote fuel build-up for wildfire risk in the event of drought (Veldman et al. 2019 ). Some workshop participants highlighted the role of population density and its knock-on effects on the establishment of fire posts and intensive agricultural activities (including the use of tractors and herbicides to clear land) in determining burning patterns. The number of people per square kilometre varies greatly between regions: Upper East (147.2), Northern (87.1), North-East (72.6), Upper West (49.0) and Savannah (18.7) (GSS 2021). The most densely populated Upper East Region has the lowest amount of MODIS burned area which also shows a significantly decreasing trend. This is consistent with previous studies that suggest high population densities are linked to smaller fires, as it limits the amount of available fuel for combustion and spread (Archibald and Roy 2009 , Bistinas et al. 2013 ). Conversely, the very low population density and associated lack of intensive agricultural activity of the Savannah Region may contribute to its high count of active fires and large amount of burned area for both MODIS and Landsat datasets. Field observations by one of the authors in the Savannah Region revealed that the observed high counts of active fires and extensively burned areas were associated with burning activities around the Mole National Park (MNP) to enhance livelihoods. The MNP also extends to the Northern and Upper West Regions. The predominant livelihoods in these regions revolved around charcoal production, hunting and clearing new lands for yam ( Dioscorea alata ) cultivation, which all required fire use. These activities typically involve small fires, which can grow into large fires if not properly controlled, especially during the windy late dry season. Moreover, prescribed burning takes place in the park to regenerate grasses for grazing animals. There has also been an influx of migrant herders from neighbouring countries who bring their cattle to graze in the Savannah, Northern and Upper West Regions. These herders set fires to resprout grasses for grazing during the dry season (Kpienbaareh and Luginaah 2019 , Amoako and Gambiza 2020). For Landsat burned area, participants attributed the lack of decline in smaller fires to livelihood-related activities such as hunting, herding, charcoal burning, and using fires to clear new farmlands. The term ‘smaller fires’ had various interpretations among the workshop participants. For example, the Ghana National Fire Service (GNFS) defined it specifically as fires at the ignition stage. Other participants perceived that fires could be small but intense and could spread and become large destructive fires. These perceptions about smaller fires could be attributed to the discursive framing of traditional burning practices as a ‘backward’ practice and a threat to sustainable land management (Amanor 2002 , Snyder et al. 2019). Thus, most participants at the workshop expressed a shared objective of suppressing all types of fires in Northern Ghana through the range of interventions indicated above. However, existing studies in Northern Ghana suggest that small fires, particularly early prescribed burns, can promote species diversity and reduce the stem density of woody species (Amoako et al. 2023 ). Therefore, the timing, frequency and extent of burning have more significant implications for land management (Laris 2002 ) than merely evaluating the total extent of the burned area. Fire and rainfall seasonality The seasonality of fires observed in this study aligned with the movement of the Inter-Tropical Convergence Zone (ITCZ), resulting in a dry season from November to March in Northern Ghana. Within this period, the annual dry dusty wind from the Sahel, locally referred to as the Harmattan, dries vegetation and increases the spread of fires influenced by anthropogenic activities (Kugbe et al. 2012 , Dahan et al. 2023 ). Our results revealed that fires peaked in December which is consistent with observations of Dahan et al. ( 2023 )d Datchoh et al. ( 2015 ). However, a detailed analysis revealed shifts in the fire season. Active fires showed a decreasing trend in November and an increasing trend for January and February. These shifts could not be attributed to shifts in rainfall patterns in Northern Ghana, which suggests that other human activities could be driving these changes. The observed shifts in the fire season imply a reduction in early burning and an increase in mid-dry season burning. These observations support findings by Laris et al. ( 2017 ) who identified that peak annual fires in West Africa occur in the mid-dry season. Field observation and interviews in the West Gonja Municipal and the North Gonja District of the Savannah Region revealed that the mid-dry season burning is associated with livelihood activities like land preparation for farming and hunting. Such mid-dry season fires have been identified as having no significant adverse effects on trees because trees shed their leaves around this period (N’Dri et al. 2018). Laris et al. ( 2017 ) contend that a more nuanced understanding of the effects of the different phases of burning or fire seasonality can only be achieved if studied within a broader context that takes into account multiple variables such as fuel type, fuel bed load, soil conditions, land use, time of the day, wind direction, humidity and temperature. Such studies in Northern Ghana could also enhance effective fire management. Currently, fire policies and projects in Ghana often advocate for early burning without clearly establishing what early burning entails. This has led to various interpretations regarding the timing of early burns and the appropriate times to conduct them. Interviews with local communities in the Savannah Region suggested that a more tailored message around early burning focusing on the moisture content of the vegetation and weather conditions rather than the specific months would be more useful and in line with traditional community decision-making. Moreover, it is important to note that currently, Northern Ghana is experiencing shifting and erratic rainfall patterns, with the early onset occurring in the western areas and late onset in the eastern part of Northern Ghana (Atiah et al. 2021 ). This shift could also account for the difference in MODIS burned area and active fire trends for the specific regions. The changes in rainfall timing and distribution, landscape scale, land use land cover types (LULC), fire suppression interventions as well as the types and sizes of fires collectively shape fire regime dynamics in savanna landscapes (Laris 2005 , N’Datchoh et al. 2015 , Laris et al. 2018 ). However, the shift in rainfall patterns also has implications for various livelihoods dependent on fires in this area. Therefore, understanding shifts in fire regimes and their impacts on local communities is essential for developing effective strategies for fire management, land use planning and sustainable resource utilisation. Land cover change and changing fire regimes in Northern Ghana The findings from Landsat LULC suggested an increase in agricultural and built-up expansions in Northern Ghana. This increase aligned with the land use changes perceived by stakeholders. It also supports findings of existing studies in the Upper Guinean Region of West Africa and Africa which reported increases in agricultural lands (Andela et al. 2017 , Dwomoh and Wimberly 2017 ). However, while their findings attributed agricultural expansion to decreasing burned area in these places, our study adds nuance to the explanation. Drawing on the finer-resolution Landsat data, our research shows that the observed agriculture expansion and built-up areas are associated with a decrease in larger fires and no change in smaller fires in Northern Ghana. As indicated in earlier discussions, fire is used in clearing new lands for farming, contributing to cropland expansion in these areas. Conclusion This study draws on evidence from remote sensing and institutional interviews to assess changing fire regimes in Northern Ghana. When considering Northern Ghana as a whole, Landsat and MODIS burned areas depicted different fire trends and types; MODIS showed a decrease in large fires, whereas Landsat revealed no change in small fires. Despite numerous project interventions on fires, there is little agreement on how they are defined. This lack of agreement overlooks the various types of fires in the savanna landscape and characterises them solely as destructive fires. Therefore, local-level decision-makers may need to determine the types of fires they are interested in monitoring and consistently evaluate them using the most appropriate method. MODIS satellite data would be suitable for monitoring large fires, while on-the-ground reporting or the use of alternative fine spatial resolution datasets such as Landsat or Sentinel-2 satellite data would be more appropriate for small fires. Region-specific trend analysis indicates that the Savannah Region stands out as the region with the most extensive burned area and higher active fire counts. Our results highlight the importance of scale in fire regime analysis since aggregating to the whole of Northern Ghana masked the varying trends in different regions. Thus, we argue that disaggregation of data is important to ensure fire interventions are cost-effective by responding to region-specific fire trends. The study also highlighted shifts in fire seasonality in Northern Ghana. It showed a decrease in early burning, and an increase in fire activities in the mid-dry season, specifically in January and February. This has important implications for fire management in Northern Ghana, as policies and projects tend to promote early burning without explicitly defining what early burning entails and rely on November and December as months for early burning. This suggests that current fire management strategies may not be aligned with the actual fire dynamics of Northern Ghana. Therefore, there is a need for more adaptable and responsive fire management practices tailored to the changing fire regimes in this area. Additionally, changes in fire regimes are not only impacted by biophysical factors but also mediated through political and socio-cultural factors. Insights from institutional interviews and the stakeholder engagement workshop indicated that institutions implemented measures to suppress fires in their project communities. This may have contributed to the reduction of large fires. To determine the effectiveness of these interventions, institutions need to consider the perspective of local communities. While we recommend upscaling these interventions to other communities, it is also important to recognise that fire shapes and maintains the composition and the functioning of savanna ecosystem services. Therefore, to ensure the sustainability of such projects, and reduce the risk of later devastating fires, it is important to establish appropriate land use systems and acknowledge the significance of prescribed and controlled burning practices in savanna ecosystems. Declarations Acknowledgement The authors thank the London Interdisciplinary Social Science Doctoral Training Programme (LISS-DTP) for funding this work. The authors are also grateful to Henry Thompson, Davide Lomeo, and Aidan Byrne for assisting in coding the remote sensing analysis. Special thanks also to all individuals from the various institutions in Ghana who participated in the study. Author’s contributions RSA designed the research, collected and analysed the data, and drafted the paper. Both KS and ET reviewed the methodology and the results and contributed to the editing of the paper. Funding This research work is funded by the Economic Social and Research Council (ESRC) London Interdisciplinary Social Science Doctoral Training Programme (LISS-DTP), Grant Ref: ES/P000703/1. Availability of data and materials The datasets used to estimate burned area, active fires, rainfall, land use and land cover are openly accessible and free to use as detailed in the methods section. However, the datasets developed and used in this manuscript can be obtained from the authors upon request. Ethics approval and consent to participate The studies involving human participants were reviewed and approved by King’s College London Research Ethics Office (Ethical clearance reference number: MRSP-21/22-30162). Competing interests The authors declare that they have no competing interests. References Alare, Rahinatu S., Emma Tebbs, and Kate Schreckenberg. n.d. Navigating Discourses: Unpacking the Problematisation of Traditional Burning Practices in Northern Ghana. In Preparation [Unpublished] . Amanor, K. S. 2002. Bushfire Management, Culture and Ecological Modernisation in Ghana . vol. 33 United Kingdom: Brighton. Amoako, Esther Ekua, James Gambiza. 2019. Effects of Anthropogenic Fires on Soil Properties and the Implications of Fire Frequency for the Guinea Savanna Ecological Zone, Ghana. Scientific African 6. https://doi.org/10.1016/j.sciaf.2019.e00201 Amoako, Esther Ekua, James Gambiza. 2020. Traditional Practices, Knowledge and Perceptions of Fire Use in a West African Savanna Parkland. BioRxiv 60:53–77. https://doi.org/10.35979/alj.2020.02.60.53 Amoako, Esther, Hamza Ekua, and Issifu, Rikiatu Husseini. 2023. The Effects of Prescribed Dry Season Burning on Woody Species Composition, Mole National Park, Ghana. Tropical Conservation Science 16:1–15. https://doi.org/10.1177/19400829231164936 Ampadu-Agyei, Okyeame. 1988. Bushfires and Management Policies in Ghana. The Environmentalist 8(3):221–228. https://doi.org/10.1007/BF02240254 Andela, N., D. C. Morton, L. Giglio, Y. Chen, G. R. Van Der Werf, P. S. Kasibhatla, and R. S. DeFries et al. 2017. A Human-Driven Decline in Global Burned Area. Science 356(6345):1356–1362. https://doi.org/10.1126/science.aal4108 Andrew Wardell, D., and Christian Lund. 2006. Governing Access to Forests in Northern Ghana: Micro-Politics and the Rents of Non-Enforcement. World Development 34(11):1887–1906. https://doi.org/10.1016/j.worlddev.2005.11.021 Apusigah, Agnes Atia. 2007. Promoting Sustainable Wildfire Management in Northern Ghana: Learning from History. European Journal of Social Sciences 5(1):61–76. Archibald, S., A. Nickless, N. Govender, R. J. Scholes, and V. Lehsten. 2010. Global Ecology and Biogeography 19:794–809. https://doi.org/10.1111/j.1466-8238.2010.00568.x . Climate and the Inter-Annual Variability of Fire in Southern Africa: A Meta-Analysis Using Long-Term Field Data and Satellite-Derived Burnt Area Data. Archibald, S., and D. P. Roy. 2009. Identifying Individual Fires from Satellite-Derived Burned Area Data. International Geoscience and Remote Sensing Symposium (IGARSS) 3:160–63. https://doi.org/10.1109/IGARSS.2009.5417974 Atiah, Winifred, Francis K. Ayinpogbilla, Bekele Muthoni, Fred Kotu, Kizito, and Leonard K. Amekudzi. 2021. Trends of Rainfall Onset, Cessation, and Length of Growing Season in Northern Ghana: Comparing the Rain Gauge, Satellite, and Farmer’s Perceptions. Atmosphere 12(12). https://doi.org/10.3390/atmos12121674 Benjaminsen, Tor A., Ian Bryceson, and Faustin Maganga. 2011. and Tonje Refseth. Conservation and Land Grabbing in Tanzania Global Land Grabbing. In International Conference on Global Land Grabbing . Bistinas, Ioannis, Duarte Oom, C. L. Ana, P. Sá, Sandy, I. Harrison, Colin Prentice, M. C. José, and Pereira. 2013. Relationships between Human Population Density and Burned Area at Continental and Global Scales. Plos One 8(12):1–12. https://doi.org/10.1371/journal.pone.0081188 Bonye, Samuel, Gordon Yenglier Ziem, and Yiridomoh, Vivian Nsiah. 2023. Our Forest, Our Livelihood: Natural Resources’ Use Controversies and Community Livelihood Sustainability in the Mole National Park, Ghana. Land Use Policy 127. https://doi.org/10.1016/j.landusepol.2023.106589 Boschetti, Luigi, Stéphane P. Flasse, and Pietro A. Brivio. 2004. Analysis of the Conflict between Omission and Commission in Low Spatial Resolution Dichotomic Thematic Products: The Pareto Boundary. Remote Sensing of Environment 91(3–4):280–292. https://doi.org/10.1016/j.rse.2004.02.015 Bowman, David M.J.S., K. Jennifer, Paulo Balch, William J. Artaxo, Jean M. Bond, Mark A. Carlson, M. Cochrane, Carla, and D’Antonio et al. 2009. Fire in the Earth System. Science 324(5926):481–484. https://doi.org/10.1126/science.1163886 Carmenta, Rachel, and Emilie Coudel, Angela M. Steward. 2019. Forbidden Fire: Does Criminalising Fire Hinder Conservation Efforts in Swidden Landscapes of the Brazilian Amazon? Geographical Journal 185(1):23–37. https://doi.org/10.1111/geoj.12255 Chuvieco, Emilio, Florent Mouillot, R. Guido, and van der Werf et al. 2019. Jesús San Miguel, Mihai Tanasse, Nikos Koutsias, Mariano García,. Historical Background and Current Developments for Mapping Burned Area from Satellite Earth Observation. Remote Sensing of Environment 225 (February): 45–64. https://doi.org/10.1016/j.rse.2019.02.013 Dahan, Kueshi, Raymond Abudu Sémanou, and Kasei, Rikiatu Husseini. 2023. Contribution of Remote Sensing to Wildfire Trend and Dynamic Analysis in Two of Ghana’s Ecological Zones: Guinea-Savanna and Forest-Savanna Mosaic. Fire Ecology 19(1). https://doi.org/10.1186/s42408-023-00198-z Dennis, Rona A., Judith Mayer, Grahame Applegate, Unna Chokkalingam, Carol J.Pierce Colfer, Iwan Kurniawan, and Henry Lachowski et al. 2005. Fire, People and Pixels: Linking Social Science and Remote Sensing to Understand Underlying Causes and Impacts of Fires in Indonesia. Human Ecology 33(4):465–504. https://doi.org/10.1007/s10745-005-5156-z Doe, Sylvanus Sefenu Per. 2008. Fire Management Policy and Environmental NGO Movement: Critical Contemporary Perspectives, Indicators and Prospects in Savannas, Ghana. GeoJournal 71 (2–3): 159–70. https://doi.org/10.1007/s10708-008-9153-9 Dwomoh, Francis K., and Michael C. Wimberly. 2017. Fire Regimes and Their Drivers in the Upper Guinean Region of West Africa. Remote Sensing 9(11). https://doi.org/10.3390/rs9111117 Frey, L., C. H. Botan, and G. Kreps. 2000. Investigating Communication: An Introduction to Research Methods . Allyn and Bacon. Funk, Chris, Pete Peterson, Martin Landsfeld, Diego Pedreros, James Verdin, Shraddhanand Shukla, and Gregory Husak et al. 2015. The Climate Hazards Infrared Precipitation with Stations - A New Environmental Record for Monitoring Extremes. Scientific Data 2:1–21. https://doi.org/10.1038/sdata.2015.66 Fusco, Emily J., John T. Finn, John T. Abatzoglou, Jennifer K. Balch, Sepideh Dadashi, and Bethany A. Bradley. 2019. Detection Rates and Biases of Fire Observations from MODIS and Agency Reports in the Conterminous United States. Remote Sensing of Environment 220(January):30–40. https://doi.org/10.1016/j.rse.2018.10.028 Giglio, L., J. T. Randerson, G. R. Van Der, P. S. Werf, G. J. Kasibhatla, D. C. Collatz, R. S. Morton, and Defries et al. 2009. Assessing Variability and Long-Term Trends in Burned Area by Merging Multiple Satellite Fire Products. Biogeosciences Discussions 6(6):11577. https://doi.org/10.5194/bgd-6-11577-2009 Giglio, L., W. Schroeder, J. Hall, and C. Justice. 2021. MODIS Collection 6 and Collection 6.1 Active Fire Product User’s Guide. Washington D.C., USA. Glaser, B. G., and A. L. Strauss. 1967. The Discovery of Grounded Theory: Strategies for Qualitative Research . New Brunswick, USA and London, UK: Aldine Transaction. GSS. 2021. Ghana 2021 Population and Housing Census : Preliminary Report. Accra, Ghana. Hantson, Stijn, Gitta Lasslop, Silvia Kloster, and Emilio Chuvieco. 2015. Anthropogenic Effects on Global Mean Fire Size. International Journal of Wildland Fire 24(5):589–596. https://doi.org/10.1071/WF14208 Harwell, Emily E. 2000. Remote Sensibilities: Discourses of Technology and the Making of Indonesia’s Natural Disasters. Development and Change 31(1):307–340. https://doi.org/10.1111/1467-7660.00156 Jerven, Morten. 2016. Benefits and Costs of the Data for Development Targets for the Post-2015 Development Agenda. United Nations Working Paper. Kandel, Matt, Genevieve Agaba, Rahinatu S. Alare, and Thomas Addoah, Kate Schreckenberg. 2021. Assessing Social Equity in Farmer-Managed Natural Regeneration (FMNR) Interventions: Findings from Ghana. Ecological Restoration 39(1–2):64–76. https://doi.org/10.3368/er.39.1-2.64 Kerner, Hannah, Catherine Nakalembe, Adam Yang, Ivan Zvonkov, Ryan McWeeny, and Gabriel Tseng, Inbal Becker-Reshef. 2024. How Accurate Are Existing Land Cover Maps for Agriculture in Sub-Saharan Africa? Scientific Data 11(1). https://doi.org/10.1038/s41597-024-03306-z Kleemann, Janina, Gülendam Baysal, N. N. Henry, Bulley, and Christine Fürst. 2017. Assessing Driving Forces of Land Use and Land Cover Change by a Mixed-Method Approach in North-Eastern Ghana, West Africa. Journal of Environmental Management 196:411–442. https://doi.org/10.1016/j.jenvman.2017.01.053 Kpienbaareh, D., and I. Luginaah. 2019. After the Flames Then What? Exploring the Linkages between Wildfires and Household Food Security in the Northern Savannah of Ghana. International Journal of Sustainable Development and World Ecology 26(7):612–624. https://doi.org/10.1080/13504509.2019.1640311 Kugbe, Joseph X., Fosu Mathias, Tamene L. Desta, Manfred Denich, L. G. Paul, and Vlek. 2012. Annual Vegetation Burns across the Northern Savanna Region of Ghana: Period of Occurrence, Area Burns, Nutrient Losses and Emissions. Nutrient Cycling in Agroecosystems . https://doi.org/10.1007/s10705-012-9514-0 Kull, Christian A. 2009. and Paul Laris. Fire Ecology and Fire Politics in Mali and Madagascar. In Tropical Fire Ecology , edited by M. A Cochrane, 171–226. New York: Springer. https://doi.org/10.1007/978-3-540-77381-8_7 Laris, P. 2002. Burning the Seasonal Mosaic: Preventative Burning Strategies in the Wooded Savanna of Southern Mali. Human Ecology 30(2):155–186. https://doi.org/10.1023/A:1015685529180 Laris, P., S. Dadashi, A. Jo, and S. Wechsler. 2016. Buffering the Savanna: Fire Regimes and Disequilibrium Ecology in West Africa. Plant Ecology 217(5):583–596. https://doi.org/10.1007/s11258-016-0602-0 Laris, Paul. 2011. Humanizing Savanna Biogeography: Linking Human Practices with Ecological Patterns in a Frequently Burned Savanna of Southern Mali. Annals of the Association of American Geographers 101(5):1067–1088. https://doi.org/10.1080/00045608.2011.560063 Laris, Paul, Aurahm Jo, and Suzanne P. Wechsler. 2018. Effects of Landscape Pattern and Vegetation Type on the Fire Regime of a Mesic Savanna in Mali. Journal of Environmental Management 227(March):134–145. https://doi.org/10.1016/j.jenvman.2018.08.091 Laris, Paul, Moussa Koné, and Sepideh Dadashi, Fadiala Dembele. 2017. The Early/Late Fire Dichotomy: Time for a Reassessment of Aubréville’s Savanna Fire Experiments. Progress in Physical Geography 41(1):68–94. https://doi.org/10.1177/0309133316665570 Laris, Paul S. 2005. Spatiotemporal Problems with Detecting and Mapping Mosaic Fire Regimes with Coarse-Resolution Satellite Data in Savanna Environments. Remote Sensing of Environment 99(4):412–424. https://doi.org/10.1016/j.rse.2005.09.012 Laris, Paul, David Andrew Wardell. 2006. Good, Bad or ‘Necessary Evil’? Reinterpreting the Colonial Burning Experiment …. The Geographical Journal 172(4):271–290. Leach, Melissa, Ian Scoones. 2015. Political Ecologies of Carbon in Africa. In Carbon Conflicts and Forest Landscapes in Africa , Routledge. https://doi.org/10.4324/9781315740416 Long, Tengfei, Zhaoming Zhang, Guojin He, Weili Jiao, Chao Tang, Bingfang Wu, Xiaomei Zhang, Guizhou Wang, and Ranyu Yin. 2019. 30m Resolution Global Annual Burned Area Mapping Based on Landsat Images and Google Earth Engine. Remote Sensing 11(5):1–30. https://doi.org/10.3390/rs11050489 McCaffrey, Sarah, Eric Toman, and Melanie Stidham, Bruce Shindler. 2013. Social Science Research Related to Wildfire Management: An Overview of Recent Findings and Future Research Needs. International Journal of Wildland Fire 22:15–24. https://doi.org/http://dx.doi.org/10.1071/WF11115 Mendes, Mehmet, Akin Pala. 2003. Type I Error Rate and Power of Three Normality Tests. Information Technology Journal 2(2):135–139. https://doi.org/10.3923/itj.2003.135.139 Mistry, Jayalaxshmi, Andrea Berardi. 2016. Bridging Indigenous and Scientific Knowledge. Science 352(6291):1274–1275. https://doi.org/10.1126/science.aaf1160 MLFM. 2006. National Wildfire Management Policy. Accra, Ghana. Mohd Razali, and Nornadiah, Yap Bee Wah. 2011. Power Comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and Anderson-Darling Tests. Journal of Statistical Modeling and Analytics 2(1):13–14. Moura, Livia C., O. Aldicir, Isabel B. Scariot, Robin Schmidt, and Beatty. 2019. and Jeremy Russell-Smith. The Legacy of Colonial Fire Management Policies on Traditional Livelihoods and Ecological Sustainability in Savannas: Impacts, Consequences, New Directions. Journal of Environmental Management 232 (December 2018): 600–606. https://doi.org/10.1016/j.jenvman.2018.11.057 N’Datchoh, E. T., A. Konaré, A. Diedhiou, A. Diawara, E. Quansah, and P. Assamoi. 2015. Effects of Climate Variability on Savannah Fire Regimes in West Africa. Earth System Dynamics 6(1):161–174. https://doi.org/10.5194/esd-6-161-2015 N’Dri, Aya, Tionhonkélé Drissa Brigitte, Jacques Soro, Kanvaly Gignoux, and Dosso. Mouhamadou Koné, Julien Kouadio N’Dri, N’golo Abdoulaye Koné, and Sébastien Barot. 2018. Season Affects Fire Behavior in Annually Burned Humid Savanna of West Africa. Fire Ecology 14 (2). https://doi.org/10.1186/s42408-018-0005-9 Potapov, Peter, Matthew C. Hansen, Indrani Kommareddy, Anil Kommareddy, Svetlana Turubanova, Amy Pickens, Bernard Adusei, and Alexandra Tyukavina, Qing Ying. 2020. Remote Sensing 12(3). https://doi.org/10.3390/rs12030426 . Landsat Analysis Ready Data for Global Land Cover and Land Cover Change Mapping. Randerson, J. T., Y. Chen, G. R. Van Der Werf, B. M. Rogers, and D. C. Morton. 2012. Global Burned Area and Biomass Burning Emissions from Small Fires. Journal of Geophysical Research: Biogeosciences 117(4):1–23. https://doi.org/10.1029/2012JG002128 Roteta, E., A. Bastarrika, M. Padilla, T. Storm, and E. Chuvieco. 2019. Development of a Sentinel-2 Burned Area Algorithm: Generation of a Small Fire Database for Sub-Saharan Africa. Remote Sensing of Environment 222 (December 2018): 1–17. https://doi.org/10.1016/j.rse.2018.12.011 Silva, João M.N., C. L. Ana, Sá, M. C. José, and Pereira. 2005. Comparison of Burned Area Estimates Derived from SPOT-VEGETATION and Landsat ETM + Data in Africa: Influence of Spatial Pattern and Vegetation Type. Remote Sensing of Environment 96(2):188–201. https://doi.org/10.1016/j.rse.2005.02.004 Smith, C., J. Ainscough, R. S. Alare, A. R. Croker, K. M. De Freitas, James D.A. Millington, and J. Mistry et al. 2024. How Policy Interventions Influence Burning to Meet Cultural and Small-Scale. Ecology and Society 29(1). https://doi.org/https://doi.org/10.5751/ES-14850-290135 Snyder, Katherine A., and Beth Cullen, Juliet Braslow. 2019. Farmers as Experts: Interpreting the ‘Hidden’ Messages of Participatory Video across African Contexts. Area 51(4):779–787. https://doi.org/10.1111/area.12538 Soest, Christian Von. 2023. Why Do We Speak to Experts? Reviving the Strength of the Expert Interview Method. Perspectives on Politics 21(1):277–287. https://doi.org/10.1017/S1537592722001116 Steen-Adams, Michelle M., and Susan Charnley, Mark D. Adams. 2017. Historical Perspective on the Influence of Wildfire Policy, Law, and Informal Institutions on Management and Forest Resilience in a Multiownership, Frequent-Fire, Coupled Human and Natural System in Oregon, USA. Ecology and Society 22(3). https://doi.org/10.5751/ES-09399-220323 Szpakowski, D. M., and J. L. R. Jensen. 2019. A Review of the Applications of Remote Sensing in Fire Ecology. Remote Sensing 11(2638):1–31. https://doi.org/10.2307/4445278 Veldman, Joseph W., J. C. Aleman, S. T. Alvarado, T. M. Anderson, S. Archibald, W. J. Bond, and T. W. Boutton, N Buchmann. 2019. Comment on ‘The Global Tree Restoration Potential’. Science 366:1–4. https://doi.org/10.1126/science.aaz0111 Wardell, D., Thomas Theis Andrew, Kjeld Nielsen, and Rasmussen. 2004. and Cheikh Mbow. Fire History, Fire Regimes and Fire Management in West Africa: An Overview. In Wildland Fire Management Handbook for Sub-Sahara Africa , edited by Johann G. Goldammer and Cornelis de Ronde, 350–81. Global Fire Monitoring Center (GFMC). Werf, G. R., James T. Van Der, Louis Randerson, T. Giglio, Thijs, Yang Van Leeuwen, Brendan M. Chen, Mingquan Rogers, and Mu et al. 2017. Global Fire Emissions Estimates during 1997–2016. Earth System Science Data 9(2):697–720. https://doi.org/10.5194/essd-9-697-2017 Wilgen, B. W., N. Van, H. C. Govender, D. Biggs, Ntsala, and X. N. Funda. 2004. Response of Savanna Fire Regimes to Changing Fire-Management Policies in a Large African National Park. Conservation Biology 18(6):1533–1540. https://doi.org/10.1111/j.1523-1739.2004.00362.x Wooster, Martin J., J. Gareth, Louis Roberts, David Giglio, Patrick Roy, Luigi Freeborn, Christ Boschetti, and Justice et al. 2021. Satellite Remote Sensing of Active Fires: History and Current Status, Applications and Future Requirements. Remote Sensing of Environment 267(May):112694. https://doi.org/10.1016/j.rse.2021.112694 Yaro, J. A., J Hesselberg. 2010. The Contours of Poverty in Northern Ghana: Policy Implications for Combating Food Insecurity. Research Review of the Institute of African Studies 26(1). https://doi.org/10.4314/rrias.v26i1.56957 Additional Declarations No competing interests reported. Supplementary Files SupplementarydocumentRevised.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 27 Mar, 2026 Reviews received at journal 25 Feb, 2026 Reviews received at journal 22 Feb, 2026 Reviewers agreed at journal 20 Feb, 2026 Reviewers agreed at journal 10 Feb, 2026 Reviewers agreed at journal 10 Feb, 2026 Reviewers invited by journal 16 Jul, 2025 Editor assigned by journal 14 Apr, 2025 Submission checks completed at journal 14 Apr, 2025 First submitted to journal 13 Apr, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6441082","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":486568638,"identity":"48842b33-515f-4013-9bb6-febce1aed859","order_by":0,"name":"Rahinatu Sidiki Alare","email":"data:image/png;base64,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","orcid":"","institution":"King's College London","correspondingAuthor":true,"prefix":"","firstName":"Rahinatu","middleName":"Sidiki","lastName":"Alare","suffix":""},{"id":486568639,"identity":"a9a28124-201b-4c9a-9d33-7638ac593b86","order_by":1,"name":"Emma Tebbs","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Emma","middleName":"","lastName":"Tebbs","suffix":""},{"id":486568642,"identity":"fb8e03b3-f733-4db3-a45e-c0c9ea3d1587","order_by":2,"name":"Kate Schreckenberg","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Kate","middleName":"","lastName":"Schreckenberg","suffix":""}],"badges":[],"createdAt":"2025-04-13 20:53:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6441082/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6441082/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87280380,"identity":"2189e2d8-20f6-4421-8d7e-8217aecf69e1","added_by":"auto","created_at":"2025-07-22 09:36:59","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":150120,"visible":true,"origin":"","legend":"\u003cp\u003eMap of the study area\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/0f55b37abd082e335fd4f38b.png"},{"id":87281394,"identity":"eed3d5fa-400e-4909-9606-d54d64e40bce","added_by":"auto","created_at":"2025-07-22 09:45:00","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":393738,"visible":true,"origin":"","legend":"\u003cp\u003eTime series of annual MODIS burned area (A), Landsat burned area (B), number of active fires (C), and rainfall (D) in Northern Ghana.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/162b723328c372b162c9fc8e.png"},{"id":87281774,"identity":"6738a870-96b0-49f5-8617-11d43976df25","added_by":"auto","created_at":"2025-07-22 09:53:00","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":204572,"visible":true,"origin":"","legend":"\u003cp\u003eSpatio-temporal trends of MODIS burned areas for each region of Northern Ghana, including the North-East Region (NE), Northern Region (NR), Savannah Region (SR), Upper East Region (UE) and Upper West Region (UW).\u003c/p\u003e","description":"","filename":"image3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/b4e76f9f63107e2e443c2903.jpeg"},{"id":87281391,"identity":"a76eccfc-0643-49a5-b80b-a848ea2a8968","added_by":"auto","created_at":"2025-07-22 09:45:00","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":312367,"visible":true,"origin":"","legend":"\u003cp\u003eTime series of Landsat burned areas for each region in Northern Ghana including the North-East (NE), Northern (NR), Savannah (SR), Upper East (UE), and Upper West (UW) regions. All trends were statistically insignificant for all regions.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/e6325988edaea66989aafaa7.png"},{"id":87281775,"identity":"aa18c349-6900-4ba7-80e2-026cfbbc24bd","added_by":"auto","created_at":"2025-07-22 09:53:00","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":274984,"visible":true,"origin":"","legend":"\u003cp\u003eTime series of MODIS active fires for each region in Northern Ghana including North-East (NE), Northern (NR), Savannah (SR), Upper East (UE), and Upper West (UW) regions. The trend was statistically significant for NR (p=0.031) which revealed an increasing trend. All other regions demonstrated a decreasing, but statistically insignificant trend.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/cea2642263676d1eba612158.png"},{"id":87280396,"identity":"e0417fd7-9ba0-4ba4-bcd1-7a99df9724f9","added_by":"auto","created_at":"2025-07-22 09:37:00","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":189911,"visible":true,"origin":"","legend":"\u003cp\u003eSeasonality of MODIS active fire (A) and rainfall (B) for Northern Ghana from 2001 to 2022\u003c/p\u003e","description":"","filename":"image6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/a95c9955ed771a9d368b3842.jpeg"},{"id":87281776,"identity":"faf7f491-6295-42b9-833d-137ceeafd406","added_by":"auto","created_at":"2025-07-22 09:53:00","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":203825,"visible":true,"origin":"","legend":"\u003cp\u003eMonthly trends of fire seasonality in Northern Ghana between 2000 and 2022. Fig. 7A, 7B, 7C and 7D depict monthly trends of active fires in November, December, January, and February between 2001 and 2022, respectively.\u003c/p\u003e","description":"","filename":"image7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/ad7447d3d1fa3d75837ad6af.jpeg"},{"id":87280385,"identity":"0bc42842-a221-4a33-95c7-01054ed0fa40","added_by":"auto","created_at":"2025-07-22 09:37:00","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":76397,"visible":true,"origin":"","legend":"\u003cp\u003eLandsat land use land cover analysis for Northern Ghana between 2000 (red) and 2020 (green).\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/228e13650fc14168a94de266.png"},{"id":87467089,"identity":"d9606c21-280d-4a23-8e6d-d30f334089a5","added_by":"auto","created_at":"2025-07-24 07:58:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2781262,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/fa494c10-f138-4d28-9f80-a72e28af8023.pdf"},{"id":87280382,"identity":"f5d0fab4-daa6-4889-84d3-621df1869ee7","added_by":"auto","created_at":"2025-07-22 09:37:00","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":447891,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementarydocumentRevised.docx","url":"https://assets-eu.researchsquare.com/files/rs-6441082/v1/72861873e16b2c810bdab693.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Diverging fire trends in Northern Ghana’s savanna landscape: Insights from remote sensing and institutional perspectives","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAfrican savannas are known for their high flammability, and fire plays a significant role in shaping the composition and structure of these ecosystems (Bowman et al. \u003cspan class=\"CitationRef\"\u003e2009\u003c/span\u003e, Laris et al. \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e). It is estimated that approximately 250\u0026nbsp;million hectares (Mha) of land are burned in Africa each year, accounting for an average of 70% of the global annual burned area (Giglio et al. \u003cspan class=\"CitationRef\"\u003e2009\u003c/span\u003e). These fires also contribute to land use modifications and greenhouse gas emissions (Hantson et al. \u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e, Van Der Werf et al. \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eIn the past, studies on fires were spatially limited to regional scales, but satellite remote sensing (SRS) now allows for extensive detection of burned and active fires, enabling analysis of conditions and monitoring of landscape changes (Chuvieco et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e, Szpakowski and Jensen \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e, Wooster et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). This study focuses on two key applications of SRS data to understand shifting fire regimes in Northern Ghana\u0026rsquo;s savanna landscape: active fire and burned area. Fire regimes in this context refer to the spatiotemporal trends in active fire, burned area and fire seasonality. Active fire involves real-time monitoring of fire through a technique that monitors the thermal contrast between burnings and the background, while burned area is a post-fire analysis that relies on the changes in the spectral signatures of fire-affected vegetation and soil in parts of the electromagnetic spectrum (Chuvieco et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e, Wooster et al. \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). Currently, there is an increase in the number of global active fires and burned area products from different satellite sensors, advancing studies on fire regimes, fire risk, climate change, and air quality studies (Szpakowski and Jensen \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eDespite this progress, existing global fire products have been based on coarse spatial resolution sensors (e.g., MODIS with a resolution of 250 m to 1 km) that contain a \u0026lsquo;low-resolution bias\u0026rsquo; where larger fires are detected while the smaller ones go undetected (Boschetti et al. \u003cspan class=\"CitationRef\"\u003e2004\u003c/span\u003e, Laris \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e, Laris et al. \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). Yet, these small fires account for 24% (0.78M km\u003csup\u003e2\u003c/sup\u003e/year) of African burned areas (Randerson et al. \u003cspan class=\"CitationRef\"\u003e2012\u003c/span\u003e). They are often used to achieve a range of livelihood activities such as agricultural land preparation, hunting, sprouting fresh grasses for grazing, clearing bushes and deterring wild reptiles from entering homes (Laris \u003cspan class=\"CitationRef\"\u003e2002\u003c/span\u003e, Amoako and Gambiza 2020). Sometimes, these fires can become uncontrollable and expand into large, destructive ones (Amoako and Gambiza 2020). Hence, both large and small fires are present in this landscape. However, the sole reliance on coarse resolution sensors by existing studies to analyse fire regimes and their drivers in Africa, including Ghana(Kugbe et al. \u003cspan class=\"CitationRef\"\u003e2012\u003c/span\u003e, Andela et al. \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e, Dwomoh and Wimberly \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e, Dahan et al. \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e) poses challenges in developing policies for effective fire management in these landscapes.Mistry and Berardi (2016) contend that such studies and the prevailing antifire policies undermine traditional fire knowledge and marginalise local resource users who depend on fire for their livelihoods. Additionally, proper data is crucial for creating tailored policies and interventions and assessing their effectiveness (Jerven \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eMeanwhile, interdisciplinary researchers have highlighted the limitations of SRS of fires in explaining the drivers of burning in African savanna landscapes (Harwell \u003cspan class=\"CitationRef\"\u003e2000\u003c/span\u003e, Dennis et al. \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e, Laris \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e). This perspective is valid because while biophysical factors account for fire ignition, fire management is inherently social (McCaffrey et al. \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e). For instance, the actual start of fires is influenced by biophysical factors such as weather conditions, fuel load, and ignition sources. However, managing and controlling fires is significantly influenced by human activities, including socio-economic factors, and policies. An important area of fire management research is the influence of institutions (McCaffrey et al. \u003cspan class=\"CitationRef\"\u003e2013\u003c/span\u003e, Steen-Adams et al. 2017). Institutions play a key role in influencing fire regimes through decision-making, governing land access, and implementing policies and interventions (Van Wilgen et al. \u003cspan class=\"CitationRef\"\u003e2004\u003c/span\u003e, Kull and Laris 2009, Carmenta et al. 2019, Moura et al. \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). These interventions include actions taken by state and non-state actors to achieve specific aims, including regulations, economic incentives, and interventions that elicit voluntary responses, such as communication campaigns (Smith et al. \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eGiven the complexities associated with establishing a direct correlation between the drivers and the observed spatiotemporal trends of fire regimes in African savanna landscapes using SRS (Laris \u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e), integrating institutional perspectives can enhance our understanding of the drivers in this area. Expert interviews from formal institutions provide professional knowledge and interpretation (Kleemann et al. \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e, Von Soest \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e) of fire regime drivers and spatiotemporal trends. Therefore, this study addresses these gaps by employing a mixed methods approach to understanding the types of fires and their drivers in Northern Ghana\u0026rsquo;s savanna landscape.\u003c/p\u003e\n\u003cp\u003eNorthern Ghana\u0026apos;s savanna landscape provides an important context for this study, as 90% of the total land cover experiences annual bush burning during the dry season (MLFM \u003cspan class=\"CitationRef\"\u003e2006\u003c/span\u003e), leading to the implementation of various policies and interventions to manage fires in these areas. The study focuses on analysing the spatiotemporal trends in fire regimes and institutional perspectives for a more comprehensive understanding of the drivers, given that the reasoning behind these regimes is often under-studied (Laris \u003cspan class=\"CitationRef\"\u003e2011\u003c/span\u003e). Specifically, we asked:\u003c/p\u003e\n\u003cp\u003e\u003cspan\u003ei. What are the spatiotemporal trends in burned areas, active fires, and fire seasonality over the last 21 years (2000\u0026ndash;2022) in Northern Ghana as assessed by MODIS and Landsat satellite imagery?\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003e\u003cspan\u003eii.\u0026nbsp;\u003c/span\u003e\u003cspan\u003eHow do institutional perspectives on shifting fire regimes relate to trends identified with remote sensing?\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003e\u003cspan\u003eiii.\u0026nbsp;\u003c/span\u003e\u003cspan\u003eWhat are the drivers of these shifting fire regimes, including landcover, precipitation and social factors?\u003c/span\u003e\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStudy Site\u003c/h2\u003e\u003cp\u003eThis study focused on the five regions of Northern Ghana (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003e), where a greater proportion of the landscape (90%) experiences annual vegetation burning (MLFM \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). Before 2017, Northern Ghana consisted of Northern, Upper East and Upper West Regions. In 2018, two new regions, Savannah and North-East were created by carving them out of the Northern Region. This administrative change led to the five regions in Northern Ghana. Northern Ghana is situated within the Guinea and Sudan savanna ecological zones. The region\u0026rsquo;s vegetation is characterised by fire and drought-resistant woody species including Shea (\u003cem\u003eVitelleria paradoxa), bushwillow (Combretum spp), wild syringa (Burkea africana)\u003c/em\u003e and \u0026lsquo;abogo\u0026rsquo; (\u003cem\u003eIsoberlinia doka)\u003c/em\u003e with common grasses such as \u003cem\u003eAndropogon, Heteropogon\u003c/em\u003e and \u003cem\u003eHyparrhenia spp\u003c/em\u003e (Amoako and Gambiza 2019). It covers a total landmass of about 97,250 km\u003csup\u003e2\u003c/sup\u003e, with thin soils overlaid with an iron pan. An iron pan is a hard soil layer cemented by iron oxides and usually consists of sand or sand gravel. The area is bounded by Burkina Faso to the North and North-west, Togo to the East and C\u0026ocirc;te D\u0026rsquo;Ivoire to the West. The southern edge of the zone is bordered by the Oti and Bono Regions. Northern Ghana has a tropical climate with unimodal rainfall patterns influenced by two air masses: the North-east trade winds and the South-westerly. However, because of its proximity to the Sahel, Northern Ghana experiences a dry climate caused by the north-easterly winds (harmattan) between November and April, facilitating vegetative burning.\u003c/p\u003e\u003cp\u003eFire suppression policies have a long history in Northern Ghana\u0026rsquo;s savanna landscapes. These policies were initially driven by the colonial rulers\u0026rsquo; interest in forest resources and their concerns about the effects of fire on vegetation, as well as its impacts on local climate, hydrology and downstream flow of major rivers like the Niger and Volta (Wardell et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2004\u003c/span\u003e, Wardell and Lund 2006). Realising that complete fire suppression in this area was undesirable and impracticable, scientific research by colonial botanists embraced an early burning regime as it affected trees slightly less than late fires (Laris et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Thus, early burning was recognised as a necessary tool in savanna land management towards the end of the colonial rule (Laris and Wardell 2006). However, post-independence, the country witnessed a return to restrictive fire policies because of the government's interest in protecting its important projects like the hydroelectric dam, irrigated agriculture, and industrial forest plantations (Wardell et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Moreover, the drought events in the Sahel region in the 1980s triggered wildfires, destroying lives and properties in the country (Ampadu-Agyei \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e1988\u003c/span\u003e). This reignited discussion about bushfires and led to key fire policies like the Provisional National Defence Council (PNDC) Law 46 in the 1980s, PNDC Law 229 in the 1990s, the 2006 Wildfire Policy and other related sectoral policies. The PNDC Laws 46 and 229 had several stipulations and restrictions, with punishment for non-compliance, but never achieved their goal because local communities resisted and set more fires (Apusigah \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). The 2006 wildfire policy is thought to be more inclusive and recognises traditional fire use, aiming to reduce a perceived bushfire menace in the country (Apusigah \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Doe \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). However, the various terms used to describe fires (e.g., bushfire, wildfire, traditional burning) in policy documents and academic papers, along with the prevailing narratives about fires, have led to a misunderstanding of the types of fires and the ones requiring management (Alare, Tebbs, and Schreckenberg, n.d.).\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eBiophysical data collection and analyses\u003c/h3\u003e\n\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eWe employed the Moderate Resolution Imaging Spectroradiometer\u0026rsquo;s (MODIS) burned area dataset (MCD64A1, 500 m spatial resolution) to analyse spatiotemporal burned area trends. Additionally, we used MODIS Active fire dataset (MOD14A1 V6, 1 km spatial resolution) to investigate temporal trends in active fires and fire seasonality (Giglio et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) between 2001 and 2022. It is worth noting that MODIS active fire products are more sensitive to smaller fires (\u0026ge;\u0026thinsp;10 ha) than MODIS-burned area products (\u0026ge;\u0026thinsp;169 ha) under pristine conditions (Fusco et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). We also used the Global Annual Burned Area Mapping (GABAM) dataset, derived from 30 m Landsat 8 Surface Reflectance imagery as described by Long et al. (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). This enabled us to identify the types of fires and to compare and assess the spatiotemporal burned area trends with the MODIS burned area data, considering the limitations of MODIS satellite sensors in detecting small fires (Laris \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2005\u003c/span\u003e, Roteta et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The trend analysis was limited to a 22-year duration because there was no satellite data available beyond this point.\u003c/p\u003e\u003cp\u003eAlthough Landsat has a fine spatial resolution (30 m), it is constrained by low temporal resolutions (with observations only once every 16 days) and for areas like Northern Ghana where vegetation recovery is very quick after a fire, long temporal gaps usually result in higher omission errors (Bowman et al. 2003). Additionally, Long et al. (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) assert that one of the drawbacks of the Landsat burned area data is the uncertainty associated with agricultural lands. In these areas, harvested or ploughed surfaces can be mistaken for burned areas due to similarities in spectral and temporal characteristics. While Sentinel-2 would have been ideal for this study due to its 10 m spatial resolution and high revisit frequency coverage, the short temporal coverage (data available since 2015) makes it less suitable for long-term fire regime analysis.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eThe connection between cropland expansion and the decline of the global and regional burned areas (Andela et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Dwomoh and Wimberly \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), requires an analysis of the spatiotemporal trends of land use land cover (LULC) in Northern Ghana. Like remotely sensed burned area products, LULC maps for Africa must provide accurate spatial and temporal resolution of land cover types and their extent for effective policy development (Silva et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2005\u003c/span\u003e, Kerner et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Therefore, we chose the Global Land and Aridity Discovery (GLAD) LULC dataset because of its fine spatial resolution (Landsat satellite image, 30 m) and long-term temporal coverage (data available for the years 2000 and 2020) (Potapov et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Moreover, it is one of the most accurate cropland products (Kerner et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eRainfall data was used as a predictor variable to analyse the mean annual rainfall and its influence on fire seasonality and fuel load (Dwomoh and Wimberly \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Due to rain gauge data sparsity in Ghana, we relied on Climate Hazards Group Infrared Precipitation with Station Data (CHIRPS) data for rainfall analysis from 2001 to 2022. CHIRPS gridded daily rainfall product (UCSB-CHG/CHIRPS/DAILY) combines satellite infrared data and in-situ precipitation station data to produce accurate, detailed and extended periods of rainfall estimates (Funk et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Observed climate data (rainfall) from Tamale (Northern Region), Yendi (Northern Region), Bole (Savannah Region), Navrongo (Upper East Region) and Wa (Upper West Region) were also obtained from Ghana Meteorological Agency rain gauges to validate findings of the CHIRPS Rainfall analysis.\u003c/p\u003e\u003cp\u003eFollowing the data collection, we used Google Earth Engine (GEE) to analyse all remotely sensed images. Each region's time series was exported as a CSV file for further statistical analysis and visualisation in R studio. For the statistical analysis in R studio, we performed a Shapiro-Wilk normality test to determine if the data distribution of burned area, active fires and rainfall required a parametric statistical analysis. We considered the Shapiro-Wilk test of normality due to its effectiveness in analysing small datasets (n\u0026thinsp;\u0026lt;\u0026thinsp;50) and being less impacted by outliers (Mendes and Pala 2003, Mohd Razali and Bee Wah 2011). Since the results revealed normally distributed data for the burned area (W\u0026thinsp;=\u0026thinsp;0.97, p\u0026thinsp;=\u0026thinsp;0.73) and active fires (W\u0026thinsp;=\u0026thinsp;0.93, p\u0026thinsp;=\u0026thinsp;0.11) at a 95% confidence level, we employed linear regression analysis to evaluate the significance of the observed trends in the time series data. Concerning rainfall, the Shapiro-Wilk test showed that the data was not normally distributed (W\u0026thinsp;=\u0026thinsp;0.9, p\u0026thinsp;=\u0026thinsp;1.70 e\u003csup\u003e\u0026minus;\u0026thinsp;12\u003c/sup\u003e). Therefore, a non-parametric test such as the Mann-Kendall test was used to assess whether there was a monotonic trend in the time series data. Landcover data was presented as histogram charts.\u003c/p\u003e\n\u003ch3\u003eInstitutional interviews\u003c/h3\u003e\n\u003cp\u003eTo acquire expert knowledge on the perceived trends of burned areas, active fires, and cropland areas, as well as the drivers of these spatial and temporal trends in Northern Ghana\u0026rsquo;s savanna landscapes, we conducted semi-structured interviews with local government institutions (n\u0026thinsp;=\u0026thinsp;6), and non-governmental organisations (NGOs) (n\u0026thinsp;=\u0026thinsp;4) (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The study employed a combination of different sampling approaches to select interview participants. Purposive sampling was used in the selection of Government institutions given their formal responsibility in fire management as outlined in the 2006 Ghana Wildfire Policy. However, NGOs involved in forest and biodiversity conservation-related projects were identified based on snowball sampling while participants\u0026rsquo; availability and willingness to participate determined the selection process through convenience sampling (Frey et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). The interviews were conducted between November 1 and December 7, 2022, and took place both virtually and face-to-face. The interviews focused on institutional perspectives of traditional burning practices and perceived changes in fire regimes. Additionally, a stakeholder engagement workshop was conducted on 16\u0026ndash;17 October 2024 in Damongo in the Savannah Region to discuss pathways for promoting equitable fire management in Northern Ghana. During the workshop, stakeholders identified additional social drivers that influenced the contrasting trends observed in the MODIS and Landsat satellite data. This helped to improve the reliability and consensus of the responses.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eInstitutional perceptions about changing fire regimes in Northern Ghana\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eInstitution (government, unless otherwise stated)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLocation\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003ePerception of trends in burned areas and active fires\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eExplanation for the observed trend\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eForestry Commission\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUpper East Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDecreasing in project communities\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eIntroduction of multilateral projects such as Sustainable Land and Water Management Projects (SLWMP), Ghana Landscape Restoration and Small-Scale Mining Project, (GLRSSMP) and Ghana Shea Landscape Emission Reduction Project (GSLERP) (2021\u0026ndash;2028) have and are: sensitising communities on the effects of bush burning, forming community fire volunteers, restoring degraded lands through tree planting, providing improved crop varieties to farmers and strengthening local resource governance with chiefs playing active roles.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGhana National Fire Service\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUpper West Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFluctuating\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eThis is influenced by a range of factors including livelihoods driven, low awareness of the impacts of bush burning.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSavannah Integrated Biodiversity and Conservation Initiative (NGO)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSavannah Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDecreasing\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eIntroduction of technologies such as tractors for ploughing and herbicides\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGhana National Fire Service (West Gonja Municipal Office)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSavannah Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eRemained the same or increased for West Gonja District over the last four years\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eBehavioural (not \u0026lsquo;rational\u0026rsquo; and just part of their lifestyle).\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWest Gonja Municipal Assembly\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSavannah Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eRemained the same\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eBehavioural (not \u0026lsquo;rational\u0026rsquo; and just part of their lifestyle)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMinistry of Agriculture\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNorthern Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDecreasing in project sites\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSensitizations, introduction of community fire volunteers and training on how to combat bushfires, presence of more elite chiefs who are environmental enthusiasts\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNational Disaster Management Organisation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUpper West Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFires have remained the same in the region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eLivelihood-driven\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWorld Vision International (NGO)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUpper East Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDecreasing in project communities and districts\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eCommunity awareness, radio discussions, collaboration with GNFS to train communities on how to combat fires and the establishment of community fire volunteers.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eA Rocha Ghana (NGO)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNorthern Ghana\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDecreasing in project communities\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eInfluence of high-profile projects such GSLERP, GLRSSMP and Landscape and Environmental Agility across the Nation (LEAN) projects focused on land restoration and fire management.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNandom Deanery Integrated Rural Development Project (NGO)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUpper West Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eDecreasing in project communities\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eCommunity awareness programmes,\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGhana National Fire Service\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eNorthern Region\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFluctuating\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eLivelihood-driven, lack of awareness\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWe transcribed the virtual interviews using Zoom and Microsoft Teams\u0026rsquo; transcription software, while face-to-face interviews were recorded and transcribed using Otter AI software. However, these transcription tools could not accurately transcribe the interviews due to accent-related issues, necessitating additional manual editing. This involved listening to the recording and editing sentences and words that were not transcribed well. The transcripts were then coded into themes based on the principles of grounded theory (Glaser and Strauss \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1967\u003c/span\u003e). This approach aims to develop theories that emanate from the knowledge and perceptions shared by interviewees concerning the subject matter. The NVivo 14 software aided the analysis of codes, memos, and quotations. Additionally, data from the stakeholder engagement workshop were used to explain the observed spatiotemporal trends.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003eSpatio-temporal patterns of burned area, fire seasonality, rainfall and land use land cover\u003c/h2\u003e\u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents trends in burned areas, active fires, and rainfall averaged across the whole of Northern Ghana from 2001 to 2022. MODIS 500 m burned area (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eA) declined from approximately 3750 km\u003csup\u003e2\u003c/sup\u003e in 2001 to 3305 km\u003csup\u003e2\u003c/sup\u003e in 2022. The trend was significant (p\u0026thinsp;=\u0026thinsp;0.076) at a 90% confidence level, with notable peaks in 2005 (4729 km\u003csup\u003e2\u003c/sup\u003e), 2011 (4279 km\u003csup\u003e2\u003c/sup\u003e) and 2017 (3859 km\u003csup\u003e2\u003c/sup\u003e). Conversely, Landsat 30 m burned area data (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eB) showed an increase from 1914 km\u003csup\u003e2\u003c/sup\u003e in 2001 to 2250 km\u003csup\u003e2\u003c/sup\u003e in 2021, but the trend was not statistically significant (p\u0026thinsp;=\u0026thinsp;0.54) at a 90% confidence level. The largest mean burned area of about 3400 km\u003csup\u003e2\u003c/sup\u003e occurred in 2002, whereas 2004 recorded the smallest mean burned area of about 200 km\u003csup\u003e2\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eMODIS burned area disaggregated by regions of Northern Ghana revealed contrasting temporal trends (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Notably, there were statistically significant decreasing trends at a 95% confidence level for the Savannah Region (p\u0026thinsp;=\u0026thinsp;0.01), the Upper East Region (p\u0026thinsp;=\u0026thinsp;0.02), and the Upper West Region (p\u0026thinsp;=\u0026thinsp;0.04). The North-East Region also showed a decrease in the burned area, but this trend was not statistically significant (p\u0026thinsp;=\u0026thinsp;0.12) at a 90% confidence level. In contrast, the Northern Region had increased burned area, albeit insignificant (p\u0026thinsp;=\u0026thinsp;0.26). For Landsat (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e4\u003c/span\u003e), no significant trend was observed for the North-East (p\u0026thinsp;=\u0026thinsp;0.19), Northern (p\u0026thinsp;=\u0026thinsp;0.44), Savannah (p\u0026thinsp;=\u0026thinsp;0.51), Upper East (p\u0026thinsp;=\u0026thinsp;0.69), and the Upper West Region (p\u0026thinsp;=\u0026thinsp;0.97).\u003c/p\u003e\u003cp\u003eRegarding active fires, the average number of MODIS active fires increased from 123 in 2001 to 128 in 2022 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Active fires peaked in 2005 (254), 2011 (274), 2017 (251), and 2020 (244). However, no significant trend (p\u0026thinsp;=\u0026thinsp;0.73) was observed at a 90% confidence level. When analysing the trends within the individual regions of Northern Ghana (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e5\u003c/span\u003e), the Northern Region exhibited a significant upward trend (p\u0026thinsp;=\u0026thinsp;0.031) at a 95% confidence level, while the trends in other regions were statistically insignificant.\u003c/p\u003e\u003cp\u003eThe fire season spanned from October to April with peak fires mostly occurring in December, however, the seasonal pattern varied between years (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Further analysis revealed a shift in the timing of the fire season, with a significant declining trend in active fires for November (p\u0026thinsp;=\u0026thinsp;0.038) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e7\u003c/span\u003eA), no significant trend for December (p\u0026thinsp;=\u0026thinsp;0.67) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e7\u003c/span\u003eB), and a significant increasing trend for January (p\u0026thinsp;=\u0026thinsp;0.0018) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e7\u003c/span\u003eC) and February (p\u0026thinsp;=\u0026thinsp;0.028) (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e7\u003c/span\u003eD) at a 95% confidence level.\u003c/p\u003e\u003cp\u003eRainfall in Northern Ghana between 2001 and 2022 showed considerable variability between years, with no significant trend (p\u0026thinsp;=\u0026thinsp;0.34) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). Notable peaks in mean annual rainfall occurred in 2003 (11575 mm), 2010 (11771 mm), and 2018 (11475 mm), while significant dips occurred in 2013 (7972 mm) and 2017 (8544 mm). The onset of rainfall typically starts in April and ends in October with shifts in the peak rainfall months from year to year (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). However, further analysis of the mean rainfall trends for all months (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003e in the supplementary document) revealed no significant trends in rainfall for any month. Rainfall in Northern Ghana showed no significant correlation with Landsat Burned Areas (r = -0.06, p\u0026thinsp;=\u0026thinsp;0.80) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003e in the supplementary document). However, a statistically significant but weak negative correlation was observed between rainfall and MODIS burned areas (r = -0.44) at a 95% confidence level (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003e. in the supplementary document).\u003c/p\u003e\u003cp\u003eLULC data for 2000 and 2020 revealed a cropland increase from about 13,000 km\u003csup\u003e2\u003c/sup\u003e to 19,000 km\u003csup\u003e2\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e8\u003c/span\u003e). Similarly, built-up areas experienced expansion from 789 km\u003csup\u003e2\u003c/sup\u003e in 2000 to 1,999 km\u003csup\u003e2\u003c/sup\u003e in 2020. Conversely, there was a decrease in area for dense short vegetation and tree cover over the same period.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eInstitutional perspectives on changing fire regimes\u003c/h2\u003e\u003cp\u003eInitial responses from interviews revealed varying perceptions about the changing fire regimes in Northern Ghana. However, in a subsequent stakeholder engagement workshop, participants perceived a decline in large fires (locally referred to as bushfires) in the area. They attributed this decline to several interventions, including intensified public sensitisations and government flagship programmes such as the Sustainable Land and Water Management Project (SLWMP), Planting for Food and Jobs (PFJ), Ghana Shea Landscape Emission Reduction Project (GSLERP), Ghana Land Restoration and Small-Scale Mining Project (GLRSSMP). Additionally, the establishment of fire posts at the various Metropolitan, Municipal and District Assemblies, as well as the provision of alternative mechanisms for land preparations using tractors and herbicides have contributed to managing fires. These perceptions aligned with the observed MODIS burned area trends for Northern Ghana. For instance, an interview with a project manager at a conservation NGO highlighted that:\u003c/p\u003e\u003cp\u003e\u003cem\u003eGenerally, bush burning increased in the landscape within the period and peaked during the logging and charcoal production activities. Burning is used to make the landscape more open for undertaking these activities. Also, with the increasing and expanding agricultural activities, more fires were being used to clear and prepare lands. However, this has decreased a bit within the last five years due to national and local bans on logging (2012\u0026ndash;2019) and charcoal burning (2017\u0026ndash;2021), respectively. Additionally, within this period, the presence of large-scale multinational programmes such as the SLWMP in 2011 and the Landscape and Environmental Agility Across Nation (LEAN) in 2020 has provided financial support for fire management and restoration efforts in Northern Ghana. (Project Manager, A Rocha Northern Ghana Sector, 2023)\u003c/em\u003e\u003c/p\u003e\u003cp\u003eIn terms of region-specific trends for MODIS burned areas, participants perceived the observed increasing trend in the Northern Region as due to lower population density coupled with increased hunting activities and the lack of wildlife-protected areas or community resource management areas (CREMAs). For example, the Environmental Specialist for the Savannah Agriculture Value Chain Development Project (SADEP) noted that:\u003c/p\u003e\u003cp\u003e\u003cem\u003eDuring the dry season in the Northern Region, you will find a lot of young men in about 3\u0026ndash;5 trucks going hunting. They target a large area for burning to hunt for game. (Environmental Specialist- SADEP, 2024)\u003c/em\u003e\u003c/p\u003e\u003cp\u003eConversely, participants attributed the lack of significant trends in smaller fires (Landsat burned area) in Northern Ghana to livelihood-related activities such as hunting, charcoal burning and using fires to clear new farmlands. These activities are particularly prevalent in the North-East, Northern, Savannah and Upper West Regions. The low levels of burned areas in the Upper East Region were attributed to low fuel availability. This scarcity is largely attributed to the region\u0026rsquo;s geology, which is predominantly rocky. Additionally, the Upper East Region has seen an increase in built-up areas and a decrease in tree cover due to charcoal production. In this region, many farmers also clear their fields by hoes (rather than using fire) so they can use the resulting crop residues as fodder for their livestock. For instance, a Lecturer at the University for Development Studies emphasised that:\u003c/p\u003e\u003cp\u003e\u003cem\u003ePeople are increasingly turning to charcoal production as an alternative livelihood, as farming has become less lucrative due to climate change. About 20 years ago, there was only one community along the route from Tamale to Accra where you could buy charcoal. Currently, numerous charcoal-selling spots exist, contributing to the prevalence of small fires. Moreover, existing studies indicate that increasing populations and built-up areas are limiting the areas available for large fires. As communities expand, they convert lands for settlements, which restricts the burning activities to small plots used for farming, ensuring that the fires do not spread uncontrollably. In the Upper East Region, for example, the land area is smaller and has a high population density. Therefore, there are not enough farms for burning activities, and there is less tree cover for charcoal production. (Lecturer, UDS)\u003c/em\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003eSpatio-temporal disparities in burned area and active fire trends in Northern Ghana\u0026rsquo;s savanna landscape\u003c/h2\u003e\u003cp\u003eOur results highlight the importance of scale and spatial resolution in fire regime analysis. MODIS and Landsat burned areas in Northern Ghana over the last 22 years revealed contrasting spatial and temporal trends. The MODIS dataset showed a decreasing trend in burned area, while the Landsat dataset showed no significant trend. Aggregating to the whole of Northern Ghana masked the contrasting trends occurring in different regions. The decreasing MODIS burned area trend (e.g., for Savannah, Upper East and Upper West Regions) aligns with previous research \u0026ndash; also using MODIS \u0026ndash; which shows both West Africa and global decreases in burned areas (Andela et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Dwomoh and Wimberly \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The coarse resolution (500 m) of MODIS hampers its ability to detect small fires (\u0026le;\u0026thinsp;25 ha) in heterogeneous landscapes(Laris \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2005\u003c/span\u003e, Roteta et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) which are commonly found in Northern Ghana, suggesting that the decreasing trend refers to larger fires. Conversely, the lack of a significant trend in Landsat burned areas for Northern Ghana and the different regions suggests a more complex picture. Taking both datasets into account, this indicates that large-scale fires may have been decreasing across Northern Ghana as a whole, while small fires showed no such decline.\u003c/p\u003e\u003cp\u003eHowever, it is important to acknowledge a key limitation. The trends we described are based on 22-year satellite data, constrained by MODIS and Landsat data availability. As such, the analysis captures only a relatively short window in the region's long-term trajectory of fire regimes. This makes it difficult to assess whether the observed pattern\u0026mdash;particularly the decrease in large fires\u0026mdash;represents sustained shifts or is part of a broader, more cyclical trend. However, historical records discussed in the study area section reveal significant shifts in fire governance\u0026mdash;from colonial-era suppression policies to more adaptive approaches that recognised early burning practices. In post-independence, the country witnessed a return to restrictive fire policies because of the government's interest in protecting its important projects and the Sahelian drought in the 1980s that triggered bushfire events in the country (Wardell et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). These governance changes likely influenced fire use patterns before satellite data became available. While we cannot quantify these historical dynamics using SRS, they provide important context for interpreting the observed trends and suggest that contemporary patterns may be shaped by deeper, policy-driven legacies of fire management (Alare et al. n.d.).\u003c/p\u003e\u003cp\u003eTo investigate the factors influencing the observed year-to-year variability in the burned areas of Northern Ghana for both MODIS and Landsat, we conducted a correlation analysis to examine the relationship between rainfall and the burned areas. The results revealed a weak but significant negative correlation between rainfall and MODIS burned area, suggesting that rainfall inhibited burning. This is consistent with the findings ofN\u0026rsquo;Datchoh et al. (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) who revealed that a wet rainfall season leads to high humidity, suppressing fire spread. As a result, no fires were observed during the rainy season (April-October), but in the dry season (November to March). Conversely, there was no significant relationship between rainfall and the Landsat burned area. This suggests that anthropogenic activities, primarily livelihood-driven, mostly control the year-to-year variability of small fires. For instance, participants from the stakeholder engagement workshop attributed the increasing number of small fires in Northern Ghana to hunting, farm preparations and charcoal production activities. Hunting and charcoal production activities provide alternate livelihoods for communities, helping to diversify income sources amid growing food insecurity caused by poverty and vulnerability to climate change impacts (Yaro and Hesselberg 2010, Bonye et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This aligns with the findings ofArchibald et al. (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) andN\u0026rsquo;Datchoh et al. (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) who contend that human influence on fire regimes in African savannas is more influential than the effect of climate in driving variation between years. However, when these activities coincide with extreme weather conditions like the hot and strong windy conditions in the dry season where fuels are sufficiently dried with no fire breaks or belts, the latter can exacerbate large fire risk in these areas. Participants also emphasised that recent national bans on charcoal production and logging have reduced the occurrence of large fires. However, communities tend to revert to these practices when enforcement decreases.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eSocial drivers of region-specific shifting fire regimes\u003c/h2\u003e\u003cp\u003eIn regions (North-East, Savannah, Upper East and West Regions) where large fires were decreasing based on MODIS burned area and active fire data, the workshop participants credited the decrease to interventions such as intensified public sensitisations, and government and multilateral flagship programmes such as the SLWMP, PFJ, GSLERP, LEAN and GLRSSMP. These multilateral projects have implemented tree planting and protection projects in selected Districts and communities in Northern Ghana. A key aspect of these initiatives is the incorporation of fire management plans to suppress burning that could potentially destroy investments in these trees. The fire management plans typically include bushfire sensitisation, the establishment of community fire volunteer groups, and the provision of incentives to discourage landscape burning. Despite the win-win discourses that dominate tree planting initiatives, they can potentially promote land grabbing, constraining local people\u0026rsquo;s access to resources and other resource-dependent livelihoods (Benjaminsen et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2011\u003c/span\u003e, Leach and Scoones 2015, Kandel et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). It was, therefore, not surprising to learn from institutional interviews that some tree planting projects in the Savannah Region of Northern Ghana were destroyed by arson. This occurred particularly in communities that lacked sufficient lands for farming. Moreover, extensive tree planting projects along with fire suppression policies in these areas can also promote fuel build-up for wildfire risk in the event of drought (Veldman et al. \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eSome workshop participants highlighted the role of population density and its knock-on effects on the establishment of fire posts and intensive agricultural activities (including the use of tractors and herbicides to clear land) in determining burning patterns. The number of people per square kilometre varies greatly between regions: Upper East (147.2), Northern (87.1), North-East (72.6), Upper West (49.0) and Savannah (18.7) (GSS 2021). The most densely populated Upper East Region has the lowest amount of MODIS burned area which also shows a significantly decreasing trend. This is consistent with previous studies that suggest high population densities are linked to smaller fires, as it limits the amount of available fuel for combustion and spread (Archibald and Roy \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2009\u003c/span\u003e, Bistinas et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Conversely, the very low population density and associated lack of intensive agricultural activity of the Savannah Region may contribute to its high count of active fires and large amount of burned area for both MODIS and Landsat datasets.\u003c/p\u003e\u003cp\u003eField observations by one of the authors in the Savannah Region revealed that the observed high counts of active fires and extensively burned areas were associated with burning activities around the Mole National Park (MNP) to enhance livelihoods. The MNP also extends to the Northern and Upper West Regions. The predominant livelihoods in these regions revolved around charcoal production, hunting and clearing new lands for yam (\u003cem\u003eDioscorea alata\u003c/em\u003e) cultivation, which all required fire use. These activities typically involve small fires, which can grow into large fires if not properly controlled, especially during the windy late dry season. Moreover, prescribed burning takes place in the park to regenerate grasses for grazing animals. There has also been an influx of migrant herders from neighbouring countries who bring their cattle to graze in the Savannah, Northern and Upper West Regions. These herders set fires to resprout grasses for grazing during the dry season (Kpienbaareh and Luginaah \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, Amoako and Gambiza 2020).\u003c/p\u003e\u003cp\u003eFor Landsat burned area, participants attributed the lack of decline in smaller fires to livelihood-related activities such as hunting, herding, charcoal burning, and using fires to clear new farmlands. The term \u0026lsquo;smaller fires\u0026rsquo; had various interpretations among the workshop participants. For example, the Ghana National Fire Service (GNFS) defined it specifically as fires at the ignition stage. Other participants perceived that fires could be small but intense and could spread and become large destructive fires. These perceptions about smaller fires could be attributed to the discursive framing of traditional burning practices as a \u0026lsquo;backward\u0026rsquo; practice and a threat to sustainable land management (Amanor \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2002\u003c/span\u003e, Snyder et al. 2019). Thus, most participants at the workshop expressed a shared objective of suppressing all types of fires in Northern Ghana through the range of interventions indicated above. However, existing studies in Northern Ghana suggest that small fires, particularly early prescribed burns, can promote species diversity and reduce the stem density of woody species (Amoako et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Therefore, the timing, frequency and extent of burning have more significant implications for land management (Laris \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2002\u003c/span\u003e) than merely evaluating the total extent of the burned area.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eFire and rainfall seasonality\u003c/h2\u003e\u003cp\u003eThe seasonality of fires observed in this study aligned with the movement of the Inter-Tropical Convergence Zone (ITCZ), resulting in a dry season from November to March in Northern Ghana. Within this period, the annual dry dusty wind from the Sahel, locally referred to as the Harmattan, dries vegetation and increases the spread of fires influenced by anthropogenic activities (Kugbe et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e, Dahan et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Our results revealed that fires peaked in December which is consistent with observations of Dahan et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)d Datchoh et al. (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). However, a detailed analysis revealed shifts in the fire season. Active fires showed a decreasing trend in November and an increasing trend for January and February. These shifts could not be attributed to shifts in rainfall patterns in Northern Ghana, which suggests that other human activities could be driving these changes.\u003c/p\u003e\u003cp\u003eThe observed shifts in the fire season imply a reduction in early burning and an increase in mid-dry season burning. These observations support findings by Laris et al. (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) who identified that peak annual fires in West Africa occur in the mid-dry season. Field observation and interviews in the West Gonja Municipal and the North Gonja District of the Savannah Region revealed that the mid-dry season burning is associated with livelihood activities like land preparation for farming and hunting. Such mid-dry season fires have been identified as having no significant adverse effects on trees because trees shed their leaves around this period (N\u0026rsquo;Dri et al. 2018). Laris et al. (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) contend that a more nuanced understanding of the effects of the different phases of burning or fire seasonality can only be achieved if studied within a broader context that takes into account multiple variables such as fuel type, fuel bed load, soil conditions, land use, time of the day, wind direction, humidity and temperature. Such studies in Northern Ghana could also enhance effective fire management. Currently, fire policies and projects in Ghana often advocate for early burning without clearly establishing what early burning entails. This has led to various interpretations regarding the timing of early burns and the appropriate times to conduct them. Interviews with local communities in the Savannah Region suggested that a more tailored message around early burning focusing on the moisture content of the vegetation and weather conditions rather than the specific months would be more useful and in line with traditional community decision-making.\u003c/p\u003e\u003cp\u003eMoreover, it is important to note that currently, Northern Ghana is experiencing shifting and erratic rainfall patterns, with the early onset occurring in the western areas and late onset in the eastern part of Northern Ghana (Atiah et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). This shift could also account for the difference in MODIS burned area and active fire trends for the specific regions. The changes in rainfall timing and distribution, landscape scale, land use land cover types (LULC), fire suppression interventions as well as the types and sizes of fires collectively shape fire regime dynamics in savanna landscapes (Laris \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2005\u003c/span\u003e, N\u0026rsquo;Datchoh et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2015\u003c/span\u003e, Laris et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). However, the shift in rainfall patterns also has implications for various livelihoods dependent on fires in this area. Therefore, understanding shifts in fire regimes and their impacts on local communities is essential for developing effective strategies for fire management, land use planning and sustainable resource utilisation.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eLand cover change and changing fire regimes in Northern Ghana\u003c/h2\u003e\u003cp\u003eThe findings from Landsat LULC suggested an increase in agricultural and built-up expansions in Northern Ghana. This increase aligned with the land use changes perceived by stakeholders. It also supports findings of existing studies in the Upper Guinean Region of West Africa and Africa which reported increases in agricultural lands (Andela et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Dwomoh and Wimberly \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). However, while their findings attributed agricultural expansion to decreasing burned area in these places, our study adds nuance to the explanation. Drawing on the finer-resolution Landsat data, our research shows that the observed agriculture expansion and built-up areas are associated with a decrease in larger fires and no change in smaller fires in Northern Ghana. As indicated in earlier discussions, fire is used in clearing new lands for farming, contributing to cropland expansion in these areas.\u003c/p\u003e\u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study draws on evidence from remote sensing and institutional interviews to assess changing fire regimes in Northern Ghana. When considering Northern Ghana as a whole, Landsat and MODIS burned areas depicted different fire trends and types; MODIS showed a decrease in large fires, whereas Landsat revealed no change in small fires. Despite numerous project interventions on fires, there is little agreement on how they are defined. This lack of agreement overlooks the various types of fires in the savanna landscape and characterises them solely as destructive fires. Therefore, local-level decision-makers may need to determine the types of fires they are interested in monitoring and consistently evaluate them using the most appropriate method. MODIS satellite data would be suitable for monitoring large fires, while on-the-ground reporting or the use of alternative fine spatial resolution datasets such as Landsat or Sentinel-2 satellite data would be more appropriate for small fires.\u003c/p\u003e\u003cp\u003eRegion-specific trend analysis indicates that the Savannah Region stands out as the region with the most extensive burned area and higher active fire counts. Our results highlight the importance of scale in fire regime analysis since aggregating to the whole of Northern Ghana masked the varying trends in different regions. Thus, we argue that disaggregation of data is important to ensure fire interventions are cost-effective by responding to region-specific fire trends.\u003c/p\u003e\u003cp\u003eThe study also highlighted shifts in fire seasonality in Northern Ghana. It showed a decrease in early burning, and an increase in fire activities in the mid-dry season, specifically in January and February. This has important implications for fire management in Northern Ghana, as policies and projects tend to promote early burning without explicitly defining what early burning entails and rely on November and December as months for early burning. This suggests that current fire management strategies may not be aligned with the actual fire dynamics of Northern Ghana. Therefore, there is a need for more adaptable and responsive fire management practices tailored to the changing fire regimes in this area.\u003c/p\u003e\u003cp\u003eAdditionally, changes in fire regimes are not only impacted by biophysical factors but also mediated through political and socio-cultural factors. Insights from institutional interviews and the stakeholder engagement workshop indicated that institutions implemented measures to suppress fires in their project communities. This may have contributed to the reduction of large fires. To determine the effectiveness of these interventions, institutions need to consider the perspective of local communities. While we recommend upscaling these interventions to other communities, it is also important to recognise that fire shapes and maintains the composition and the functioning of savanna ecosystem services. Therefore, to ensure the sustainability of such projects, and reduce the risk of later devastating fires, it is important to establish appropriate land use systems and acknowledge the significance of prescribed and controlled burning practices in savanna ecosystems.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors thank the London Interdisciplinary Social Science Doctoral Training Programme (LISS-DTP) for funding this work. The authors are also grateful to Henry Thompson, Davide Lomeo, and Aidan Byrne for assisting in coding the remote sensing analysis. Special thanks also to all individuals from the various institutions in Ghana who participated in the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor\u0026rsquo;s contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRSA designed the research, collected and analysed the data, and drafted the paper.\u003c/p\u003e\n\u003cp\u003eBoth KS and ET reviewed the methodology and the results and contributed to the editing of the paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research work is funded by the Economic Social and Research Council (ESRC) London Interdisciplinary Social Science Doctoral Training Programme (LISS-DTP), Grant Ref: ES/P000703/1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used to estimate burned area, active fires, rainfall, land use and land cover are openly accessible and free to use as detailed in the methods section. However, the datasets developed and used in this manuscript can be obtained from the authors upon request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe studies involving human participants were reviewed and approved by King\u0026rsquo;s College London Research Ethics Office (Ethical clearance reference number: MRSP-21/22-30162).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAlare, Rahinatu S., Emma Tebbs, and Kate Schreckenberg. n.d. Navigating Discourses: Unpacking the Problematisation of Traditional Burning Practices in Northern Ghana. \u003cem\u003eIn Preparation [Unpublished]\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmanor, K. S. 2002. \u003cem\u003eBushfire Management, Culture and Ecological Modernisation in Ghana\u003c/em\u003e. vol. 33 United Kingdom: Brighton.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmoako, Esther Ekua, James Gambiza. 2019. Effects of Anthropogenic Fires on Soil Properties and the Implications of Fire Frequency for the Guinea Savanna Ecological Zone, Ghana. \u003cem\u003eScientific African\u003c/em\u003e 6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.sciaf.2019.e00201\u003c/span\u003e\u003cspan address=\"10.1016/j.sciaf.2019.e00201\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmoako, Esther Ekua, James Gambiza. 2020. Traditional Practices, Knowledge and Perceptions of Fire Use in a West African Savanna Parkland. \u003cem\u003eBioRxiv\u003c/em\u003e 60:53\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.35979/alj.2020.02.60.53\u003c/span\u003e\u003cspan address=\"10.35979/alj.2020.02.60.53\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmoako, Esther, Hamza Ekua, and Issifu, Rikiatu Husseini. 2023. The Effects of Prescribed Dry Season Burning on Woody Species Composition, Mole National Park, Ghana. \u003cem\u003eTropical Conservation Science\u003c/em\u003e 16:1\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/19400829231164936\u003c/span\u003e\u003cspan address=\"10.1177/19400829231164936\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAmpadu-Agyei, Okyeame. 1988. Bushfires and Management Policies in Ghana. \u003cem\u003eThe Environmentalist\u003c/em\u003e 8(3):221\u0026ndash;228. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/BF02240254\u003c/span\u003e\u003cspan address=\"10.1007/BF02240254\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAndela, N., D. C. Morton, L. Giglio, Y. Chen, G. R. Van Der Werf, P. S. Kasibhatla, and R. S. DeFries et al. 2017. A Human-Driven Decline in Global Burned Area. \u003cem\u003eScience\u003c/em\u003e 356(6345):1356\u0026ndash;1362. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.aal4108\u003c/span\u003e\u003cspan address=\"10.1126/science.aal4108\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAndrew Wardell, D., and Christian Lund. 2006. Governing Access to Forests in Northern Ghana: Micro-Politics and the Rents of Non-Enforcement. \u003cem\u003eWorld Development\u003c/em\u003e 34(11):1887\u0026ndash;1906. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.worlddev.2005.11.021\u003c/span\u003e\u003cspan address=\"10.1016/j.worlddev.2005.11.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eApusigah, Agnes Atia. 2007. Promoting Sustainable Wildfire Management in Northern Ghana: Learning from History. \u003cem\u003eEuropean Journal of Social Sciences\u003c/em\u003e 5(1):61\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArchibald, S., A. Nickless, N. Govender, R. J. Scholes, and V. Lehsten. 2010. \u003cem\u003eGlobal Ecology and Biogeography\u003c/em\u003e 19:794\u0026ndash;809. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1466-8238.2010.00568.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1466-8238.2010.00568.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Climate and the Inter-Annual Variability of Fire in Southern Africa: A Meta-Analysis Using Long-Term Field Data and Satellite-Derived Burnt Area Data.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArchibald, S., and D. P. Roy. 2009. Identifying Individual Fires from Satellite-Derived Burned Area Data. \u003cem\u003eInternational Geoscience and Remote Sensing Symposium (IGARSS)\u003c/em\u003e 3:160\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/IGARSS.2009.5417974\u003c/span\u003e\u003cspan address=\"10.1109/IGARSS.2009.5417974\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAtiah, Winifred, Francis K. Ayinpogbilla, Bekele Muthoni, Fred Kotu, Kizito, and Leonard K. Amekudzi. 2021. Trends of Rainfall Onset, Cessation, and Length of Growing Season in Northern Ghana: Comparing the Rain Gauge, Satellite, and Farmer\u0026rsquo;s Perceptions. \u003cem\u003eAtmosphere\u003c/em\u003e 12(12). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/atmos12121674\u003c/span\u003e\u003cspan address=\"10.3390/atmos12121674\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBenjaminsen, Tor A., Ian Bryceson, and Faustin Maganga. 2011. and Tonje Refseth. Conservation and Land Grabbing in Tanzania Global Land Grabbing. In \u003cem\u003eInternational Conference on Global Land Grabbing\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBistinas, Ioannis, Duarte Oom, C. L. Ana, P. S\u0026aacute;, Sandy, I. Harrison, Colin Prentice, M. C. Jos\u0026eacute;, and Pereira. 2013. Relationships between Human Population Density and Burned Area at Continental and Global Scales. \u003cem\u003ePlos One\u003c/em\u003e 8(12):1\u0026ndash;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0081188\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0081188\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBonye, Samuel, Gordon Yenglier Ziem, and Yiridomoh, Vivian Nsiah. 2023. Our Forest, Our Livelihood: Natural Resources\u0026rsquo; Use Controversies and Community Livelihood Sustainability in the Mole National Park, Ghana. \u003cem\u003eLand Use Policy\u003c/em\u003e 127. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.landusepol.2023.106589\u003c/span\u003e\u003cspan address=\"10.1016/j.landusepol.2023.106589\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoschetti, Luigi, St\u0026eacute;phane P. Flasse, and Pietro A. Brivio. 2004. Analysis of the Conflict between Omission and Commission in Low Spatial Resolution Dichotomic Thematic Products: The Pareto Boundary. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e 91(3\u0026ndash;4):280\u0026ndash;292. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rse.2004.02.015\u003c/span\u003e\u003cspan address=\"10.1016/j.rse.2004.02.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBowman, David M.J.S., K. Jennifer, Paulo Balch, William J. Artaxo, Jean M. Bond, Mark A. Carlson, M. Cochrane, Carla, and D\u0026rsquo;Antonio et al. 2009. Fire in the Earth System. \u003cem\u003eScience\u003c/em\u003e 324(5926):481\u0026ndash;484. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1163886\u003c/span\u003e\u003cspan address=\"10.1126/science.1163886\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCarmenta, Rachel, and Emilie Coudel, Angela M. Steward. 2019. Forbidden Fire: Does Criminalising Fire Hinder Conservation Efforts in Swidden Landscapes of the Brazilian Amazon? \u003cem\u003eGeographical Journal\u003c/em\u003e 185(1):23\u0026ndash;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/geoj.12255\u003c/span\u003e\u003cspan address=\"10.1111/geoj.12255\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChuvieco, Emilio, Florent Mouillot, R. Guido, and van der Werf et al. 2019. Jes\u0026uacute;s San Miguel, Mihai Tanasse, Nikos Koutsias, Mariano Garc\u0026iacute;a,. Historical Background and Current Developments for Mapping Burned Area from Satellite Earth Observation. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e 225 (February): 45\u0026ndash;64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rse.2019.02.013\u003c/span\u003e\u003cspan address=\"10.1016/j.rse.2019.02.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDahan, Kueshi, Raymond Abudu S\u0026eacute;manou, and Kasei, Rikiatu Husseini. 2023. Contribution of Remote Sensing to Wildfire Trend and Dynamic Analysis in Two of Ghana\u0026rsquo;s Ecological Zones: Guinea-Savanna and Forest-Savanna Mosaic. \u003cem\u003eFire Ecology\u003c/em\u003e 19(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s42408-023-00198-z\u003c/span\u003e\u003cspan address=\"10.1186/s42408-023-00198-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDennis, Rona A., Judith Mayer, Grahame Applegate, Unna Chokkalingam, Carol J.Pierce Colfer, Iwan Kurniawan, and Henry Lachowski et al. 2005. Fire, People and Pixels: Linking Social Science and Remote Sensing to Understand Underlying Causes and Impacts of Fires in Indonesia. \u003cem\u003eHuman Ecology\u003c/em\u003e 33(4):465\u0026ndash;504. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10745-005-5156-z\u003c/span\u003e\u003cspan address=\"10.1007/s10745-005-5156-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDoe, Sylvanus Sefenu Per. 2008. Fire Management Policy and Environmental NGO Movement: Critical Contemporary Perspectives, Indicators and Prospects in Savannas, Ghana. \u003cem\u003eGeoJournal\u003c/em\u003e 71 (2\u0026ndash;3): 159\u0026ndash;70. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10708-008-9153-9\u003c/span\u003e\u003cspan address=\"10.1007/s10708-008-9153-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDwomoh, Francis K., and Michael C. Wimberly. 2017. Fire Regimes and Their Drivers in the Upper Guinean Region of West Africa. \u003cem\u003eRemote Sensing\u003c/em\u003e 9(11). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs9111117\u003c/span\u003e\u003cspan address=\"10.3390/rs9111117\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFrey, L., C. H. Botan, and G. Kreps. 2000. \u003cem\u003eInvestigating Communication: An Introduction to Research Methods\u003c/em\u003e. Allyn and Bacon.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFunk, Chris, Pete Peterson, Martin Landsfeld, Diego Pedreros, James Verdin, Shraddhanand Shukla, and Gregory Husak et al. 2015. The Climate Hazards Infrared Precipitation with Stations - A New Environmental Record for Monitoring Extremes. \u003cem\u003eScientific Data\u003c/em\u003e 2:1\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/sdata.2015.66\u003c/span\u003e\u003cspan address=\"10.1038/sdata.2015.66\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFusco, Emily J., John T. Finn, John T. Abatzoglou, Jennifer K. Balch, Sepideh Dadashi, and Bethany A. Bradley. 2019. Detection Rates and Biases of Fire Observations from MODIS and Agency Reports in the Conterminous United States. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e 220(January):30\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rse.2018.10.028\u003c/span\u003e\u003cspan address=\"10.1016/j.rse.2018.10.028\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGiglio, L., J. T. Randerson, G. R. Van Der, P. S. Werf, G. J. Kasibhatla, D. C. Collatz, R. S. Morton, and Defries et al. 2009. Assessing Variability and Long-Term Trends in Burned Area by Merging Multiple Satellite Fire Products. \u003cem\u003eBiogeosciences Discussions\u003c/em\u003e 6(6):11577. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/bgd-6-11577-2009\u003c/span\u003e\u003cspan address=\"10.5194/bgd-6-11577-2009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGiglio, L., W. Schroeder, J. Hall, and C. Justice. 2021. MODIS Collection 6 and Collection 6.1 Active Fire Product User\u0026rsquo;s Guide. Washington D.C., USA.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGlaser, B. G., and A. L. Strauss. 1967. \u003cem\u003eThe Discovery of Grounded Theory: Strategies for Qualitative Research\u003c/em\u003e. New Brunswick, USA and London, UK: Aldine Transaction.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGSS. 2021. \u003cem\u003eGhana 2021 Population and Housing Census\u003c/em\u003e: Preliminary Report. Accra, Ghana.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHantson, Stijn, Gitta Lasslop, Silvia Kloster, and Emilio Chuvieco. 2015. Anthropogenic Effects on Global Mean Fire Size. \u003cem\u003eInternational Journal of Wildland Fire\u003c/em\u003e 24(5):589\u0026ndash;596. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1071/WF14208\u003c/span\u003e\u003cspan address=\"10.1071/WF14208\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHarwell, Emily E. 2000. Remote Sensibilities: Discourses of Technology and the Making of Indonesia\u0026rsquo;s Natural Disasters. \u003cem\u003eDevelopment and Change\u003c/em\u003e 31(1):307\u0026ndash;340. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1467-7660.00156\u003c/span\u003e\u003cspan address=\"10.1111/1467-7660.00156\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJerven, Morten. 2016. Benefits and Costs of the Data for Development Targets for the Post-2015 Development Agenda. United Nations Working Paper.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKandel, Matt, Genevieve Agaba, Rahinatu S. Alare, and Thomas Addoah, Kate Schreckenberg. 2021. Assessing Social Equity in Farmer-Managed Natural Regeneration (FMNR) Interventions: Findings from Ghana. \u003cem\u003eEcological Restoration\u003c/em\u003e 39(1\u0026ndash;2):64\u0026ndash;76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3368/er.39.1-2.64\u003c/span\u003e\u003cspan address=\"10.3368/er.39.1-2.64\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKerner, Hannah, Catherine Nakalembe, Adam Yang, Ivan Zvonkov, Ryan McWeeny, and Gabriel Tseng, Inbal Becker-Reshef. 2024. How Accurate Are Existing Land Cover Maps for Agriculture in Sub-Saharan Africa? \u003cem\u003eScientific Data\u003c/em\u003e 11(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41597-024-03306-z\u003c/span\u003e\u003cspan address=\"10.1038/s41597-024-03306-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKleemann, Janina, G\u0026uuml;lendam Baysal, N. N. Henry, Bulley, and Christine F\u0026uuml;rst. 2017. Assessing Driving Forces of Land Use and Land Cover Change by a Mixed-Method Approach in North-Eastern Ghana, West Africa. \u003cem\u003eJournal of Environmental Management\u003c/em\u003e 196:411\u0026ndash;442. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jenvman.2017.01.053\u003c/span\u003e\u003cspan address=\"10.1016/j.jenvman.2017.01.053\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKpienbaareh, D., and I. Luginaah. 2019. After the Flames Then What? Exploring the Linkages between Wildfires and Household Food Security in the Northern Savannah of Ghana. \u003cem\u003eInternational Journal of Sustainable Development and World Ecology\u003c/em\u003e 26(7):612\u0026ndash;624. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13504509.2019.1640311\u003c/span\u003e\u003cspan address=\"10.1080/13504509.2019.1640311\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKugbe, Joseph X., Fosu Mathias, Tamene L. Desta, Manfred Denich, L. G. Paul, and Vlek. 2012. Annual Vegetation Burns across the Northern Savanna Region of Ghana: Period of Occurrence, Area Burns, Nutrient Losses and Emissions. \u003cem\u003eNutrient Cycling in Agroecosystems\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10705-012-9514-0\u003c/span\u003e\u003cspan address=\"10.1007/s10705-012-9514-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKull, Christian A. 2009. and Paul Laris. Fire Ecology and Fire Politics in Mali and Madagascar. In \u003cem\u003eTropical Fire Ecology\u003c/em\u003e, edited by M. A Cochrane, 171\u0026ndash;226. New York: Springer. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-540-77381-8_7\u003c/span\u003e\u003cspan address=\"10.1007/978-3-540-77381-8_7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLaris, P. 2002. Burning the Seasonal Mosaic: Preventative Burning Strategies in the Wooded Savanna of Southern Mali. \u003cem\u003eHuman Ecology\u003c/em\u003e 30(2):155\u0026ndash;186. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1023/A:1015685529180\u003c/span\u003e\u003cspan address=\"10.1023/A:1015685529180\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLaris, P., S. Dadashi, A. Jo, and S. Wechsler. 2016. Buffering the Savanna: Fire Regimes and Disequilibrium Ecology in West Africa. \u003cem\u003ePlant Ecology\u003c/em\u003e 217(5):583\u0026ndash;596. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11258-016-0602-0\u003c/span\u003e\u003cspan address=\"10.1007/s11258-016-0602-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLaris, Paul. 2011. Humanizing Savanna Biogeography: Linking Human Practices with Ecological Patterns in a Frequently Burned Savanna of Southern Mali. \u003cem\u003eAnnals of the Association of American Geographers\u003c/em\u003e 101(5):1067\u0026ndash;1088. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/00045608.2011.560063\u003c/span\u003e\u003cspan address=\"10.1080/00045608.2011.560063\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLaris, Paul, Aurahm Jo, and Suzanne P. Wechsler. 2018. Effects of Landscape Pattern and Vegetation Type on the Fire Regime of a Mesic Savanna in Mali. \u003cem\u003eJournal of Environmental Management\u003c/em\u003e 227(March):134\u0026ndash;145. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jenvman.2018.08.091\u003c/span\u003e\u003cspan address=\"10.1016/j.jenvman.2018.08.091\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLaris, Paul, Moussa Kon\u0026eacute;, and Sepideh Dadashi, Fadiala Dembele. 2017. The Early/Late Fire Dichotomy: Time for a Reassessment of Aubr\u0026eacute;ville\u0026rsquo;s Savanna Fire Experiments. \u003cem\u003eProgress in Physical Geography\u003c/em\u003e 41(1):68\u0026ndash;94. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0309133316665570\u003c/span\u003e\u003cspan address=\"10.1177/0309133316665570\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLaris, Paul S. 2005. Spatiotemporal Problems with Detecting and Mapping Mosaic Fire Regimes with Coarse-Resolution Satellite Data in Savanna Environments. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e 99(4):412\u0026ndash;424. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rse.2005.09.012\u003c/span\u003e\u003cspan address=\"10.1016/j.rse.2005.09.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLaris, Paul, David Andrew Wardell. 2006. Good, Bad or \u0026lsquo;Necessary Evil\u0026rsquo;? Reinterpreting the Colonial Burning Experiment \u0026hellip;. \u003cem\u003eThe Geographical Journal\u003c/em\u003e 172(4):271\u0026ndash;290.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeach, Melissa, Ian Scoones. 2015. Political Ecologies of Carbon in Africa. In \u003cem\u003eCarbon Conflicts and Forest Landscapes in Africa\u003c/em\u003e, Routledge. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4324/9781315740416\u003c/span\u003e\u003cspan address=\"10.4324/9781315740416\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLong, Tengfei, Zhaoming Zhang, Guojin He, Weili Jiao, Chao Tang, Bingfang Wu, Xiaomei Zhang, Guizhou Wang, and Ranyu Yin. 2019. 30m Resolution Global Annual Burned Area Mapping Based on Landsat Images and Google Earth Engine. \u003cem\u003eRemote Sensing\u003c/em\u003e 11(5):1\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs11050489\u003c/span\u003e\u003cspan address=\"10.3390/rs11050489\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMcCaffrey, Sarah, Eric Toman, and Melanie Stidham, Bruce Shindler. 2013. Social Science Research Related to Wildfire Management: An Overview of Recent Findings and Future Research Needs. \u003cem\u003eInternational Journal of Wildland Fire\u003c/em\u003e 22:15\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/http://dx.doi.org/10.1071/WF11115\u003c/span\u003e\u003cspan address=\"10.1071/WF11115\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMendes, Mehmet, Akin Pala. 2003. Type I Error Rate and Power of Three Normality Tests. \u003cem\u003eInformation Technology Journal\u003c/em\u003e 2(2):135\u0026ndash;139. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3923/itj.2003.135.139\u003c/span\u003e\u003cspan address=\"10.3923/itj.2003.135.139\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMistry, Jayalaxshmi, Andrea Berardi. 2016. Bridging Indigenous and Scientific Knowledge. \u003cem\u003eScience\u003c/em\u003e 352(6291):1274\u0026ndash;1275. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.aaf1160\u003c/span\u003e\u003cspan address=\"10.1126/science.aaf1160\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMLFM. 2006. National Wildfire Management Policy. Accra, Ghana.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMohd Razali, and Nornadiah, Yap Bee Wah. 2011. Power Comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and Anderson-Darling Tests. \u003cem\u003eJournal of Statistical Modeling and Analytics\u003c/em\u003e 2(1):13\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoura, Livia C., O. Aldicir, Isabel B. Scariot, Robin Schmidt, and Beatty. 2019. and Jeremy Russell-Smith. The Legacy of Colonial Fire Management Policies on Traditional Livelihoods and Ecological Sustainability in Savannas: Impacts, Consequences, New Directions. \u003cem\u003eJournal of Environmental Management\u003c/em\u003e 232 (December 2018): 600\u0026ndash;606. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jenvman.2018.11.057\u003c/span\u003e\u003cspan address=\"10.1016/j.jenvman.2018.11.057\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eN\u0026rsquo;Datchoh, E. T., A. Konar\u0026eacute;, A. Diedhiou, A. Diawara, E. Quansah, and P. Assamoi. 2015. Effects of Climate Variability on Savannah Fire Regimes in West Africa. \u003cem\u003eEarth System Dynamics\u003c/em\u003e 6(1):161\u0026ndash;174. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/esd-6-161-2015\u003c/span\u003e\u003cspan address=\"10.5194/esd-6-161-2015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eN\u0026rsquo;Dri, Aya, Tionhonk\u0026eacute;l\u0026eacute; Drissa Brigitte, Jacques Soro, Kanvaly Gignoux, and Dosso. Mouhamadou Kon\u0026eacute;, Julien Kouadio N\u0026rsquo;Dri, N\u0026rsquo;golo Abdoulaye Kon\u0026eacute;, and S\u0026eacute;bastien Barot. 2018. Season Affects Fire Behavior in Annually Burned Humid Savanna of West Africa. \u003cem\u003eFire Ecology\u003c/em\u003e 14 (2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s42408-018-0005-9\u003c/span\u003e\u003cspan address=\"10.1186/s42408-018-0005-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePotapov, Peter, Matthew C. Hansen, Indrani Kommareddy, Anil Kommareddy, Svetlana Turubanova, Amy Pickens, Bernard Adusei, and Alexandra Tyukavina, Qing Ying. 2020. \u003cem\u003eRemote Sensing\u003c/em\u003e 12(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs12030426\u003c/span\u003e\u003cspan address=\"10.3390/rs12030426\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Landsat Analysis Ready Data for Global Land Cover and Land Cover Change Mapping.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRanderson, J. T., Y. Chen, G. R. Van Der Werf, B. M. Rogers, and D. C. Morton. 2012. Global Burned Area and Biomass Burning Emissions from Small Fires. \u003cem\u003eJournal of Geophysical Research: Biogeosciences\u003c/em\u003e 117(4):1\u0026ndash;23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2012JG002128\u003c/span\u003e\u003cspan address=\"10.1029/2012JG002128\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoteta, E., A. Bastarrika, M. Padilla, T. Storm, and E. Chuvieco. 2019. Development of a Sentinel-2 Burned Area Algorithm: Generation of a Small Fire Database for Sub-Saharan Africa. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e 222 (December 2018): 1\u0026ndash;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rse.2018.12.011\u003c/span\u003e\u003cspan address=\"10.1016/j.rse.2018.12.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSilva, Jo\u0026atilde;o M.N., C. L. Ana, S\u0026aacute;, M. C. Jos\u0026eacute;, and Pereira. 2005. Comparison of Burned Area Estimates Derived from SPOT-VEGETATION and Landsat ETM\u0026thinsp;+\u0026thinsp;Data in Africa: Influence of Spatial Pattern and Vegetation Type. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e 96(2):188\u0026ndash;201. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rse.2005.02.004\u003c/span\u003e\u003cspan address=\"10.1016/j.rse.2005.02.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSmith, C., J. Ainscough, R. S. Alare, A. R. Croker, K. M. De Freitas, James D.A. Millington, and J. Mistry et al. 2024. How Policy Interventions Influence Burning to Meet Cultural and Small-Scale. \u003cem\u003eEcology and Society\u003c/em\u003e 29(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/https://doi.org/10.5751/ES-14850-290135\u003c/span\u003e\u003cspan address=\"10.5751/ES-14850-290135\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSnyder, Katherine A., and Beth Cullen, Juliet Braslow. 2019. Farmers as Experts: Interpreting the \u0026lsquo;Hidden\u0026rsquo; Messages of Participatory Video across African Contexts. \u003cem\u003eArea\u003c/em\u003e 51(4):779\u0026ndash;787. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/area.12538\u003c/span\u003e\u003cspan address=\"10.1111/area.12538\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSoest, Christian Von. 2023. Why Do We Speak to Experts? Reviving the Strength of the Expert Interview Method. \u003cem\u003ePerspectives on Politics\u003c/em\u003e 21(1):277\u0026ndash;287. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S1537592722001116\u003c/span\u003e\u003cspan address=\"10.1017/S1537592722001116\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSteen-Adams, Michelle M., and Susan Charnley, Mark D. Adams. 2017. Historical Perspective on the Influence of Wildfire Policy, Law, and Informal Institutions on Management and Forest Resilience in a Multiownership, Frequent-Fire, Coupled Human and Natural System in Oregon, USA. \u003cem\u003eEcology and Society\u003c/em\u003e 22(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5751/ES-09399-220323\u003c/span\u003e\u003cspan address=\"10.5751/ES-09399-220323\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSzpakowski, D. M., and J. L. R. Jensen. 2019. A Review of the Applications of Remote Sensing in Fire Ecology. \u003cem\u003eRemote Sensing\u003c/em\u003e 11(2638):1\u0026ndash;31. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/4445278\u003c/span\u003e\u003cspan address=\"10.2307/4445278\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVeldman, Joseph W., J. C. Aleman, S. T. Alvarado, T. M. Anderson, S. Archibald, W. J. Bond, and T. W. Boutton, N Buchmann. 2019. Comment on \u0026lsquo;The Global Tree Restoration Potential\u0026rsquo;. \u003cem\u003eScience\u003c/em\u003e 366:1\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.aaz0111\u003c/span\u003e\u003cspan address=\"10.1126/science.aaz0111\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWardell, D., Thomas Theis Andrew, Kjeld Nielsen, and Rasmussen. 2004. and Cheikh Mbow. Fire History, Fire Regimes and Fire Management in West Africa: An Overview. In \u003cem\u003eWildland Fire Management Handbook for Sub-Sahara Africa\u003c/em\u003e, edited by Johann G. Goldammer and Cornelis de Ronde, 350\u0026ndash;81. Global Fire Monitoring Center (GFMC). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003c/span\u003e\u003cspan address=\"http://www.oneworldbooks.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWerf, G. R., James T. Van Der, Louis Randerson, T. Giglio, Thijs, Yang Van Leeuwen, Brendan M. Chen, Mingquan Rogers, and Mu et al. 2017. Global Fire Emissions Estimates during 1997\u0026ndash;2016. \u003cem\u003eEarth System Science Data\u003c/em\u003e 9(2):697\u0026ndash;720. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/essd-9-697-2017\u003c/span\u003e\u003cspan address=\"10.5194/essd-9-697-2017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWilgen, B. W., N. Van, H. C. Govender, D. Biggs, Ntsala, and X. N. Funda. 2004. Response of Savanna Fire Regimes to Changing Fire-Management Policies in a Large African National Park. \u003cem\u003eConservation Biology\u003c/em\u003e 18(6):1533\u0026ndash;1540. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1523-1739.2004.00362.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1523-1739.2004.00362.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWooster, Martin J., J. Gareth, Louis Roberts, David Giglio, Patrick Roy, Luigi Freeborn, Christ Boschetti, and Justice et al. 2021. Satellite Remote Sensing of Active Fires: History and Current Status, Applications and Future Requirements. \u003cem\u003eRemote Sensing of Environment\u003c/em\u003e 267(May):112694. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rse.2021.112694\u003c/span\u003e\u003cspan address=\"10.1016/j.rse.2021.112694\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYaro, J. A., J Hesselberg. 2010. The Contours of Poverty in Northern Ghana: Policy Implications for Combating Food Insecurity. \u003cem\u003eResearch Review of the Institute of African Studies\u003c/em\u003e 26(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4314/rrias.v26i1.56957\u003c/span\u003e\u003cspan address=\"10.4314/rrias.v26i1.56957\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"fire-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"feco","sideBox":"Learn more about [Fire Ecology](https://www.springer.com/journal/42408)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/feco/default.aspx","title":"Fire Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Northern Ghana, savanna, fire regimes, institutions, MODIS burned area MODIS active fires, Landsat burned area","lastPublishedDoi":"10.21203/rs.3.rs-6441082/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6441082/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eGlobally, there are growing demands for evidence-based policies to manage wildfires. Currently, fire management in Northern Ghana relies on policies and projects developed without a comprehensive understanding of fire trends and drivers. This study analysed spatio-temporal trends in burned areas (500m MODIS and 30m Landsat products), active fires and fire seasonality using linear regression analysis to investigate shifts in fire regimes between 2000 and 2022. Rainfall and land cover changes during this period and institutional perspectives of the observed trends were also examined.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eWhen averaged across Northern Ghana, MODIS burned area data revealed a significant decreasing trend, while Landsat burned area, and active fires showed no significant trend. When disaggregated by region, MODIS burned area showed significant decreasing trends for the Savannah Region, Upper East and Upper West Regions. Conversely, Landsat burned area showed no trend in all regions of Northern Ghana. Active fires increased significantly in the Northern Region. Active fire data also revealed a significant shift in fire seasonality in Northern Ghana towards more mid-dry season fires. Institutional perspectives attributed the decline in large fires (identified by MODIS) to the success of interventions designed to reduce uncontrolled burning (locally referred to as bushfires). Conversely, increasing small fires (Landsat burned area and MODIS active fires) were perceived as being associated with smallholder livelihoods dependent on burning, which aligned with the increase in cropland extent observed in land cover data.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eOur results highlight that the scale and resolution of fire datasets are crucial considerations for analysing fire trends. Aggregating data across the whole of Northern Ghana obscured contrasting trends in individual regions. The different trends observed between MODIS and Landsat burned areas suggest a shift from large fires towards smaller ones in Northern Ghana. Institutions expressed a shared objective of suppressing all fires; however, when designing fire management policies, it is important to consider the type of fire, since fire trends, drivers and impacts can vary depending on the size and timing of burning, and the associated land use.\u003c/p\u003e","manuscriptTitle":"Diverging fire trends in Northern Ghana’s savanna landscape: Insights from remote sensing and institutional perspectives","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-22 09:36:55","doi":"10.21203/rs.3.rs-6441082/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-27T14:15:35+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-25T17:08:10+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-22T14:46:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"96667236215917996250476020438169304764","date":"2026-02-20T05:31:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"184004078312413186697957366258669156064","date":"2026-02-10T17:13:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"184963752834247719391864789034647871114","date":"2026-02-10T08:26:50+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-16T23:59:37+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-04-15T00:22:38+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-15T00:21:01+00:00","index":"","fulltext":""},{"type":"submitted","content":"Fire Ecology","date":"2025-04-13T20:49:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"fire-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"feco","sideBox":"Learn more about [Fire Ecology](https://www.springer.com/journal/42408)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/feco/default.aspx","title":"Fire Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"5f42ad51-4e10-4b73-9daa-224cef582443","owner":[],"postedDate":"July 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-20T14:25:46+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-22 09:36:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6441082","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6441082","identity":"rs-6441082","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}