Severe Forest Fires Research Articles

A two-stage Cox process model with spatial and nonspatial covariates

Rich new marked point process data allow researchers to consider disparate problems such as the factors affecting the location and type of police use of force incidents, and the characteristics that impact the location and size of forest fires. We develop a two-stage log Gaussian Cox process that models these data in terms of both spatial (community-level) and nonspatial (individual or event-level) characteristics; both types of covariates are present in the examples we consider and are not easy to incorporate via existing methods. Via simulated and real data examples we find that our model is easy to interpret and flexible, accommodating multiple types of marks and multiple types of spatial covariates. In the first example we consider, our approach allows us to study the impact of community-level socioeconomic features such as unemployment as well as event-level features such as officer tenure on force used by police, illustrated through simulated examples. In our second example we consider factors that impact the locations and severity of forest fires from the Castilla-La Mancha region of Spain between 2004–2007.

Mapping Susceptibility of Forest in Taman Raya Raden Soerjo Using Machine Learning Random Forest

The increasing incidence of forest fires in Raden Soerjo’s Grand Park over the last 5 years has become a concern for conducting research on fire predictions. The availability of free remote sensing data makes it easier to analyze forest fires using Geographic Information Systems (GIS). and remote sensing. Varying publicly available spatial datasets were used to classify forest fires affected areas with an approach of Machine Learning (ML) using Random Forest (RF). Nine (9) variables suspected as potential cause of forest fires (NDVI. ET. LST. Aspect. Slope. Altitude. Distance to Road (ED_Road). Distance to Built-up area (ED_BUA). and Distance to river (ED_River) were used as determinants. About 80% of the data were used as Trained data and 20% as a validation. The results of the model produced an accuracy of 0.96. The model result was checked for its sensitivity with the Area Under the Curve (AUC) AUC results with a value of 0.89. Findings show that Random Forest could be applied to map forest fire severity classes with a good result.

Unravelling the Local Dynamics of Increasing Fires in Community Forests of Mid-hills of Nepal

Forest fire is a global phenomenon and is having unprecedented impacts across continents. Nepal is not an exception to the increasing impacts, where hundreds of acres of forests is lost, or damaged, to forest fire. The severity of forest fire is on the rise in recent decades, wherein community forests have witnessed effects of the calamity over the recent years. This paper investigates the local factors behind the increasing frequency and severity of forest fires in mid-hills of Nepal. The paper draws on three separate cases from the research sites, involving five years of ‘Enhancing Livelihoods from Improved Forest Management in Nepal (EnLiFT2)’ project's team observations in Bhumlu rural municipality and Chautara Sangachokgadhi municipality of Kavrepalanchok and Sindhupalchok district respectively. This is complemented by informant interviews taken with 25 individuals. The paper primarily examines forest-people-fire relationship to demonstrate the weakening collective actions among the community forest user groups (CFUGs) as an important factor for increasing forest fires in the mid-hills of Nepal. We found the changing forest-people relationship, weakening CFUG governance, and increasing regulatory impositions as important factors driving the alienation of users from the forest thus, undermining collective action in forest management. We argue that strengthening collective action on better forest management to prevent forest fire is crucial over adopting mitigation techniques. The alienation of forest user groups from their forest in the changing socio-economic and forest management context can be addressed to strengthening the collective action for better forest management and ultimately to forest fire prevention and management.

Carbon dioxide and particulate emissions from the 2013 Tasmanian firestorm: implications for Australian carbon accounting

BackgroundUncontrolled wildfires in Australian temperate Eucalyptus forests produce significant smoke emissions, particularly carbon dioxide (CO2) and particulates. Emissions from fires in these ecosystems, however, have received less research attention than the fires in North American conifer forests or frequently burned Australian tropical savannas. Here, we use the 2013 Forcett–Dunalley fire that caused the first recorded pyrocumulonimbus event in Tasmania, to understand CO2 and particulate matter (PM2.5) emissions from a severe Eucalyptus forest fire. We investigate the spatial patterns of the two emissions using a fine scale mapping of vegetation and fire severity (50 m resolution), and utilising available emission factors suitable for Australian vegetation types. We compare the results with coarse-scale (28 km resolution) emissions estimates from Global Fire Emissions Database (GFED) to determine the reliability of the global model in emissions estimation.ResultsThe fine scale inventory yielded total CO2 emission of 1.125 ± 0.232 Tg and PM2.5 emission of 0.022 ± 0.006 Tg, representing a loss of 56 t CO2 ha−1 and 1 t PM2.5 ha−1. The CO2 emissions were comparable to GFED estimates, but GFED PM2.5 estimates were lower by a factor of three. This study highlights the reliability of GFED for CO2 but not PM2.5 for estimating emissions from Eucalyptus forest fires. Our fine scale and GFED estimates showed that the Forcett–Dunalley fire produced 30% of 2013 fire carbon emissions in Tasmania, and 26–36% of mean annual fire emissions for the State, representing a significant single source of emissions.ConclusionsOur analyses highlight the need for improved PM2.5 emission factors specific to Australian vegetation, and better characterisation of fuel loads, particularly coarse fuel loads, to quantify wildfire particulate and greenhouse gas emissions more accurately. Current Australian carbon accountancy approach of excluding large wildfires from final GHG accounts likely exaggerates Tasmania’s claim to carbon neutrality; we therefore recommend that planned and unplanned emissions are included in the final national and state greenhouse gas accounting to international conventions. Advancing these issues is important given the trajectory of more frequent large fires driven by anthropogenic climate change.

ANALYZING SPATIO-TEMPORAL PATTERN OF THE FOREST FIRE BURNT AREA IN UTTARAKHAND USING SENTINEL-2 DATA

Abstract. Forest fire burnt area estimation using Normalized Burn Ratio at regional level helps in understanding the pattern of the frequency and severity of forest fires. In this study, burnt area is estimated for all the thirteen districts of Indian state Uttarakhand for last six years from 2016 to 2021 using Sentinel 2A and 2B datasets. The spatial and temporal pattern of the burnt area was analyzed by incorporating different parameters such as meteorological parameters like land surface temperature, rainfall; edaphic parameter like surface soil moisture and vegetation parameters like Normalized Difference Vegetation Index & Enhanced Vegetation Index. The estimated burnt area was statistically analyzed with respect to the parameters stated and the relationship among them was quantified. It was found that burnt area is positively correlated with the land surface temperature, while it showed negative correlation with the pre-fire precipitation, pre-fire NDVI & EVI and the surface soil moisture for 11 out of 13 districts. The district-wise forest fire burnt area assessment and analysis of its spatio-temporal pattern can be used in the preparedness and mitigation planning to prevent drastic ecological impacts of forest fires on the landscape.

Fire severity assessment on peatland vegetation diversity

Forest fires have a significant impact on peatland ecosystem, including its vegetation diversity. Several approaches applied for forest fire severity assessment are available in the literature. This study aimed to implement field-based and spatial-aware fire severity assessments and to compare both approaches. Peat swamp forests of Ogan Komering Ilir, South Sumatera Province, for field assessment and Siak, Riau Province, for spatial assessment were selected in this research. Field-based fire severity assessment used indicators of vegetation analyses (individual tree damages, burned vegetation severity, and diversity index), which were measured and analyzed from sample plots in both burned and unburned areas. Spatial analyses of fires severity assessment employed Sentinel-2 satellite imagery with Normalized Burn Ratio (NBR) approach. The research revealed a significant change in tree species diversity, indicating a lower diversity in burned areas than the ones in pristine fields. Magnitude of changes in diversity was influenced by fire severity levels in burned areas and pre-fire condition of burned areas. Spatial analyses indicated low to high fire severity with burned area estimation for each fire severity class. Both approaches have their own benefits and challenges. Field-based severity assessment is beneficial for a detailed assessment of fire damages and losses, whereas spatial analyses would be considered for a broader estimation of fire severity. Both assessments are mutually implemented. Field-based assessment can serve as a verification tool for spatial assessment.

Warmer and drier conditions have increased the potential for large and severe fire seasons across south‐eastern Australia

AbstractAimThe aims were: (1) to identify the environmental drivers of interannual variation in wildfire extent and severity; (2) to examine temporal trends in climatic potential for large and severe wildfires; and (3) to assess whether environmental conditions experienced during the 2019–2020 mega‐fire season were anomalous.LocationSouth‐eastern Australia.Time period1953–2020.Major taxa studiedTemperate forests.MethodsWe used satellite‐derived fire severity mapping from 1988 to 2020 to model the effects of drought, weather and fuels on the annual area burned and the proportion of the area burned that was impacted by high‐severity fire across four bioregions. Trends in wildfire extent and severity were then estimated from 1953 to 2020 using these derived models and gridded climate data to assess changes in climatic potential for large and severe wildfires. Estimates of wildfire extent and severity for the 2019–2020 fire season were then assessed against prior seasons (1953–2019).ResultsAnnual area burned was positively related to the severity of seasonal drought and frequency of fire weather conditions that promote substantial daily fire growth. Wildfire severity was elevated in years with severe fire weather and increased with increasing antecedent drought in years without severe fire weather. Fuels had a lesser effect on wildfire extent and severity than climate. Potential fire extent and severity have increased over time in response to an increased severity of drought and worsening fire weather conditions. Estimates of wildfire extent and severity during the 2019–2020 fire season approached the upper extreme within each bioregion, owing to widespread extreme climatic conditions.Main conclusionsThe climatic potential for large and severe forest fires has increased across south‐eastern Australia since the 1950s, probably because of anthropogenic climate change. The magnitude and severity of the 2019–2020 fires reflected climatic conditions that are driving an increase in the size and severity of wildfires.

Forest avifauna exhibit enduring responses to historical high-severity wildfires

Forest fire size, frequency and severity are increasing worldwide, with corresponding reductions in long-unburnt habitat and greater modification of forest structure over wider areas. Understanding the implications for animals is imperative in optimizing management for species persistence and overall biodiversity. We investigated how avian responses to historical high-severity fire differ in forests at short (five years) and mid-range (16 years) time since fire, including whether increased time since fire mitigates any negative responses to high-severity fire. Sites were established in fire-prone dry forests of the Greater Blue Mountains World Heritage Area, Australia. A Bayesian latent variable analysis of bird data obtained from acoustic recordings was used to estimate the occurrence of 74 species in relation to time since fire (short, mid-range), the spatial extent of historical high-severity fire (limited, extensive), and their interaction. Time since fire influenced the number of species present, but only where high-severity fire had been extensive. Here, the lowest and highest number of species in the study occurred where time since fire was short and mid-range, respectively. At least ten species responded either positively or negatively to high-severity fire, but for nine of these species the response did not change with time since fire, potentially implicating persistent effects of such fires on habitat. Six other species were unlikely to occur at short time since fire, requiring habitat at mid-range time since fire. This finding suggests that these species would benefit from strategic retention of forest with longer fire-ages under increased fire activity associated with climate change.

EVALUATION OF FIELD-BASED BURN INDICES FOR ASSESSING FOREST FIRE SEVERITY IN LUHANSK REGION, UKRAINE

Evaluation of forest fire severity is a basis of post-fire forest management. Remote sensing-based methods enable reliable delineation of fire perimeters, however, assessments of the degree of forest damage need to be verified and adjusted through field sampling. The forest damage assessment conducted in this study is useful for practitioners to understand and justify the design of clear cuts for restoration purposes. Thus, the aim of the study is to verify the different approaches to field assessment of forest fire severity. In this paper, the authors present a site-specific assessment of large wildfires in Luhansk oblast, Ukraine occurred in 2020 using field-based burn severity indices. The Composite Burn Index (CBI) and the Geometrically Structured Composite Burn Index (GeoCBI) were used to estimate the extent of forest damage. The Burned Area Emergency Response (BAER) methodology was also tested to assess the extent of soil damage. The authors used PlanetScope images to delineate perimeters of burned areas. These perimeters were overlaid over a forest inventory database to extract forest attributes and site characteristics for all forested and unforested areas affected by fires. Within the fire perimeters, the burned area was stratified into six strata to independently account for forest damage in diverse types of land cover. In total 73 test plots were proportionally distributed among different classes of land cover to assess fire severity using CBI, GeoCBI, and BAER approaches. It was found that the fire’s footprints covered 39,782 hectares. Among that area, 21.2% were forested lands. About 78% of burned forests were pine plantations. The highest fire intensity levels were estimated within pure pine plantations that were grown in very dry sites, while the lowest ones were associated with hardwoods forests in moisture site conditions. The average estimates of fire severity using the field-based indices varied within strata (CBI>GeoCBI) which could be an issue for assessing burn severity using remote sensing-based approaches. The authors also concluded that the BAER methodology contributed less to assessing the fire intensity because soil burn severity is not directly related to vegetation damage. This work creates a foundation for further assessment of fire severity using satellite imagery. As a result of this study, a spatial data set of sample plots was proposed that can facilitate calibrating approaches used to map fire severity in the region

Predicting natural hyperdense regeneration after wildfires in Pinus halepensis (Mill.) forests using prefire site factors, forest structure and fire severity

Postfire Pinus halepensis (Aleppo pine) regeneration is often hyperdense. The overstocked stands created by this hyperdense regeneration considerably increase the risk of biotic and abiotic disturbances, especially fires, by increasing the potential for widespread forest losses. Our aim was to understand the relation between prefire site factors (climate, geographical position, topography, soil), prefire forest structure variables and fire severity with regeneration density after fire. We specifically wondered: (1) what are the general drivers of natural regeneration in these forests after fire?; (2) what are the necessary prefire conditions for establishing Aleppo pine hyperdense regenerations (>4,000 plants/ha)? To answer these questions, we sampled 147 plots in 15 wildfires located in the Comunitat Valenciana, which were representative of Aleppo pine Mediterranean forests. We used full and partial redundancy analyses (RDAs) for variance partitioning, and a decision tree analysis to look for the key site factors that drive regeneration density after fire. We found that all the site factors measured in the study explained 34.4 % of total variation in regeneration density. Prefire site factors and fire severity together explained 28.4 % of total variability, while the measured postfire factors explained only 7.5 %. Forest structure and climate explained 8.3 % and 6.7 % of variation, respectively. Five specific site factors drove regeneration density after fire: average minimum temperature, tree density before fire, resprouting shrubs coverage before fire, soil depth and bedrock type. The conclusions of this study were: (i) the average minimum temperature was the main significant variable that classified regeneration density and split data into three significant groups of Aleppo pine burned sites; (ii) the prefire forest structure (overstorey density and understorey coverage) controls regeneration density at colder burned sites, but soil depth and bedrock can be more important at warmer sites; (iii) fire severity relates positively to pine regeneration density, but negatively to resprouting vegetation coverage after fire; (iv) overstocked stands are not expected if prefire stand density is below 100 trees/ha at colder burned sites. These results may facilitate the planning of forest management and restoration actions because it may be used to identify those areas more likely to regenerate overstocked stands when faced with a changing fire regime.

Self-thinning forest understoreys reduce wildfire risk, even in a warming climate

As climatic changes continue to drive increases in the frequency and severity of forest fires, it is critical to understand all of the factors influencing the risk of forest fire. Using a spatial dataset of areas burnt over a 65 year period in a 528 343 ha study area, we examined three possible drivers of flammability dynamics. These were: that forests became more flammable as fine biomass (fuel) returned following disturbance (H1), that disturbance increased flammability by initiating dense understorey growth that later self-thinned (H2), and that climatic effects were more important than either of these internal dynamics (H3). We found that forests were unlikely to burn for a short ‘young’ period (5–7 years) following fire, very likely to burn as the regrowing understorey became taller and denser (regrowth period), then after a total post-disturbance period of 43–56 years (young + regrowth periods), fire became unlikely and continued to decrease in likelihood (mature period). This trend did not change as the climate warmed, although increases in synoptic variability (mean changes in synoptic systems per season) had a pronounced effect on wildfire likelihood overall. Young forest and regrowth forest became increasingly likely to burn in years of greater synoptic variability and the time taken for forests to mature increased, but in years with the most severe synoptic variability, mature forests were the least likely to burn. Our findings offer an explanation for fire behaviour in numerous long-term studies in diverse forest types globally and indicate that, even in the face of a warming climate, ‘ecologically-cooperative’ approaches may be employed that reinforce rather than disrupt natural ecological controls on forest fire. These range from traditional indigenous fire knowledge, to modern targeting of suppression resources to capitalise on the benefits of self-thinning, and minimise the extent of dense regrowth in the landscape.

Composition and diversity of tree species after fire disturbance in a lowland tropical forest in East Kalimantan, Indonesia

Abstract. Sakai A, Arbain, Sugiarto, Rahmawati K, Mirmanto E, Takahashi M, Ueda A. 2022. Composition and diversity of tree species after fire disturbance in a lowland tropical forest in East Kalimantan, Indonesia. Biodiversitas 23: 1576-1587. Lowland dipterocarp forests in East Kalimantan have suffered severe forest fires after prolonged drought at least twice in recent decades. We investigated species richness, stand structure and species composition in Sungai Wain Protection Forest using circle plots (each 0.04 ha) along multiple long transects (0.5 - 4.5 km long) from the edge to the interior of this forest. The impact of forest fires remained in the form of significantly smaller tree sizes in burnt areas, although species richness and the number of stems appeared to be recovering. NMDS (non-metric multidimensional scaling) ordination revealed that the species composition of burnt areas was still considerably different from that of intact forest. However, nearly 20 years have elapsed, suggesting that it takes a long time for species composition to recover, if at all. The species composition was affected by the basal area ratio of pioneer trees, the basal area ratio of dipterocarp trees and maximum diameter at breast height, in order of its strength. Distance from the edge of the Protection Forest also had a minor effect on species composition, suggesting that forest fires had damaged marginal areas of the forest.

Combustion of Aboveground Wood from Live Trees in Megafires, CA, USA

Biomass combustion is a major biogeochemical process, but uncertain in magnitude. We examined multiple levels of organization (twigs, branches, trees, stands, and landscapes) in large, severe forest fires to see how combustion rates for live aboveground woody parts varied with tree species, size, and fire severity in Ponderosa pine (Pinus ponderosa Dougl. ex Laws.) and mixed conifer-dominated forests of the Sierra Nevada, California, USA. In high severity fire patches, most combustion loss was from branches < 2 cm diameter; in low to moderate severity patches, most was from bole charring. Combustion rates decreased as fire severity declined and with increasing tree size. Pinus species had little branch combustion, leading them to have ≈50% the combustion rate of other taxa. Combustion rates could be 100% for small branch segments and up to 57% for small tree aboveground woody biomass in high severity fire patches. However, combustion rates are very low overall at the stand (0.1%–3.2%) and landscape level (0.6%–1.8%), because large trees with low combustion rates comprise the majority of biomass, and high severity fire patches are less than half of the area burned. Our findings of low live wood combustion rates have important implications for policies related to wildfire emissions and forest management.

Using geoinformatics to link forest fire severity and fragmentation in India's Dalma Wildlife Sanctuary

Using geoinformatics to link forest fire severity and fragmentation in India's Dalma Wildlife Sanctuary

Forest and Peatland Fire Severity Assessment at Siak Regency, Riau Province using Sentinel-2 Imagery

Siak Regency, Riau Province is one of the most forest and land fire-prone regencies in Indonesia. Most of the fires occur in peatland areas which contributes to the transboundary haze pollution in the region. Despite limited studies, fire severity assessment is an essential step in post-fire activities to estimate ecological impacts and economic impacts and law enforcement. This study aims to estimate fire severity using Sentinel-2 imagery at Siak Regency, Riau Province. The methods applied Normalized Burn Ratio on Sentinel-2 Imagery as an identification model based on reflectance value for 2019 imagery. The study revealed that burned areas in Siak Regency could be classified into four fire severity classes: low fire severity, moderate-low fire severity, moderate-high fire severity, and high fire severity. High fire severity was found mainly at Sungai Apit and Mempura Districts.

Assessing the severity of forest fire in Sungai Buluh Protected Peat Forest, Jambi

Jambi is one of the provinces that has a high level of forest fires in Indonesia. Sungai Buluh Peat Protected Forest is one of the protected peat forest areas in Jambi Province which experienced severe forest and land fires in 2019. Fire severity was estimated using the NBR Index from the Sentinel-2 satellite imagery data. The results showed that fire severity in the HLG Sungai Buluh in 2019 was dominated by high-class fire severity, covering an area of 207.9 ha. The NBR Index has a good ability to estimate the severity of forest fires with a D-value of 2.816 with an accuracy rate of 91.984%. The severity of the fires that occur is linked to the groundwater level dynamics which are strongly and significantly (p < 0.05) influenced by precipitation. Fires were detected in September when groundwater levels averaged -57.66 cm and very low amounts of precipitation (17.43 mm) occurred. It is important to maintain high groundwater levels, less than 40 cm below the surface, to reduce the risk of fire and prevent future peatland degradation.

Hotspot Analysis in 2015 and 2019 at Ogan Komering Ilir District South Sumatera Province

Ogan Komering Ilir (OKI) is one of the districts in South Sumatra Province that experience severe forest and land fires with the highest hotspots number during the period of 2015-2019. The objective of this study was to analyze the distribution of hotspots and their relationship with rainfall in 2015 and 2019 in the OKI district. The data sources used in this study included hotspot data derived from MODIS Terra/Aqua satellite imagery and rainfall data from BMKG. The results of this study indicate that the number of hotspots in the OKI District was 17,532 in 2015 and 4,583 in 2019. The distribution of hotspots from 2015 and 2019 was mostly found on peatlands. The highest number of hotspots in 2015 was in Tulung Selapan Sub-district and in 2019 it was in Cengal Sub-district 2019. The relationship between rainfall and hotspots number in 2015 and 2019 is not significant. The lowest rainfall in 2015 occurred in September with 0 mm/month and in 2019 the lowest rainfall occurred in August with 0,5 mm/month. Rainfall and hotspots number has an inverse correlation.

Class Activation Map-Based Data Augmentation for Satellite Smoke Scene Detection

With global warming, increasingly severe forest fires have caused serious damage to the Earth&#x2019;s ecological environment. Remote sensing (RS) technology has become an important means of forest fire monitoring owing to its unique advantages of multiple cycles and wide coverage. In smoke scene detection, the current simple mixed sample data augmentation (MSDA) methods lead to the loss of key objects in images, and training the samples whose images do not match the labels is the main reason for the decline in classifier accuracy. We propose a Class Activation Map-based Mixing method (CAMMix), a new MSDA method. In contrast to previous data augmentation (DA) methods, CAMMix can select the region and degree of mixing throughout the significance map. With the aid of an auxiliary classifier, CAMMix generates a mixed mask with class significance, such that the newly generated mixed sample distribution is close to the original data distribution. We also propose an intervention for the coverage method to further prevent the loss of smoke objects. On a smoke dataset (USTC&#x005F;SmokeRS), CAMMix achieves the best accuracy of 94.95% and 83% for 64% and 8% training set respectively. It results that CAMMix and variants provides high classification accuracy and preserves key object information, outperforming the Mixup, CutMix, ResizeMix and FMix.

Detection of forest fire burn-area using landsat-8 and sentinel-2: a case study of Nivale (Kolhapur) beat of Chandoli National Park, Maharashtra, India

Forests play an important role in maintaining environmental equilibrium in the ecosystem. Forest fires, which can occur for a variety of reasons, are the greatest threat to forests. In order to control forest fires, it is critical to assess the formation and behavioral characteristics of forest fires. Detecting fire areas and the severity is much easier with satellite images obtained with advancing technologies. The main objective of this study is to extract burn area from Landsat-8 and Sentinel-2 satellite images of 2017 using six vegetation indices. Remote sensing data was used to study a forest fire in the Nivale (Kolhapur) beat of Chandoli National Park of Maharashtra. The Department of Divisional Forest Officer (Wildlife)- Chandoli National Park at Karad (DDFO-CNP) provided reference data indicating that the fire had damaged 194 hectares of Scrub forest. In addition, forest fire areas were determined using an objectbased image classification technique. When the findings of the study are compared to the values obtained by DDFO- CNP, it is found that Sentinel-2's object-based analysis provided the highest accuracy with an overall accuracy of 84 % and 0.795 Kappa statistics. Landsat-8 image has 82 % overall accuracy and 0.765 Kappa value. The findings of Sentinel-2 and Landsat-8 spectral indices showed the Sentinel-2 had better results in all indices. Differenced Normalized Difference Vegetation Index (dNDVI) and Relative Difference Normalized Burn Ratio (RdNBR) performed better than other indexes with a difference of only 18.27 and 30.88 hectares respectively. According to the fire severity analysis, a burn area with high intensity in Sentinel-2 was identified as moderate-high in Landsat-8. As per the research findings, sentinel-2 had a high severity area of 28.72 ha and a low severity area of 37.04 ha. It shows that satellite images of Sentinel-2 are highly suitable than Landsat-8 for estimating scrub forest fire areas. Finally, the findings of this study could be useful to forest managers to monitor burn regions quickly after the fire and in reducing severity and frequency of forest fires.

RETRACTED: The Temporal-Spatial Distribution and Information-Diffusion-Based Risk Assessment of Forest Fires in China

As forest fires are becoming a recurrent and severe issue in China, their temporal-spatial information and risk assessment are crucial for forest fire prevention and reduction. Based on provincial-level forest fire data during 1998–2017, this study adopts principal component analysis, clustering analysis, and the information diffusion theory to estimate the temporal-spatial distribution and risk of forest fires in China. Viewed from temporality, China’s forest fires reveal a trend of increasing first and then decreasing. Viewed from spatiality, provinces characterized by high population density and high coverage density are seriously affected, while eastern coastal provinces with strong fire management capabilities or western provinces with a low forest coverage rate are slightly affected. Through the principal component analysis, Hunan (1.33), Guizhou (0.74), Guangxi (0.51), Heilongjiang (0.48), and Zhejiang (0.46) are found to rank in the top five for the severity of forest fires. Further, Hunan (1089), Guizhou (659), and Guanxi (416) are the top three in the expected number of general forest fires, Fujian (4.70), Inner Mongolia (4.60), and Heilongjiang (3.73) are the top three in the expected number of large forest fires, and Heilongjiang (59,290), Inner Mongolia (20,665), and Hunan (5816) are the top three in the expected area of the burnt forest.