中南半岛旱季VIIRS活跃火的空间特征与国别差异

热带是全球活跃火(active fire)的集中发生区,客观认识其空间特征、国别差异及其动态变化对评估区域生物质燃烧及其碳排放等具有重要意义。作为热带季风气候典型区,中南半岛旱季活跃火发生发展空间特征及其动态变化仍缺乏清晰认识。为此,论文利用可见光红外成像辐射仪(VIIRS) S-NPP 2012—2019年活跃火矢量数据,基于核密度与空间自相关评价了中南半岛及国别旱季尤其是其特征月份(2—4月)活跃火发生发展的密集程度、集聚特征及其动态变化。结果表明：① 中南半岛活跃火核密度低值区占比最大(79%),高值区最小(4%);柬埔寨、缅甸、老挝等经济落后国家的核密度均值明显高于泰国和越南;2012—2019年核密度高值区具有朝高海拔、向内陆与趋边境等分布特征,且柬埔寨东北部长居高值区。② 活跃火核密度中值区变化集中在1—4月,且多分布在低、高值区周围;高值区变化集中在2—4月,由柬埔寨东北部逐渐向缅甸东/西部、泰国西北部以及老挝北/南部转移。③ 半岛与5国活跃火核密度在旱季具有显著空间正相关性,空间集聚类型以“高—高”型和“低—低”型集聚为主,越南、柬埔寨等国局部自相关性强于泰国和老挝。     

Estimating the spatial pattern of human-caused forest fires using a generalized linear mixed model with spatial autocorrelation in South Korea

Most forest fires in Korea are spatially concentrated in certain areas and are highly related to human activities. These site-specific characteristics of forest fires are analyzed by spatial regression analysis using the R-module generalized linear mixed model (GLMM), which can consider spatial autocorrelation. We examined the quantitative effect of topology, human accessibility, and forest cover without and with spatial autocorrelation. Under the assumption that slope, elevation, aspect, population density, distance from road, and forest cover are related to forest fire occurrence, the explanatory variables of each of these factors were prepared using a Geographic Information System-based process. First, we tried to test the influence of fixed effects on the occurrence of forest fires using a generalized linear model (GLM) with Poisson distribution. In addition, the overdispersion of the response data was also detected, and variogram analysis was performed using the standardized residuals of GLM. Second, GLMM was applied to consider the obvious residual autocorrelation structure. The fitted models were validated and compared using the multiple correlation and root mean square error (RMSE). Results showed that slope, elevation, aspect index, population density, and distance from road were significant factors capable of explaining the forest fire occurrence. Positive spatial autocorrelation was estimated up to a distance of 32 km. The kriging predictions based on GLMM were smoother than those of the GLM. Finally, a forest fire occurrence map was prepared using the results from both models. The fire risk decreases with increasing distance to areas with high population densities, and increasing elevation showed a suppressing effect on fire occurrence. Both variables are in accordance with the significance tests.     

三种活跃火遥感产品（MODIS C6、VIIRS V1和VIIRS J1）一致性与差异性对比

MODIS C6、VIIRS V1和VIIRS J1是NASA火灾信息资源管理系统（FIRMS）2020年首次同步发布的三套全球活跃火遥感产品。鉴于卫星过赤道时间、空间分辨率、火情监测算法有别,分析MODIS和VIIRS不同活跃火产品监测结果一致性和差异性,对明确不同火产品的适用性至关重要。利用中南半岛同年活跃火矢量产品,通过GIS渔网（1 km）定量分析不同时间尺度（全年、旱/雨季和昼/夜）活跃火发生的频次与强度特征,横向对比了三套活跃火产品的一致性与差异。结果表明：① 中南半岛三套产品逐月发生频次一致性最高,逐日次之,逐时最低。全年“有火区”格网占比37.23%,其中VIIRS相应比重三倍于MODIS,且J1略高于V1。三套产品两两比较时,VIIRS V1/VIIRS J1组合总体一致性最高,VIIRS J1/MODIS C6组合一致性最低。② 三套产品旱季“有火区”占比、活跃火发生频次与强度均显著高于雨季,且VIIRS V1与J1活跃火发生强度空间分布一致性更高。旱季V1活跃火累积发生频次最高、发生强度更强;雨季J1相应频次最高,发生强度较强。③ 三套产品白天“有火区”占比、活跃火累积发生频次均高于夜间,发生强度空间分布一致性亦高于夜间,其中VIIRS V1与J1夜间一致性最高。且昼/夜V1活跃火累积发生频次和“有火区”占比均最高,但J1发生强度最强。④ VIIRS V1与J1监测精度更高,空间覆盖范围更广,且VIIRS V1更适用于旱季、白天和夜间活跃火相关研究,而VIIRS J1对雨季活跃火的探测水平更高,但MODIS C6的优势在于历史数据丰富。     

Beyond slash-and-burn: The roles of human activities,altered hydrology and fuels in peat fires in Central Kalimantan,Indonesia

Near-annual landscape-scale fires in Indonesia's peatlands have caused severe air pollution, economic losses, and health impacts for millions of Southeast Asia residents. While the extent of fires across the peatland surface has been widely attributed to widespread peatland drainage for plantation agriculture, fires that transition from surface into sub-surface soil-based fires are the source of the most dangerous air pollution. Yet the mechanisms by which this transition occurs have rarely been considered, particularly in diversely managed landscapes. Integrating physical geography methods, including active fire scene evaluations and hydrological monitoring, with qualitative methods such as retrospective fire scene evaluations and semi-structured interviews, this article discusses how and why sub-surface peat fire transition occurs in an intensively altered peatland ecosystem in Indonesia's Central Kalimantan province. We demonstrate that variable water table levels and flammable surface vegetation (fire fuels) are co-produced socio-political and biophysical phenomena that enable the conditions in which surface fire is likely to transition into peat fire and increase landscape vulnerability to ongoing, uncontrollable annual fires. This localized understanding of peat fire transition counters normative causal narratives of tropical fire such as ‘slash-and-burn’, with implications for the management of new fire regimes in inhabited landscapes.     

Influences of forest roads on the spatial patterns of human- and lightning-caused wildfire ignitions

Understanding the spatial patterns of fire ignitions and fire sizes is essential for understanding fire regimes. Although previous studies have documented associations of human-caused fire ignitions with road corridors, less consideration has been given to understanding the multiple influences of roads on the fire regime at a broader landscape-scale. Therefore, we examined the difference between lightning- and human-caused fire ignitions in relation to forest road corridors and other anthropogenic and biophysical factors in the eastern Cascade Mountains of Washington State. We used geographical information systems and case-control logistic regression models to assess the relative importance of these explanatory variables that influence the locations of lightning versus human-caused ignitions.We found that human-caused ignitions were concentrated close to roads, in high road density areas, and near the wildland-urban interface (WUI). In contrast, lightning-caused ignitions were concentrated in low road density areas, away from WUI, and in low population density areas. Lightning-caused ignitions were also associated with fuels and climatic and topographic factors. A weak but significant relationship between lightning-caused fire and proximity to gravel roads may be related to fuels near roads or to bias in detection and reporting of lightning-caused fires near roads. Although most small fires occurred in roaded areas, they accounted for only a small proportion of the total burned area. In contrast, the large fires in roadless and wilderness areas accounted for most of the burned area. Thus, from the standpoint of the total area burned, the effect of forest roads on restricting fire size is likely greater than the impact of roads on increasing fire ignitions. The results of our study suggest that roads and their edge effect area should be more widely acknowledged as a unique type of landscape effect in fire research and management.     

ANALYSIS OF FIRE EVENTS AND CONTROLLING FACTORS IN EASTERN INDIA USING SPATIAL SCAN AND MULTIVARIATE STATISTICS

Vegetation fires are one of the most serious environmental problems in several ecoregions of India. The purpose of this study is to characterize spatio‐temporal characteristics of fire events in Orissa state, eastern India. In this study, ATSR satellite remote sensing data have been used to quantify fire events from 1997 to 2006. The spatial scan statistic that quantifies ‘hotspot’ areas of fire risk has been used to identify statistically significant fire clusters during the ten‐year time period. To assess the causative factors of fires, topographic, vegetation, climatic, anthropogenic and accessibility factors were used in a robust multivariate statistical framework. Results suggested a clear variation in hotspots of fire occurrences among districts. Of the several districts, the most likely cluster of fire events has been identified in Jharsuguda district followed by secondary clusters in Gajapati, Phulbani, Anugul, Debagarh, Balangir and Raygada. The first three principal components (PCs) from multivariate statistical analysis could explain 70.48% of variance in biophysical and socio‐economic indicators of fire events. The loadings were fairly large and highly positive for deciduous forest percentage, elevation, slope mean, aspect mean, and rural population density in the first PC explaining 40.21% of variance. The second PC included drought index, average temperature and illiteracy explaining nearly 19.2% variance. The third PC had a strong positive association with cropped area and forest cover explaining 10.98% of the total variance. Overall, by quantifying the disparities in fire events in space and time, this study demonstrates the utility of the spatial scan statistic in identifying priority areas of fire risk. Our results on fire hotspots and causative factors of risk can guide forest managers toward the best management strategies for avoiding damage to forests, human life, and personal property in the study area.     

Mapping fire regimes in China using MODIS active fire and burned area data

Relatively little is known about vegetation fire regimes in China. In this study, we investigated fire regime characteristics and their potential drivers, utilizing information extracted from the MODerate resolution Imaging Spectrometer (MODIS) satellite. Twelve fire regime variables were selected and computed on a regular grid over all of China, using MODIS burned area and active fire data during the period 2001 to 2016, to identify fire incidence and its inter-annual variability, seasonality, intensity, fire size distribution and vegetation types affected by fire. The variables were normalized and clustered to define six fire regimes with distinctive fire attributes. Results show that 78.6% of the land in China was affected by fire during the study period. The barren or sparsely vegetated lands of western China are nearly fire-free. Active fires were observed in Central China, but area burned was not detectable from MODIS. Forest fires in northeastern China are relatively large, infrequent, with a short fire season that peaks in non-winter seasons and higher inter-annual variability, implying a high likelihood of accidental causes. In contrast, forest fires in southern China are relatively small, frequent, with a long fire season that peaks in non-summer seasons, and lower inter-annual variability, suggesting regular use as a land management tool. Low inter-annual variability and low fire intensity were associated with cropland fires, whereas grassland fires generally exhibit the opposite traits. We have also discussed the potential drivers of each fire regime characteristics.     

呼伦贝尔草原火时空格局及特征分析

以Fire_CCI51为主要数据源,应用GIS空间分析技术对呼伦贝尔地区2001—2019年草原火的时空分布进行了研究,结果发现：呼伦贝尔草原火时间上具有年际波动较大、循环周期不定和春、秋两季双峰式集中的特点;空间上集中分布在东部森林草原、边境线附近,呈现出东多西少、北多南少的格局。利用相关分析方法对主要气象因子（平均降水量、最高气温、最小相对湿度、最大风速）与草原火之间的关系进行了深入分析,发现气象因子与草原火之间不存在显著相关;通过对呼伦贝尔市1981—2015年的草原起火原因统计数据分析发现,人类活动是导致草原火发生的主要因素。应用空间分析方法对边境线20 km缓冲区内的草原火进行了提取,发现境外过火面积远高于境内而且俄罗斯的过火面积远高于蒙古和中国。     

Spatial and temporal patterns of fires in tropical savannas of Indonesia

This paper examines spatial and temporal patterns of large to medium‐sized fires across the Indonesian Archipelago using remote sensing, particularly the MODIS fire products, over a 14 years period (2000–13 inclusive) in conjunction with climatic and land data in a GIS. Results showed fires (burned area and number of fires) were detected the most in year 2002 and least in 2010. The extent of burning was correlated with the annual Southern Oscillation Index with most burning occurring during times of sustained negative SOI values, which generally means drier conditions across the region. Most of the detected fires occurred in the middle of the dry season and in drier eastern Indonesia. Across Indonesia, approximately 1.5 per cent of available open vegetation area burned, whereas only 0.5 per cent of forest area burned. Most open vegetation burning was detected in the savanna with an area equivalent to 11.1 per cent of its extent being burned over the 14 years. On Sumba Island, where the pronounced dry season extends for at least four months, open vegetation is the most common land cover type, of which 99 per cent burned by area over the study period making it the island with the most burned areas in the Indonesian Archipelago.     

Monitoring deforestation with MODIS Active Fires in Neotropical dry forests: An analysis of local-scale assessments in Mexico,Brazil and Bolivia

The detection of vegetation fires using remote sensing has proven useful for highlighting areas undergoing rapid conversion in humid forests, but not in tropical dry forest (TDF). To further understand this relationship, we explored the correlation between MODIS Active Fires and forest cover change at local scales using 3 × 3 km sampling grids in three TDF landscapes in Bolivia; Mexico, and Brazil. Our analysis showed no single overall correlation among sites between the frequency of fires reported by the MODIS Active Fire Mapping product and forest cover change. Also, aggregated patterns of fire occurrence in Brazil and Bolivia did not correspond to areas with high percentage of forest loss, which indicates that the fire/deforestation relationship in TDF is not apparent in a simple fire frequency map. However, statistically significant correlations were found in sampling boxes with 50–60%, 50–70%, 50–95% forest cover at “initial state” of the time series in the Mexican site, Bolivian site and Brazilian site, respectively. Our findings suggest that complex interactions between anthropogenic fire-use, satellite-detected fires, and deforestation in highly fragmented TDF landscapes are difficult to describe at regional scales and might only be possible to analyze using finer resolution sampling grids.     

老挝VIIRS活跃火的主要自然地理要素特征

利用美国国家航空航天局火灾信息资源管理系统（FIRMS）VIIRS V1活跃火、先进星载热发射和反射辐射仪全球数字高程模型（ASTER GDEM）、MODIS NDVI/LSWI与土地覆被数据产品（FROM-GLC）,基于GIS定量分析老挝2012—2017年从分省到国家不同空间尺度活跃火频次的动态变化,并重点分析2015厄尔尼诺年活跃火频次及其与海拔、坡度、植被-水分（NDVI ^*、LSWI）指数、土地覆被等主要自然地理要素的相关特征。结果表明：① 老挝活跃火频次呈先增后减的趋势,峰值在2015厄尔尼诺年,主要集中在上、中寮,其中琅勃拉邦、沙耶武里与沙湾拿吉三省居前三位,而万象市、塞公和赛宋本则居后三位。② 老挝2012—2017年活跃火发生频次与地形因素（海拔、坡度）的关系基本相同,活跃火集中在海拔1000 m以下、坡度小于30°的低山-丘陵区。其中,上寮活跃火集中分布在500~1000 m、10~30°的斜陡坡山地,中、下寮集中在500 m以下、2~20°的缓斜坡丘陵。③ 活跃火高度集中在旱季,以3月、4月最多,并集中分布在NDVI为0.4~0.8和LSWI为0.2~0.6的中高覆盖度植被区。④ 老挝活跃火主要发生在森林这一土地覆被类型中,且以上寮最为集中,而中寮多以农田、草地和灌丛活跃火为主。综上,基于自然地理要素的活跃火特征分析可以有效识别其发生类型,即老挝活跃火主要由刀耕火种农业所引起。     

时空因素对中国城市火灾态势变化的影响

基于空间动态面板数据模型,对2000～2009年中国市域城市火灾数据进行分析,探讨经济发展与气候变化共同作用下时空因素对中国城市火灾态势变化的宏观影响。结果显示:气候变干促使火灾恶化,而经济发展扭转了这种趋势并促使火灾态势整体改善。时空因素对中国城市火灾态势变化具有显著影响,并可被引申为火灾同化效应、火灾惯性效应、火灾警示效应。火灾安全管理部门应该充分利用这些效应,采取积极措施改善火灾态势。本文补充了城市地理学在城市火灾领域的研究不足;将抽象的时空因素引申为具有实际物理含义的虚拟变量,对相关研究具有借鉴意义。     

Spatial Ecology of Pre –Euro‐American Fires in a Southern Rocky Mountain Subalpine Forest Landscape

Natural disturbances such as fires have been widely studied, but less is known about their spatial ecology than about other aspects of them. We reconstructed and mapped pre–Euro‐American fire history in a subalpine forest landscape in southeastern Wyoming, and analyzed the fires using GIS. Mean fire interval varies little with topography (elevation, aspect, slope) and is spatially autocorrelated at distances of at least 2 km. Fires often spread downslope, and spread more than expected from the north and south and less than expected from the west, under the influence of particular synoptic climatic conditions. The landscape of 1868 a.d., at the time of Euro‐American settlement, was strongly influenced by fires. However, it contained large patches of connected forest and few high‐contrast edges, unlike the modern landscape, which is fragmented by industrial forestry and roads. The spatial ecology of the natural fire regime may be a useful guide for management.     

Spatial patterns and drivers of fire occurrence in a Mediterranean environment: a case study of southern Croatia

Wildfires are an important factor of landscape dynamics in fire-prone environments of the world. In the Mediterranean, one of the most fire-susceptible environments globally, between 45,000 and 50,000 wildfires are recorded every year, causing disturbances in forest and grassland ecosystems. As a Mediterranean country, Croatia faces these problems, averaging over 1000 registered wildfires annually, with the coastal areas dominated by forest fires and continental Croatia by fires on agricultural lands. This research combines various landscape and socio-economic factors in the analysis of fire occurrence in Croatia’s southernmost region of Dalmatia. Around 275 of the largest fires (encompassing 98% of the total burnt area) registered in 2013 were investigated using OLS, and different spatial indices were employed to analyse regional variability in fire distribution. The results revealed that areas more prone to fires are the northern inland areas of Dalmatia and its entire coastal zone. Altitude and vegetation type demonstrated a correlation with fire occurrence, but an increase in population in the study area was also correlated with wildfire occurrence. Regarding vegetation, the grasslands and Mediterranean shrubland (maquis) were found to be the most fire-prone vegetation types in the study region, the distribution of which can be linked to different socio-economic and demographic processes occurring in the Eastern Adriatic.     

2001—2018年印度尼西亚MODIS活跃火的发生特征与响应

基于印度尼西亚2001—2018年FIRMS MODIS活跃火（Active Fire）位置数据,通过GIS渔网（1 km×1 km）定量分析了该国活跃火的发生概率、强度及其对厄尔尼诺的响应。结果表明：① 印度尼西亚18年间至少发生过一次活跃火的区域（即“有火区”）占22.79%,五大岛屿“有火区”以加里曼丹岛最多,巴布亚岛最少。活跃火发生以低概率为主,集中在8—10月,加里曼丹岛和苏门答腊岛分布最多;中概率分布较少,苏门答腊岛东部低地平原和加里曼丹岛南部及西南沿岸最为集中;高概率零星分布,大部分集中在爪哇岛和苏拉威西岛。② 活跃火的发生强度以1级（1次/a）和2级（2~3次/a）为主,3级（4~7次/a）次之。1~5级（1~22次/a）活跃火主要集中在加里曼丹岛和苏门答腊岛,6级（23~165次/a）活跃火集中在爪哇岛、苏门答腊岛和苏拉威西岛,且强度从中心向四周递减。此外,不同概率组内活跃火的发生强度也不同,低、中、高概率活跃火分别以1次/a、1~3次/a、2~7次/a为主。③ 印度尼西亚8—10月活跃火对强、弱厄尔尼诺的敏感度不同,其响应表现在活跃火频次与影响面积显著增加,厄尔尼诺越强,活跃火响应程度越高。其中,苏门答腊岛响应程度最高,巴布亚岛最低。     

中国东南海域活跃火强度时空动态与发生类型

自然和人为因素引发的植被火是全球生态系统的重要干扰因子,而基于卫星获取的活跃火（active fires）是探究人-火关系与活跃火发生机理、扩展机制的基础。相对于陆域活跃火,海上活跃火发生的时空动态及其强度、类型等研究较为薄弱。基于VIIRS活跃火和海洋地形等数据,通过分析中国东南海域活跃火发生的时间规律、强度特征、地形属性和距海岸线距离等特性,初步确定了海上活跃火类型界定体系。结果表明：(1)本区相同位置活跃火发生具有时序性。2015年前后两年内相同位置活跃火逐渐减少,且低纬地区更突出;而2015年前后三年内低纬地区活跃火发生强度稳定、高纬地区逐渐减少。(2)中国东南海域活跃火呈现明显的空间集-疏特征,即沿岸海域“条带状”聚集区、以曾母暗沙为中心的“团状”聚集区、南海诸岛周围的分散区。(3) 2012—2013年、2015年、2020年活跃火总强度与其月均强度变化一致,即其发生强度增强归因于活跃火的面状、分散发生;2014年、2016—2019年活跃火总强度与月均强度峰值不同步,即其强度增强归因于活跃火的点状、集中发生。(4)基于时空、强度和集聚特征,结合人类活动频繁区布局,将发生时...     

Human-ignited wildfire patterns and responses to policy shifts

Development of efficient forest wildfire policies requires an understanding of the underlying reasons behind forest fire occurrences. Globally, there is a close relationship between forest wildfires and human activities; most wildfires are human events due to negligence (e.g., agricultural burning escapes) and deliberate actions (e.g., vandalism, pyromania, revenge, land use change attempts). We model the risk of wildfire as a function of the spatial pattern of urban development and the abandonment/intensity of agricultural and forestry activities, while controlling for biophysical and climatic factors. We use a count data approach to model deliberately set fires in Galicia, N.W. Spain, where wildfire is a significant threat to forest ecosystems, with nearly 100,000 wildfires recorded during a thirteen-year period (1999–2011). The spatial units of analysis are more than 3600 parishes. Data for the human influences are derived from fine-resolution maps of wildland–urban interface (WUI), housing spatial arrangements, road density, forest ownership, and vegetation type. We found wildfire risk to be higher where there are human populations and development/urbanisation pressure, as well as in unattended forest areas due to both rural exodus and a fragmented forest ownership structure that complicates the profitability of forestry practices. To better help direct management efforts, parameter estimates from our model were used to predict wildfire counts under alternative scenarios that account for variation across space on future land-use conditions. Policies that incentivize cooperative forest management and that constrain urban development in wildlands at hotspot fire locations are shown to reduce wildfire risk. Our results highlight the need for spatially targeted fire management strategies.     

Livelihoods, fire and policy in eastern Indonesia

Indonesian legislation calls for a zero-burning policy. This approach to fire management is largely in response to significant negative impacts on the economy and the environment, not only in Indonesia but also the neighbouring region, that result annually from peat fires in Kalimantan and Sumatra. In this context, the present paper investigates the local use and management of fire in Flores and Sumba islands in eastern Indonesia. Our appraisals show that people's livelihoods depend on fire to maintain grasslands and, therefore, that the national policy and legislation for zero-burning is inappropriate and needs to be revised. This follows from the fact that not all fires cause damage and are unwanted. Through a series of rapid rural appraisal interviews, we found that the fires in grasslands are often lit intentionally to maintain the grasslands that local people use to sustain a variety of livelihood activities such as cattle rearing, hunting and farming. Although fires can damage or destroy remnant dry forests in eastern Indonesia, in order to be effective, future policy formulations need to account for this human livelihood dimension and the geographic variation in fuels, climate and land use.     

Estimating the short-term impact of fire on rangeland productivity in a semi-arid climate of South Africa

The short-term (2 years) influence of fire burning with the wind (head fire) and against the wind (back fire) on the productivity (above-ground phytomass and litter) and fire behaviour were determined over seven growing seasons (1995/96–2001/02) in a semi-arid rangeland. The results showed that head fires had significantly greater flame height and rate of spread than back fires. Fire caused a decrease of 40.5% and 22.5% in basal cover for the first and second seasons after burning respectively. The seasonal above-ground phytomass production and litter were significantly decreased by fire over all growing seasons. Seasonal production losses due to fire varied between 225 and 430 kg ha⁻¹. The relation between above-ground phytomass production loss due to fire, and two independent variables namely seasonal rainfall and fuel load (above-ground phytomass plus litter) before burning, were investigated. The multiple linear regression equations obtained for one (r=0.89) and 2 years (r=0.86) after a fire, can be used with great success in estimating seasonal production losses due to fire in semi-arid rangeland.