, 张甲珅, 郑伟, 刘诚国家卫星气象中心,北京 100081
Comparative study on the emission estimation from the forest fire based on different resolution satellite data
GAOHao, ZHANGJiashen, ZHENGWei, LIUCheng收稿日期:2016-10-13
修回日期:2017-01-17
网络出版日期:2017-05-20
版权声明:2017《地理研究》编辑部《地理研究》编辑部
基金资助:
作者简介:
-->
展开
摘要
关键词：
Abstract
Keywords：
-->0
PDF (13447KB)元数据多维度评价相关文章收藏文章
本文引用格式导出EndNoteRisBibtex收藏本文-->
1 引言
火干扰是全球森林生态系统的重要干扰因子,森林火灾破坏自然生态系统的同时,火灾过程中生物质燃烧也是大气痕量气体和颗粒物的重要来源。林火燃烧释放大量的二氧化碳（CO2）、一氧化碳（CO）、甲烷（CH4）、非甲烷总烃（NMHC）、黑炭（BC）、有机碳（OC）、氮氧化物（NOx）、二氧化硫（SO2）和颗粒物（PM）[1]。它们改变大气成分的组成,对生物地球化学循环产生重要的影响,同时林火过程中释放的大量颗粒物又改变大气辐射收支,进一步对区域和全球气候变化产生影响。此外,林火释放的细颗粒物能通过影响空气质量对人体健康也产生显著的影响[2-3]。随着全球气候的变暖和降雨量的时空分布变化,森林和草原火灾的强度和频率都在加剧,其对全球环境变化产生重要影响。研究表明,自然野火排放的一氧化碳占全球的51%[4],而全球26%~73%的细小有机颗粒物来自于自然野火排放[5]。东北大兴安岭林区是我国森林火灾的多发区域,其碳排放量占到了全国排放量的近45%[6],其对我国的碳排放量的定量评估具有重要的影响。因此,对该区域的林火排放的科学准确的计量,对于正确认识森林火灾对我国碳排放的贡献,并进一步认识其对区域气候的影响有重要的意义。
为了准确地估算由生物质燃烧引起的排放,并进一步认识其对气候变化的影响,国内外在不同尺度上开展了生物质燃烧排放大气痕量气体和颗粒物的定量研究。在局地尺度上,针对特定的森林火灾通过实地测量的方法来定量计算林火的排放量[7]。在区域尺度上,一些研究通过统计资料、调查数据和实验测定来推算生物质燃烧的排放量[8-10],同时也有不少研究通过卫星遥感开展生物质燃烧的排放估算[11,12]。近年来随着遥感技术的发展,全球尺度、不同空间分辨率（1 km和500 m）和时间分辨率的火烧迹地产品被用于开展全球生物质燃烧排放的研究[13],并且通过“自下而上”的方法取得了不同时间和空间尺度的排放清单,该模型中排放量由火烧迹地面积、可燃物载量、燃烧因子、排放因子决定,排放因子和燃烧因子可在实验室测定,火烧迹地面积和可燃物载量可通过卫星遥感获取。Van Der Werf等基于该方法获取了1997-2009年全球森林、草原、农田和泥炭野火的排放清单,并广泛应用于各类化学传输模式[14]。Wiedinmyer等利用该方法计算了2005-2010年全球的生物质燃烧排放清单FINN v1 (Fire Inventory from NCAR version 1.0)用于大气化学模式应用[15]。Randerson等采用该排放模型计算了2001-2010年全球范围内野外小火的排放清单[16]。但是,针对火烧迹地的研究表明,低分辨率的传感器存在较大的局限性,中高分辨率传感器在区域和全球尺度的估算精度有较大提高[17]。究其原因主要是遥感尺度效应,空间分辨率是影响地表特征提取精度[18-19]、空间格局[20]和模拟结果[21]的重要因素,随着空间分辨率的降低,精度逐渐降低。在“自下而上”模型中排放量对其变量因子都比较敏感,排放量计算结果存在较大的不确定性[22],当前空间分辨率的尺度效应对排放估算的影响仍然没有明确的研究结论。
本文利用30 m、500 m和1 km空间分辨率的TM、MODIS和AVHRR遥感数据获取排放模型中的火烧迹地面积、燃烧因子、可燃物载量等因子,参考已有实验中测定排放因子,对2006年5月22日发生在内蒙古鄂伦春旗大兴安岭砍都河林场森林大火的排放进行定量估算,并进一步评估不同空间分辨率的卫星遥感数据对林火排放定量估算的影响,为深入认识现有的基于卫星遥感获取的生物质燃烧排放清单的不确定性提供参考。
2 研究方法与数据来源
2.1 研究区概况
2006年5月22日,位于大兴安岭的砍都河林场因雷击发生火灾,并发展成特大森林大火。此次森林大火于2006年6月3日被扑灭,火灾造成砍都河林场大范围的森林过火（图1）。砍都河林场地处大兴安岭林场,属寒温带大陆性季风气候。地带性植被为寒温带针叶林,以兴安落叶松为优势树种,同时有白桦,散生的黑桦、杨树和樟子松以及柞树。由于各种因子的相互作用,该处区域雷击火较多,为中国森林火灾的高发区域并且危害严重。因此,对该区域的大型森林火灾进行研究具有一定的代表性。
显示原图|下载原图ZIP|生成PPT图1研究区地理位置
-->Fig. 1The geographic location of the study area
-->
2.2 排放模型
本文采用“自下而上”的方法定量估算生物质燃烧的排放,该方法中的排放与火烧迹地面积、可燃物载量、燃烧因子和排放因子有关,其公式如下[15]：式中：E为生物质排放量;BA为火烧迹地面积（m2）;FL为可燃物载量（kg/m2）,本文为地上干物质生物量密度;CF为可燃物在燃烧中实际燃烧的比例,即燃烧因子（combustion factor）;EF为不同植被类型排放特定大气痕量气体和颗粒物的排放因子（g/kg）。
2.3 火烧迹地面积
本文对火烧迹地的提取,采用过火前后的归一化燃烧指数（normalized burned ratio,NBR）的差值（dNBR）,相对于植被指数,加入了短波红外波段的NBR对于地形、云阴影和水体与过火区的识别能力更高。提取火烧迹地时,为避免人机交互式方法经验阈值的主观性,采用自适应的阈值方法自动确定阈值[23],TM和MODIS确定的dNBR阈值分别为-0.08和-0.04。由于NOAA的AVHRR传感器缺少短波红外波段,因此针对AVHRR提取火烧迹地时,基于过火前后NDVI的差值采用自适应阈值方法自动确定阈值,其阈值为-0.14。NBR和dNBR的计算公式分别为：式中：
排放模型中的火烧迹地面积的准确性直接影响排放量的精度。考虑到林场的地表并不是完全被植被覆盖,本文在计算火烧迹地面积时,引入植被覆盖度对火烧迹地的面积进行订正,基于砍都河林场过火前5月20日的TM、MODIS和AVHRR卫星数据,采用像元二分模型获取每个像元的植被覆盖度（fractional vegetation cover,FVC）,与判识的火烧迹地面积相乘,只有覆盖植被的燃烧面积是有效面积,从而得到真实的火烧迹地面积。植被覆盖度根据像元二分模型获取[24],该模型广泛应用于计算植被覆盖度：
式中：FVC为植被覆盖度;NDVI为归一化植被指数;
2.4 地上可燃物载量
本文根据土地覆盖类型来给定过火林区的可燃物载量,基于卫星遥感数据通过决策树分类获取土地覆盖类型。首先,基于林火发生前期的TM多光谱数据,计算获取归一化植被指数（NDVI）,辅以数值高程数据（DEM）,以1
显示原图|下载原图ZIP|生成PPT图2砍都河林场地表覆盖分类
-->Fig. 2Land cover classification of Kanduhe timberland
-->
Tab. 1
表1
表1砍都河林场地表覆盖类别分类混淆矩阵（%）
Tab. 1Confusion matrix of land cover classification for Kanduhe timberland (%)
| 生产者精度 | 林地 | 疏林 | 草地 | 水体 | 总计 |
|---|---|---|---|---|---|
| 林地 | (2236/2307) 96.92 | (22/1789) 1.23 | (61/1452) 4.20 | (2/162) 1.23 | 2321 |
| 疏林 | (66/2307) 2.86 | (1760/1789) 98.38 | (59/1452) 4.06 | 0 | 1885 |
| 草地 | (5/2307) 0.22 | (7/1789) 0.39 | (1330/1452) 91.60 | (5/162) 3.09 | 1347 |
| 水体 | 0 | 0 | (2/1452) 0.14 | (155/162) 95.68 | 157 |
| 总计 | 2307 | 1789 | 1452 | 162 | 5710 |
| 总体精度 = (5481/5710) = 95.98% Kappa系数 = 0.9403 | |||||
新窗口打开
根据砍都河林场的地表土地覆盖分类结果,由每个像元的土地覆盖类别来给定地上可燃物的载量。研究区域的林地以落叶针叶林为主,参考已有的研究结果[1,9-11,15],给定了研究区的每种土地覆盖类别的可燃物载量（表2）。其中林地的可燃物载量参考北方落叶针叶林,疏林参考灌木和草地的可燃物载量,草地为草本植被可燃物载量。
Tab. 2
表2
表2砍都河林场地表覆盖类别可燃物载（kg/m2）
Tab. 2Fuel loadings assigned to landcover types for Kanduhe timberland (kg/m2)
| 地表覆盖类别 | 林地[9-11,15] | 疏林[1,9-11,15] | 草地[1,9-11,15] | 水体 |
|---|---|---|---|---|
| 可燃物载量 | 8.10 | 3.06 | 0.74 | 0 |
新窗口打开
2.5 燃烧因子
燃烧因子受到可燃物种类和可燃物的含水量的影响,计算燃烧因子时,研究认为燃烧因子与绿色植被所占的比例（PGREEN）有一定的函数关系[25]。PGREEN通常由NDVI计算得到：式中：NDVIt为林火发生月的NDVI值;NDVImin和NDVImax分别为研究区前一生长季中每个像元NDVI值的最大值和最小值。本文从美国地质调查局（USGS）下载2005年生长季（4-9月）基于TM的NDVI数据,2005年缺少月份的数据选择最临近年份的数据补齐,同时选取2005年生长季500 m分辨率的NDVI数据（MOD13A1）,以及基于AVHRR的1 km分辨率的NDVI产品,统计生长季的NDVImin和NDVImax并计算PRGEEN。
Ito和Penner通过实验的方式研究获得了燃烧因子与PGREEN的函数关系[26],该模型已在中国和东亚区域的生燃烧物质排放研究中广泛应用[11,27]。对于草本植被,当像元中林木覆盖度小于15%时,通常认为是草地,其燃烧因子为：
当像元中林木覆盖度大于15%并小于60%时,通常认为是疏林,其燃烧因子为：
对于林地,实验研究认为其燃烧因子不大于0.47。本文借鉴已有的研究结果[11,26],将林地的燃烧因子设定为0.23,计算得到草地的燃烧因子根据已有研究将其范围限定为大于0.44且小于0.99,疏林的燃烧因子范围限定为大于0.01且小于0.88。
2.6 排放因子
排放因子是林火燃烧过程中每千克干物质所释放的大气痕量气体和颗粒物的量,本文根据土地覆盖类型建立一套参数表,给定各个土地覆盖类别燃烧过程中释放各种大气痕量气体和颗粒物（如CO2、CO、CH4、BC、OC、PM2.5、NOx、SO2、NMHC等）的排放因子,该排放因子都来自于已发表的研究成果[1,11,14-15]。如果排放因子的数值不止一个,本文选取平均值用于林火排放的计算（表3）,水体的排放因子都设定为0。Tab. 3
表3
表3砍都河林场不同地表覆盖类别的排放因子（g/kg）[
Tab. 3Emission factors assigned to different land cover types for Kanduhe timberland (g/kg) [
| 地表覆盖类别 | CO2 | CO | CH4 | NMHC | NOx | SO2 | PM2.5 | BC | OC |
|---|---|---|---|---|---|---|---|---|---|
| 林地 | 1594.3 | 96.9 | 4.67 | 7.4 | 2.87 | 0.8 | 12.6 | 0.49 | 7.7 |
| 疏林 | 1626.4 | 72.7 | 2.92 | 4.78 | 3.0 | 0.53 | 6.94 | 0.51 | 5.55 |
| 草地 | 1650.2 | 72.3 | 2.53 | 4.45 | 3.86 | 0.51 | 6.54 | 0.44 | 3.07 |
新窗口打开
3 结果分析
3.1 火烧迹地面积
基于2006年6月28日林场林火发生后晴空的30 m、500 m和1 km分辨率的TM、MODIS和AVHRR卫星数据,分别提取了砍都河林场的火烧迹地,如图3所示,同时基于5月20日TM、MODIS和AVHRR林火发生前的NDVI数据计算的植被覆盖度数据,分别计算了2006年5月22日砍都河林场森林大火的实际火烧迹地面积（表4）。统计显示,基于三种不同尺度卫星数据的森林实际火烧迹地面积分别为6.36×104 hm2、6.10×104 hm2和5.59×104 hm2 ,随着卫星分辨率的降低,火烧迹地面积逐渐减小。不同空间分辨率的卫星获取的火烧迹地面积的差异为4.3%~13.8%。三种不同空间分辨率的卫星监测的火烧迹地差异主要发生在火烧迹地的边缘（图4）,主要原因是火烧迹地的监测能力受到卫星传感器像元分辨率的限制,例如500 m分辨率的MODIS能监测到最小火烧迹地面积为1.2 km2 [13],MODIS、AVHRR等（1 km分辨率）中低分辨率的卫星传感器对面积较小的燃烧斑块无法监测[16],高分辨率的卫星可以监测到较小的燃烧斑块。研究表明火烧迹地的漏判随着空间分辨率的降低而增大,当火烧迹地面积大于1200 hm2时,MODIS监测的火烧迹地面积平均误差小于30%,而相同的平均误差下的1 km和5 km空间分辨率的卫星监测的火烧迹地面积至少要分别大于1800 hm2和3600 hm2,此时MODIS的监测平均误差则小于15%[19]。文中引入植被覆盖度来获取实际的火烧迹地面积,基于TM、MODIS和AVHRR数据的植被覆盖度的平均值分别为0.44、0.47和0.47。虽然已有研究表明尺度变化不会对NDVI产生重大误差[28,29],并且在植被覆盖度较高或较低的区域可以忽略NDVI的尺度效应,得到较为准确的植被覆盖度[30],但基于不同尺度遥感数据计算的植被覆盖度仍存在差异,导致实际火烧迹地面积出现差异。
显示原图|下载原图ZIP|生成PPT图32006年6月28日砍都河林场过火后TM、MODIS和AVHRR卫星影像(a-c)和过火区判识(d-f)
-->Fig. 3Satellite imagery (a-c) and burned area (d-f) of TM, MODIS andAVHRR for Kanduhe timberland on 28th, June, 2006
-->
显示原图|下载原图ZIP|生成PPT图4不同空间分辨率卫星监测的火烧迹地空间分布差异
-->Fig. 4Spatial discrepancy of the burned scar monitoring from different resolution satellite
-->
Tab. 4
表4
表4砍都河林场火烧迹地面积统计（hm2）
Tab. 4Total burned scar of different resolutionsatellites for Kanduhe timberland (hm2)
| 卫星 | TM | MODIS | AVHRR |
|---|---|---|---|
| 火烧迹地面积 | 6.36×104 | 6.10×104 | 5.59×104 |
新窗口打开
3.2 林火排放量
根据排放模型,分别基于30 m、500 m和1 km分辨率的TM、MODIS和AVHRR计算了此次发生在内蒙古自治区鄂伦春旗大兴安岭砍都河林场森林大火释放的各类痕量气体和颗粒物的排放总量。统计显示（表5）,不同空间分辨率卫星估算的此次森林大火释放的二氧化碳（CO2）、一氧化碳（CO）、甲烷（CH4）、非甲烷总烃（NMHC）、氮氧化物（NOx）、二氧化硫（SO2）、可吸入颗粒物（PM2.5）、黑炭（BC）、有机碳（OC）的排放量分别为1.01×106~1.64×106 t、6.07×104~9.58×104 t、2.91×103~4.51×103 t、4.61×103~7.18×103 t、1.83×103~3.01×103 t、5.00×102~7.79×102 t、7.82×103~12.1×103 t、3.10×102 ~5.02×102 t、4.79×103~7.46×103 t。分别统计了不同分辨率卫星估算不同植被类型在砍都河林场森林大火中释放各类大气痕量气体和颗粒物排放量的贡献（表6）,结果显示不同分辨率卫星的估算结果有一定的差异,其中TM、MODIS和AVHRR卫星估算的林地在各类大气痕量气体和颗粒物排放量的贡献分别为83.4%~90.9%、94.5%~97.4%和95.1%~97.5%,疏林的平均贡献分别为5.9%~10.4%、1.5%~2.8%和1.6%~2.9%,草地的平均贡献分别为2.1%~6.3%、0.8%~2.7%和0.7%~2.0%。Tab. 5
表5
表52006年砍都河林场森林大火不同卫星估算的排放量（t）
Tab. 5Summary of fire emissions of different resolution satellites during forest fire in 2006 for Kanduhe timberland (t)
| 卫星 | CO2 | CO | CH4 | NMHC | NOx | SO2 | PM2.5 | BC | OC |
|---|---|---|---|---|---|---|---|---|---|
| TM | 1.64×106 | 9.58×104 | 4.51×103 | 7.18×103 | 3.01×103 | 7.79×102 | 1.21×104 | 5.02×102 | 7.46×103 |
| MODIS | 1.17×106 | 7.00×104 | 3.35×103 | 5.32×103 | 2.12×103 | 5.76×102 | 9.02×103 | 3.58×102 | 5.52×103 |
| AVHRR | 1.01×106 | 6.07×104 | 2.91×103 | 4.61×103 | 1.83×103 | 5.00×102 | 7.82×103 | 3.10×102 | 4.79×103 |
新窗口打开
Tab. 6
表6
表62006年砍都河林场森林大火不同植被类型对排放量的贡献（%）
Tab. 6Contribution of the different land cover types to emissions during forest fire in 2006 for Kanduhe timberland (%)
| 卫星 | 植被类型 | CO2 | CO | CH4 | NMHC | NOx | SO2 | PM2.5 | BC | OC |
|---|---|---|---|---|---|---|---|---|---|---|
| TM | 林地 | 85.4 | 88.5 | 90.7 | 89.8 | 83.4 | 89.9 | 90.9 | 85.5 | 90.3 |
| 疏林 | 10.4 | 7.7 | 6.7 | 6.8 | 10.3 | 6.9 | 5.9 | 10.4 | 7.6 | |
| 草地 | 4.2 | 3.8 | 2.6 | 3.3 | 6.3 | 3.2 | 3.2 | 4.1 | 2.1 | |
| MODIS | 林地 | 94.9 | 96.4 | 97.3 | 96.9 | 94.5 | 96.9 | 97.4 | 95.5 | 97.3 |
| 疏林 | 2.7 | 2.0 | 1.6 | 1.7 | 2.8 | 1.8 | 1.5 | 2.8 | 1.9 | |
| 草地 | 2.4 | 1.6 | 1.1 | 1.4 | 2.7 | 1.3 | 1.1 | 1.7 | 0.8 | |
| AVHRR | 林地 | 95.4 | 96.5 | 96.9 | 97.1 | 95.1 | 96.8 | 97.5 | 95.4 | 97.3 |
| 疏林 | 2.8 | 2.2 | 1.7 | 1.8 | 2.9 | 1.9 | 1.6 | 2.9 | 2.0 | |
| 草地 | 1.8 | 1.3 | 1.4 | 1.1 | 2.0 | 1.3 | 0.9 | 1.7 | 0.7 |
新窗口打开
通过对比显示,高分辨率卫星估算的林地的排放贡献相比较低分辨率卫星估算的各类大气痕量气体和颗粒物的排放贡献低。不同植被类型对林火排放的贡献差异主要由土地覆盖分类影响,研究表明随着高分辨率到低分辨率的土地覆盖分类,其空间格局基本一致,类别边缘区域的斑块数量会减少,面积较小或者分布分散的土地覆盖类别往往会损失信息,面积较大的优势类别存在信息夸大的趋势[31,32]。虽然本文在林火排放计算中为了保持结果的准确和一致性,没有对低分辨率的影像进行分类,而是基于TM的土地覆盖分类计算出低分辨率尺度网格内的类别比例计算林火排放,由于研究区域内林地为优势地类且连片分布,疏林和草地分布较少,随着尺度的变大,优势类别的比例逐渐增大,面积较小的类别比例有所减小,从而导致优势类别排放的贡献增高,非优势类别的排放贡献降低。
针对卫星估算火烧迹地面积的精度对排放量精度的影响,本文以此次大兴安岭林区特大林火排放的CO2总量为例,将自适应阈值提取的TM的火烧迹地面积和计算得到的CO2排放量当作真值,分别选取不同的阈值提取TM监测到的火烧迹地面积和CO2的排放量。定量分析火烧迹地面积误差对计算的CO2的排放量误差的影响显示,TM提取的火烧迹地面积误差1%时,导致基于排放模型计算的CO2的排放量误差为0.95%。
3.3 空间分辨率对林火排放量的影响
通过对比30 m、500 m和1 km分辨率的TM、MODIS和AVHRR卫星获取的2006年砍都河林场森林大火的排放量发现,空间分辨率对森林火灾释放的各类大气痕量气体和颗粒物排放量有较大的影响（表7）。当卫星空间分辨率从30 m（TM）降低到500 m（MODIS）,各类大气痕量气体和颗粒物排放量差异为25.5%~29.6%;当卫星空间分辨率从30 m（TM）降低到1 km（AVHRR）,各类大气痕量气体和颗粒物的排放量差异为35.4%~39.2%;当卫星空间分辨率从500 m（MODIS）降低到1 km（AVHRR）,各类大气痕量气体和颗粒物的排放量差异为13.1%~13.7%。Tab. 7
表7
表7砍都河林场2006年森林大火不同分辨率卫星估算排放量的差异（%）
Tab. 7Discrepancy of the total emissions among different resolution satellites duringforest fire in 2006 for Kanduhe timberland (%)
| 卫星 | CO2 | CO | CH4 | NMHC | NOx | SO2 | PM2.5 | BC | OC |
|---|---|---|---|---|---|---|---|---|---|
| TM vs MODIS | 28.6 | 26.9 | 25.7 | 25.9 | 29.6 | 26.1 | 25.5 | 28.7 | 26.0 |
| TM vs AVHRR | 38.4 | 36.6 | 35.5 | 35.8 | 39.2 | 35.8 | 35.4 | 38.2 | 35.8 |
| MODIS vs AVHRR | 13.7 | 13.3 | 13.1 | 13.3 | 13.7 | 13.2 | 13.3 | 13.4 | 13.2 |
新窗口打开
卫星空间分辨率对森林大火的排放量计算有比较大的影响,随着卫星空间分辨率的降低,基于排放模型得到的大气痕量气体和颗粒物的排放量也随之减少。一方面是不同尺度的传感器对火烧迹地的监测能力产生限制,低分辨率的传感器不能识别较小的燃烧斑块,随着空间分辨率的降低,火烧迹地的漏判增大,导致得到林火排放量偏小。另一方面引入植被覆盖度来获得真实的火烧迹地面积时,虽然研究区域的高植被覆盖度的特点可忽略NDVI的尺度效应,获得较准确植被覆盖度,但得到的植被覆盖度仍存在一定的差异。已有研究表明,基于低分辨率的影像得到的植被覆盖度要高于实际植被覆盖 度[30],因而导致计算的火烧迹地面积偏大,导致林火排放量偏大。分析基于TM、MODIS和AVHRR计算的PGEEN显示,低分辨率的数据得到的PGEEN大于高分辨率数据的PGEEN,根据燃烧因子的计算公式,从而引起计算得到的燃烧因子偏小,导致计算得到的林火排放量偏小。
4 结论与讨论
随着遥感技术的不断发展,基于卫星遥感的不同时间和空间分辨率的火烧迹地产品用于全球的生物质燃烧的排放研究,然而排放量的定量结果仍有一定的不确定性,尤其是不同空间分辨率的卫星的尺度效应的影响尚未可知。因此,本文基于不同空间分辨率（30 m、500 m和1 km）的TM、MODIS和AVHRR卫星遥感数据对2006年5月22日发生在内蒙古自治区鄂伦春旗大兴安岭砍都河林场森林大火释放的各类大气痕量气体和颗粒物的排放进行了定量估算,并评估卫星空间分辨率对排放量估算的定量影响。结果显示,不同空间分辨率的卫星监测的火烧迹地面积的差异为4.3%~13.8%,随着卫星空间分辨率的降低,火烧迹地面积有一定的减少。不同空间分辨率卫星估算的砍都河林场森林大火释放的二氧化碳（CO2）、一氧化碳（CO）、甲烷（CH4）、非甲烷总烃（NMHC）、氮氧化物（NOx）、二氧化硫（SO2）、可吸入颗粒物（PM2.5）、黑炭（BC）、有机碳（OC）的排放量分别为1.01×106~1.64×106 t、6.07×104~9.58×104 t、2.91×103~4.51×103 t、4.61×103~7.18×103 t、1.83×103~3.01×103 t、5.00×102~7.79×102 t、7.82×103~12.1×103 t、3.10×102~5.02×102 t、4.79×103~7.46×103 t。本文估算的大兴安岭林区砍都河林场的森林火灾的各类大气痕量气体和颗粒物的排放量与胡海清等研究结果的年平均排放量对比显示[9,10],含碳气体CO2、CO、CH4和NMHC的排放量较为一致。其中不同空间分辨率的卫星估算的林地、疏林和草地的贡献分别为83.4%~97.5%、1.5%~10.4%和0.7%~6.3%。研究表明：当卫星空间分辨率从30 m降低到500 m,各类大气痕量气体和颗粒物排放量差异为25.5%~29.6%,从30 m降低到1 km时的排放量差异为35.4%~39.2%,从500 m降低到1 km时的排放量差异为13.1%~13.7%。卫星分辨率从十米级到公里级,排放量存在约30%的差异,当从百米级到公里级,差异减小到约13%。在森林火灾的排放模型中,其影响因子包括过火迹地面积、可燃物载量、燃烧因子和排放因子,由于森林生态系统的空间异质性和复杂性,精确测定这些影响林火排放量计算的参数并不容易。对于小尺度的森林火灾,通过调查测量是可行的,对于大尺度的森林火灾的排放量的定量化,遥感的手段更加适合。然而面对海量的卫星数据,不同分辨率的卫星数据定量计算的差异多大,在大尺度的排放计算时,什么样分辨率的卫星数据更加合适,都是值得研究的问题。本文针对一次大型林火的研究显示,卫星空间分辨率的尺度效应对生物质燃烧释放各类大气痕量气体和颗粒物的估算影响十分明显,随着卫星空间分辨率的降低,估算的排放量也随着减少。主要原因是火烧迹地的监测能力受到卫星传感器像元分辨率的限制,火烧迹地的漏判随着空间分辨率的降低而增大,如500 m分辨率的MODIS火烧迹地产品检测的最小燃烧面积为1.2 km2[13],当火烧迹地面积大于500 hm2时,其误判和漏判小于10%,当火烧迹地面积大于3600 hm2时,MODIS的监测平均误差小于15%,而5 km分辨率的卫星监测平均误差则大于30%[19]。AVHRR对于过火面积较小的火烧迹地面积估算误差较大[33],中低分辨率的卫星传感器往往无法监测到火烧迹地边缘面积较小的燃烧斑块[16]。另一方面,已有研究表明基于低分辨率的遥感影像得到的植被覆盖度要大于实际的植被覆盖度[30],因而低分辨率的植被覆盖度会引起获得的实际火烧迹地面积偏大,导致林火排放量偏大。分析基于三种不同尺度数据源得到的PGEEN显示,计算的低分辨率PGEEN偏大,从而引起燃烧因子偏小,导致得到的林火排放量偏小。此外,随着空间分辨率的增大,土地覆盖边缘面积较小的类型损失,面积较大的优势类别存在信息夸大的趋势[31],因而林地的信息变大,由于林地可燃物载量比草地大,综合考虑林地、草地的燃烧因子,最终仍引起可燃物载量变大,导致林火排放偏大。由于尺度效应的复杂性,关于土地覆盖分类、可燃物载量和燃烧因子对林火排放的影响在本文中的分析仍然有所欠缺,在未来的工作中需要分别针对不同的因子开展更加深入的研究。
The authors have declared that no competing interests exist.
参考文献 原文顺序
文献年度倒序
文中引用次数倒序
被引期刊影响因子
| [1] | Fires contribute substantial emissions of trace gases and particles to the atmosphere. These emissions can impact air quality and even climate. We have developed a modeling framework to estimate the emissions from fires in North and parts of Central America (10-71 N and 55-175 W) by taking advantage of a combination of complementary satellite and ground-based data to refine estimates of fuel loadings. Various satellite drivers, including the MODIS Thermal Anomalies Product, the Global Land Cover Characteristics 2000 dataset, and the MODIS Vegetation Continuous Fields Product were used in conjunction with data mined from literature to determine fire location and timing, fuel loadings, and emission factors. Daily emissions of particulate matter and numerous trace gases from fires were estimated using this method for three years (2002-2004). Annual emission estimates differ by as much as a factor of 2 (CO emissions for North America ranged from 22.6 to 39.5 Tg yr ). Regional variations in emissions correspond to different fire seasons within the region. For example, the highest emissions from Central America and Mexico occur in the late spring whereas the highest emissions from the United States and Canada occur during the summer months. Comparisons of these results with other published estimates of CO emission estimates from fire show reasonable agreement, but substantial uncertainties remain in the estimation techniques. We suggest methods whereby future emissions models can reduce these uncertainties. |
| [2] | Increased concentrations of ozone and fine particulate matter (PM) since preindustrial times reflect increased emissions, but also contributions of past climate change. Here we use modeled concentrations from an ensemble of chemistry-climate models to estimate the global burden of anthropogenic outdoor air pollution on present-day premature human mortality, and the component of that burden attributable to past climate change. Using simulated concentrations for 2000 and 1850 and concentration-response functions (CRFs), we estimate that, at present, 47065000 (95% confidence interval, 14065000 to 90065000) premature respiratory deaths are associated globally and annually with anthropogenic ozone, and 2.1 (1.3 to 3.0) million deaths with anthropogenic PM-related cardiopulmonary diseases (93%) and lung cancer (7%). These estimates are smaller than ones from previous studies because we use modeled 1850 air pollution rather than a counterfactual low concentration, and because of different emissions. Uncertainty in CRFs contributes more to overall uncertainty than the spread of model results. Mortality attributed to the effects of past climate change on air quality is considerably smaller than the global burden: 1500 (-2065000 to 2765000) deaths yrdue to ozone and 2200 (-35065000 to 14065000) due to PM. The small multi-model means are coincidental, as there are larger ranges of results for individual models, reflected in the large uncertainties, with some models suggesting that past climate change has reduced air pollution mortality. |
| [3] | Beijing, the capital of China, is a densely populated city with poor air quality. The impact of high pollutant concentrations, in particular of aerosol particles, on human health is of major concern. The present study uses Aerosol Optical Depth (AOD) as proxy to estimate long-term PM, and subsequently estimates the premature mortality due to PM. We use the AOD from 2001 to 2012 from the Aerosol Robotic Network (AERONET) site in Beijing and the ground-based PMobservations from the US embassy in Beijing from 2010 to 2011, to establish a relationship between PMand AOD. By including the atmospheric boundary layer height and relative humidity in the comparative analysis, the correlation (R) increases from 0.28 to 0.62. We evaluate 12 years of PMdata for the Beijing central area using an estimated linear relationship with AOD, and calculate the yearly premature mortality by different diseases attributable to PM. The estimated average total mortality due to PMis about 6100 individuals yrfor the period 2001-2012 in the Beijing central area, and for the period 2010-2012 the per capita mortality for all ages due to PMis around 17.9 per 10 000 person-year, which underscores the urgent need for air pollution abatement. |
| [4] | In 2004, the Joint Research Centre (JRC) of the European Commission, the Netherlands Environmental Assessment Agency (MNP) and the Max Plank Institute for Chemistry (MPIC) started a project to create fast (bi-)annual updates of the EDGAR global emission inventory system, based on the more detailed previous version 3.2. Here, the key features of the Emission Database for Global Atmospheric Research, EDGAR 3 are first summarized, and then the compilation of recent global trends having a major influence on variables and the new 'Fast Track' approach to estimate recent emissions of greenhouse gases and air pollutants in 2000 at a country-specific level are described. Also provided is an overview of the approaches and data sources used for this EDGAR 3.2 Fast Track 2000 dataset, the different source sectors and the accuracies achieved, with a focus on anthropogenic sources of methane and nitrous oxide. Results of global emission trends for four air pollutants are also briefly addressed. Results for various sources and greenhouse gases at regional and national scales and on 11 degree grid have been made available on the EDGAR website. |
| [5] | |
| [6] | . 、CO and CH released from forest fires to be 8.96 Tg C/a、1.12 Tg C/a and 0.109 Tg C/a,of which the components of undergrowth and litter contributed 39 %,47 % and 40 %,respectively.For each province,the amounts of CO,CO and CH emission were dominantly determined by fire area.The contribution from Heilongjiang,Yunnan and Neimenggu provinces were more than 80*!% where fire occurred very frequently.CO and CH emissions from forest fire accounted for only 1.2 % and 0.35*!% of total national emissions,respectively.CO,CO and CH emissions from froest fire in China were 0.3 %,0.5 % and 0.01 % of global emission from forest fires. 、CO and CH released from forest fires to be 8.96 Tg C/a、1.12 Tg C/a and 0.109 Tg C/a,of which the components of undergrowth and litter contributed 39 %,47 % and 40 %,respectively.For each province,the amounts of CO,CO and CH emission were dominantly determined by fire area.The contribution from Heilongjiang,Yunnan and Neimenggu provinces were more than 80*!% where fire occurred very frequently.CO and CH emissions from forest fire accounted for only 1.2 % and 0.35*!% of total national emissions,respectively.CO,CO and CH emissions from froest fire in China were 0.3 %,0.5 % and 0.01 % of global emission from forest fires. |
| [7] | We used a ground-based approach to compute the pyrogenic carbon emissions from the Biscuit Fire, an exceptionally large wildfire, which in 2002 burned over 200,000 ha of mixed conifer forest in southwestern Oregon. A combination of federal inventory data and supplementary ground measurements afforded the estimation of preburn densities for 25 separate carbon pools at 180 independent locations in the burn area. Average combustion factors for each of these pools were then compiled from the postburn assessment of thousands of individual trees, shrubs, and parcels of surface and ground fuel. Combustion factors were highest for litter, duff, and foliage, lowest for live woody pools. Combustion factors also increased with burn severity as independently assessed from remote imagery, endorsing the use of such imagery in scaling emissions to fire area. We estimate the total pyrogenic carbon emissions from the Biscuit Fire to be between 3.5 and 4.4 Tg C (17 and 22 Mg C ha) depending on uncertainty in our ability to estimate preburn litter pools and mineral soil combustion with a central estimate of 3.8 Tg C (19 Mg C ha). We estimate that this flux is approximately 16 times the annual net ecosystem production of this landscape prior to the wildfire and may have reduced mean net biome production across the state of Oregon by nearly half in the year 2002. |
| [8] | Agricultural field burning plays an important role in atmospheric pollution and climate change. This work aims to develop a detailed emission inventory for agricultural burning in China with a high spatial and temporal resolution. Province-specific statistical data, distributed by the Chinese national government, and results from scientific literature were utilized to estimate the total emissions for the base year 2006. Emissions were allocated to a 1km grid and a 10-day interval by using the Moderate Resolution Imaging Spectroradiometer (MODIS) Thermal Anomalies/Fire product (MOD/MYD14A1). The estimated annual emission ranges, with a 90% confidence interval, are 68 (51–85)TgCO 2 yr 611 , 4 (2–7)TgCOyr 611 , 0.25 (0.08–0.46)TgCH 4 yr 611 , 2.2 (1.08–3.46)TgNMOCsyr 611 , 0.23 (0.08–0.41)TgNOxyr 611 , 0.09 (0.03–0.17)TgNH 3 yr 611 , 0.02 (0.01–0.03)TgSO 2 yr 611 , 0.03 (0.01–0.05)TgBCyr 611 , 0.1 (0.04–0.17)TgOCyr 611 , 0.27 (0.13–0.42)TgPM 2.5 yr 611 , 0.31 (0.12–0.53)TgPM 10 yr 611 . Provinces with the highest emissions are Anhui, Guizhou and Hunan. Spatially, agricultural fires are mostly located in the North China Plain, where the occurrence of fires is concentrated in early and late June (over 75% of the whole year) with another smaller peak in early October. This pattern corresponds with sowing and harvesting times for the main crops: wheat and maize. The temporal fire variation of two other agricultural zones in northeast China and south China are also detailed in our study. Our inventory, with a relatively high spatiotemporal resolution (1km grid and 10 days), could meet the need of global and regional air quality simulations. |
| [9] | . 火干扰是森林生态系统的重要干扰因子, 是导致植被和土壤碳储量发生变化的重要原因。火干扰所排放的含碳气体对气候变化具有重要的影响。科学有效地对森林火灾所排放的碳进行计量, 对了解区域和全球的碳平衡及碳循环具有重要的意义。根据大兴安岭森林资源调查数据和1965–2010年森林火灾统计资料, 利用地理信息系统GIS （geographic information system）技术, 通过野外火烧迹地调查与室内控制环境实验相结合的方法确定各种计量参数, 从林分水平上, 采用排放因子法, 估算了大兴安岭1965–2010年46年间森林火灾所排放的碳和含碳气体量。结果表明： 大兴安岭46年间森林火灾排放的碳为2.93 × 107t, 年平均排放量为6.38 × 105t, 约占全国年均森林火灾碳排放量的5.64%; 含碳气体CO2、CO、CH4和非甲烷烃（NMHC）的排放量分别为1.02 × 108、9.41 × 106、5.41 × 105和2.11 × 105t, 含碳气体CO2、CO、CH4和NMHC的年均排放量分别为2.22 × 106、2.05 × 105、1.18 × 104和4.59 × 103t, 分别占全国年均森林火灾各含碳气体排放量的5.46%、7.56%、10.54%和4.06%; 针阔混交林燃烧效率较低, 虽然火烧面积占总过火面积的21.23%, 但排放的碳只占总排放量的7.81%, 为此提出了相应的林火管理策略。 火干扰是森林生态系统的重要干扰因子, 是导致植被和土壤碳储量发生变化的重要原因。火干扰所排放的含碳气体对气候变化具有重要的影响。科学有效地对森林火灾所排放的碳进行计量, 对了解区域和全球的碳平衡及碳循环具有重要的意义。根据大兴安岭森林资源调查数据和1965–2010年森林火灾统计资料, 利用地理信息系统GIS （geographic information system）技术, 通过野外火烧迹地调查与室内控制环境实验相结合的方法确定各种计量参数, 从林分水平上, 采用排放因子法, 估算了大兴安岭1965–2010年46年间森林火灾所排放的碳和含碳气体量。结果表明： 大兴安岭46年间森林火灾排放的碳为2.93 × 107t, 年平均排放量为6.38 × 105t, 约占全国年均森林火灾碳排放量的5.64%; 含碳气体CO2、CO、CH4和非甲烷烃（NMHC）的排放量分别为1.02 × 108、9.41 × 106、5.41 × 105和2.11 × 105t, 含碳气体CO2、CO、CH4和NMHC的年均排放量分别为2.22 × 106、2.05 × 105、1.18 × 104和4.59 × 103t, 分别占全国年均森林火灾各含碳气体排放量的5.46%、7.56%、10.54%和4.06%; 针阔混交林燃烧效率较低, 虽然火烧面积占总过火面积的21.23%, 但排放的碳只占总排放量的7.81%, 为此提出了相应的林火管理策略。 |
| [10] | . |
| [11] | [1] Open fires such as those within forests and grasslands, as well as crop residue burning in fields, contribute considerable amounts of trace gases and particulate matter to the atmosphere and therefore play an important role in climate change and atmospheric chemistrv. Emissions from open fires in China are estimated at a medium resolution of 1 km |
| [12] | Biomass burning (BB) emissions from forest fires, agricultural waste burning, and peatland combustion contain large amounts of greenhouse gases (e.g., CO, CH, and NO), which significantly impact ecosystem productivity, global atmospheric chemistry, and climate change. With the help of recently released satellite products, biomass density based on satellite and observation data, and spatiotemporal variable combustion factors, this study developed a new high-resolution and multi-year emissions inventory for BB in Southeast Asia (SEA) during 2001-2010. The 1-km grid was effective for quantifying emissions from small-sized fires that were frequently misinterpreted by coarse grid data due to their large smoothed pixels. The average annual BB emissions in SEA during 2001-2010 were 27702Gg SO, 112502Gg NO, 55,38802Gg CO, 383102Gg NMVOC, 55302Gg NH, 32402Gg BC, 240602Gg OC, 383202Gg CH, 817,80902Gg CO, and 9902Gg NO. Emissions were high in western Myanmar, Northern Thailand, eastern Cambodia, northern Laos, and South Sumatra and South Kalimantan of Indonesia. Emissions from forest burning were the dominant contributor to the total emissions among all land types. The spatial pattern of BB emissions was consistent with that of the burned areas. In addition, BB emissions exhibited similar temporal trends from 2001 to 2010, with strong interannual and intraannual variability. Interannual and intraannual emission peaks were seen during 2004, 2007, 2010, and January-March and August-October, respectively. |
| [13] | We present an automated method for mapping burned areas using 500-m Moderate Resolution Imaging Spectroradiometer (MODIS) imagery coupled with 1-km MODIS active fire observations. The algorithm applies dynamic thresholds to composite imagery generated from a burn-sensitive vegetation index and a measure of temporal texture. Cumulative active fire maps are used to guide the selection of burned and unburned training samples. An accuracy assessment for three geographically diverse regions (central Siberia, the western United States, and southern Africa) was performed using high resolution burned area maps derived from Landsat imagery. Mapped burned areas were accurate to within approximately 10% in all regions except the high-tree-cover sub-region of southern Africa, where the MODIS burn maps underestimated the area burned by 41%. We estimate the minimum detectable burn size for reliable detection by our algorithm to be on the order of 120ha. |
| [14] | New burned area datasets and top-down constraints from atmospheric concentration measurements of pyrogenic gases have decreased the large uncertainty in fire emissions estimates. However, significant gaps remain in our understanding of the contribution of deforestation, savanna, forest, agricultural waste, and peat fires to total global fire emissions. Here we used a revised version of the Carnegie-Ames-Stanford-Approach (CASA) biogeochemical model and improved satellite-derived estimates of area burned, fire activity, and plant productivity to calculate fire emissions for the 1997-2009 period on a 0.5 spatial resolution with a monthly time step. For November 2000 onwards, estimates were based on burned area, active fire detections, and plant productivity from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor. For the partitioning we focused on the MODIS era. We used burned area estimates based on Tropical Rainfall Measuring Mission (TRMM) Visible and Infrared Scanner (VIRS) and Along-Track Scanning Radiometer (ATSR) active fire data prior to MODIS (1997-2000) and Advanced Very High Resolution Radiometer (AVHRR) derived estimates of plant productivity during the same period. Average global fire carbon emissions were 2.0 Pg yr-1 with significant interannual variability during 1997-2001 (2.8 Pg/yr in 1998 and 1.6 Pg/yr in 2001). Emissions during 2002-2007 were relatively constant (around 2.1 Pg/yr) before declining in 2008 (1.7 Pg/yr) and 2009 (1.5 Pg/yr) partly due to lower deforestation fire emissions in South America and tropical Asia. During 2002-2007, emissions were highly variable from year-to-year in many regions, including in boreal Asia, South America, and Indonesia, but these regional differences cancelled out at a global level. During the MODIS era (2001-2009), most fire carbon emissions were from fires in grasslands and savannas (44%) with smaller contributions from tropical deforestation and degradation fires (20%), woodland fires (mostly confined to the tropics, 16%), forest fires (mostly in the extratropics, 15%), agricultural waste burning (3%), and tropical peat fires (3%). The contribution from agricultural waste fires was likely a lower bound because our approach for measuring burned area could not detect all of these relatively small fires. For reduced trace gases such as CO and CH4, deforestation, degradation, and peat fires were more important contributors because of higher emissions of reduced trace gases per unit carbon combusted compared to savanna fires. Carbon emissions from tropical deforestation, degradation, and peatland fires were on average 0.5 Pg C/yr. The carbon emissions from these fires may not be balanced by regrowth following fire. Our results provide the first global assessment of the contribution of different sources to total global fire emissions for the past decade, and supply the community with an improved 13-year fire emissions time series. |
| [15] | . The Fire INventory from NCAR version 1.0 (FINNv1) provides daily, 1 km resolution, global estimates of the trace gas and particle emissions from open burning of biomass, which includes wildfire, agricultural fires, and prescribed burning and does not include biofuel use and trash burning. Emission factors used in the calculations have been updated with recent data, particularly for the non-methane organic compounds (NMOC). The resulting global annual NMOC emission estimates are as much as a factor of 5 greater than some prior estimates. Chemical speciation profiles, necessary to allocate the total NMOC emission estimates to lumped species for use by chemical transport models, are provided for three widely used chemical mechanisms: SAPRC99, GEOS-CHEM, and MOZART-4. Using these profiles, FINNv1 also provides global estimates of key organic compounds, including formaldehyde and methanol. Uncertainties in the emissions estimates arise from several of the method steps. The use of fire hot spots, assumed area burned, land cover maps, biomass consumption estimates, and emission factors all introduce error into the model estimates. The uncertainty in the FINNv1 emission estimates are about a factor of two; but, the global estimates agree reasonably well with other global inventories of biomass burning emissions for CO, CO, and other species with less variable emission factors. FINNv1 emission estimates have been developed specifically for modeling atmospheric chemistry and air quality in a consistent framework at scales from local to global. The product is unique because of the high temporal and spatial resolution, global coverage, and the number of species estimated. FINNv1 can be used for both hindcast and forecast or near-real time model applications and the results are being critically evaluated with models and observations whenever possible. |
| [16] | [1] In several biomes, including croplands, wooded savannas, and tropical forests, many small fires occur each year that are well below the detection limit of the current generation of global burned area products derived from moderate resolution surface reflectance imagery. Although these fires often generate thermal anomalies that can be detected by satellites, their contributions to burned area and carbon fluxes have not been systematically quantified across different regions and continents. Here we developed a preliminary method for combining 1-km thermal anomalies (active fires) and 500 m burned area observations from the Moderate Resolution Imaging Spectroradiometer (MODIS) to estimate the influence of these fires. In our approach, we calculated the number of active fires inside and outside of 500 m burn scars derived from reflectance data. We estimated small fire burned area by computing the difference normalized burn ratio (dNBR) for these two sets of active fires and then combining these observations with other information. In a final step, we used the Global Fire Emissions Database version 3 (GFED3) biogeochemical model to estimate the impact of these fires on biomass burning emissions. We found that the spatial distribution of active fires and 500 m burned areas were in close agreement in ecosystems that experience large fires, including savannas across southern Africa and Australia and boreal forests in North America and Eurasia. In other areas, however, we observed many active fires outside of burned area perimeters. Fire radiative power was lower for this class of active fires. Small fires substantially increased burned area in several continental-scale regions, including Equatorial Asia (157%), Central America (143%), and Southeast Asia (90%) during 2001 2010. Globally, accounting for small fires increased total burned area by approximately by 35%, from 345 Mha/yr to 464 Mha/yr. A formal quantification of uncertainties was not possible, but sensitivity analyses of key model parameters caused estimates of global burned area increases from small fires to vary between 24% and 54%. Biomass burning carbon emissions increased by 35% at a global scale when small fires were included in GFED3, from 1.9 Pg C/yr to 2.5 Pg C/yr. The contribution of tropical forest fires to year-to-year variability in carbon fluxes increased because small fires amplified emissions from Central America, South America and Southeast Asia egions where drought stress and burned area varied considerably from year to year in response to El Nino-Southern Oscillation and other climate modes. |
| [17] | . 森林火灾是加拿大北方森林地区最主要的扰动因素,对北方生态系统起着主导作用。基于MODIS数据,采用全球扰动指数算法(MGDI),对加拿大萨斯喀彻温省和亚伯达省2004-2011年的森林火烧迹地进行检测和分析。通过与MODIS标准火烧迹地产品以及加拿大林业局数据进行比较,扰动指数算法检测的火烧迹地面积比MODIS标准产品更接近于林业局的统计数据。分析表明,在2004-2011年间,由于火灾原因,整个研究区森林面积平均每年减少76.36万hm2,占该区域森林总面积的3.36%。萨斯喀彻温省平均每年燃烧的森林面积为46.83万hm2,亚伯达省为29.53万hm2。其中,2006、2008、2010和2011年是火灾的高峰年份。火烧迹地主要集中在生态交错带的北方保护区、针叶林保护区、针叶林平原区,以及北方平原东北部的伍德布法罗国家森林保护区。 森林火灾是加拿大北方森林地区最主要的扰动因素,对北方生态系统起着主导作用。基于MODIS数据,采用全球扰动指数算法(MGDI),对加拿大萨斯喀彻温省和亚伯达省2004-2011年的森林火烧迹地进行检测和分析。通过与MODIS标准火烧迹地产品以及加拿大林业局数据进行比较,扰动指数算法检测的火烧迹地面积比MODIS标准产品更接近于林业局的统计数据。分析表明,在2004-2011年间,由于火灾原因,整个研究区森林面积平均每年减少76.36万hm2,占该区域森林总面积的3.36%。萨斯喀彻温省平均每年燃烧的森林面积为46.83万hm2,亚伯达省为29.53万hm2。其中,2006、2008、2010和2011年是火灾的高峰年份。火烧迹地主要集中在生态交错带的北方保护区、针叶林保护区、针叶林平原区,以及北方平原东北部的伍德布法罗国家森林保护区。 |
| [18] | . |
| [19] | In this work, an empirical study was carried out to evaluate the impact of the spatial resolution of satellite images on the accuracy and uncertainty of burned area detection using classification techniques based on neuro-fuzzy (NF) models. The study area was situated in the northwest of the Iberian Peninsula, where in the summer of 2006, a large number of fires occurred, razing a surface area of more than 100,00002ha. A set of 12 zones containing a burned area in their central part were selected. Landsat Thematic Mapper (TM), Terra Moderate Resolution Imaging Spectroradiometer (MODIS), Advanced Very High Resolution Radiometer Local Area Coverage (AVHRR-LAC), and Advanced Very High Resolution Radiometer Land Long Term Data Record (AVHRR-LTDR) images with a spatial resolution of 30, 250, 110002m, and 0.05° (65500002m), respectively, obtained on 20 August 2006, were used. An NF classifier at pixel level for every image was constructed, taking into account only the spectrum bands (red and near-infrared (NIR)) common to all of them. The results in the study region suggest that burned areas of 65120002ha could be detected with a mean relative error less than 30% only in the MODIS image. In the case of the LAC and LTDR images, a minimum burned area size of >180002ha and >360002ha, respectively, is required to find similar errors. Burned areas greater >360002ha can be detected in MODIS imagery with a mean relative error of 6515%. A regression model of commission and omission error intervals compared with spatial resolution is presented. The conclusion is that in regard to the conditions of the study area, both error intervals increase symmetrically and linearly with the logarithm of the pixel size. The results also suggest that red and NIR spectrum bands could be used to detect burned area in post-fire images in Iberia, but with a relative error depending on burned area size for different spatial resolutions. |
| [20] | . 地理格局尺度依赖性的客观存在，要求在连续尺度序列上对区域植被覆盖空间变异性进行考察，以真实反映植被覆盖空间特征。以贵州喀斯特高原为例，借助地统计学和GIS软件，揭示了研究区NDVI的空间变异特征，并进行了空间变异与空间尺度的耦合研究。结论如下①NDVI空间变异程度表现出明显的尺度依存性，空间尺度的粗粒化对NDVI的平滑作用非常显著，但两种重采样方法对原始数据的粗粒化作用机制不同；②基于不同遥感数据源获得的NDVI数据之间空间格局差异明显，而且传统统计结果与地统计学结果明显不同，说明空间信息对数据间的差异性统计影响显著；③NDVI空间变异性呈现显著的各向异性，并表现出对遥感数据源的敏感性。 地理格局尺度依赖性的客观存在，要求在连续尺度序列上对区域植被覆盖空间变异性进行考察，以真实反映植被覆盖空间特征。以贵州喀斯特高原为例，借助地统计学和GIS软件，揭示了研究区NDVI的空间变异特征，并进行了空间变异与空间尺度的耦合研究。结论如下①NDVI空间变异程度表现出明显的尺度依存性，空间尺度的粗粒化对NDVI的平滑作用非常显著，但两种重采样方法对原始数据的粗粒化作用机制不同；②基于不同遥感数据源获得的NDVI数据之间空间格局差异明显，而且传统统计结果与地统计学结果明显不同，说明空间信息对数据间的差异性统计影响显著；③NDVI空间变异性呈现显著的各向异性，并表现出对遥感数据源的敏感性。 |
| [21] | . 利用模型分析气候变化对陆地生态系统功能的影响,是当前全球变化生态学的研究热点,然而模型模拟不确定性来源之一就是空间异质性的问题。空间异质性是尺度的函数,基于气象和遥感数据驱动的生态系统过程模型(BEPS模型),分别模拟2003-2005年中国生态系统通量观测与研究网络(China FLUX)长白山站、千烟洲站、海北站及当雄站在1 km和8 km空间分辨率下的总初级生产力(GPP)的时间动态变化,并结合土地覆盖类型及叶面积指数(LAI)的差异,探讨两种空间分辨率输入数据对GPP模拟结果的影响。结果表明:1差异性主要是由于8 km范围内混合像元导致LAI的不同,4个站点月均差异值分别为0.85、1.60、0.13及0.04;2两种空间分辨率均能较好地反映各站点GPP的季节动态变化,与GPP观测值的相关性R2为0.79~0.97(1 km)、0.69~0.97(8 km),月均差异值为11.46~29.65 g C/m2/month(1 km)、11.87~24.81g C/m2/month(8 km);3 4个通量站点在两种空间分辨率下的GPP月均差异值分别为14.43,12.05,4.79,3.22 g C/m2/month,不同空间分辨率的模拟结果在森林站的差异大于草地站,且生长季的差异大于非生长季。因此,模型在模拟大尺度、长时间序列GPP时,为了提高模型模拟效率,适度降低空间分辨率是可行的,但应尽量减小低空间分辨率对于森林生态系统以及生长季GPP模拟上的误差。 利用模型分析气候变化对陆地生态系统功能的影响,是当前全球变化生态学的研究热点,然而模型模拟不确定性来源之一就是空间异质性的问题。空间异质性是尺度的函数,基于气象和遥感数据驱动的生态系统过程模型(BEPS模型),分别模拟2003-2005年中国生态系统通量观测与研究网络(China FLUX)长白山站、千烟洲站、海北站及当雄站在1 km和8 km空间分辨率下的总初级生产力(GPP)的时间动态变化,并结合土地覆盖类型及叶面积指数(LAI)的差异,探讨两种空间分辨率输入数据对GPP模拟结果的影响。结果表明:1差异性主要是由于8 km范围内混合像元导致LAI的不同,4个站点月均差异值分别为0.85、1.60、0.13及0.04;2两种空间分辨率均能较好地反映各站点GPP的季节动态变化,与GPP观测值的相关性R2为0.79~0.97(1 km)、0.69~0.97(8 km),月均差异值为11.46~29.65 g C/m2/month(1 km)、11.87~24.81g C/m2/month(8 km);3 4个通量站点在两种空间分辨率下的GPP月均差异值分别为14.43,12.05,4.79,3.22 g C/m2/month,不同空间分辨率的模拟结果在森林站的差异大于草地站,且生长季的差异大于非生长季。因此,模型在模拟大尺度、长时间序列GPP时,为了提高模型模拟效率,适度降低空间分辨率是可行的,但应尽量减小低空间分辨率对于森林生态系统以及生长季GPP模拟上的误差。 |
| [22] | Biomass burning emission inventories serve as critical input for atmospheric chemical transport models that are used to understand the role of biomass fires in the chemical composition of the atmosphere, air quality, and the climate system. Significant progress has been achieved in the development of regional and global biomass burning emission inventories over the past decade using satellite remote sensing technology for fire detection and burned area mapping. However, agreement among biomass burning emission inventories is frequently poor. Furthermore, the uncertainties of the emission estimates are typically not well characterized, particularly at the spatio-temporal scales pertinent to regional air quality modeling. We present the Wildland Fire Emission Inventory (WFEI), a high resolution model for non-agricultural open biomass burning (hereafter referred to as wildland fires, WF) in the contiguous United States (CONUS). The model combines observations from the MODerate Resolution Imaging Spectroradiometer (MODIS) sensors on the Terra and Aqua satellites, meteorological analyses, fuel loading maps, an emission factor database, and fuel condition and fuel consumption models to estimate emissions from WF. WFEI was used to estimate emissions of CO (ECO) and PM2.5 (EPM2.5) for the western United States from 2003 2008. The uncertainties in the inventory estimates of ECO and EPM2.5 (uECO and uEPM2.5, respectively) have been explored across spatial and temporal scales relevant to regional and global modeling applications. In order to evaluate the uncertainty in our emission estimates across multiple scales we used a figure of merit, the half mass uncertainty, EX (where X = CO or PM2.5), defined such that for a given aggregation level 50% of total emissions occurred from elements with uEX EX. The sensitivity of the WFEI estimates of ECO and EPM2.5 to uncertainties in mapped fuel loading, fuel consumption, burned area and emission factors have also been examined. The estimated annual, domain wide ECO ranged from 436 Gg yr 1 in 2004 to 3107 Gg yr 1 in 2007. The extremes in estimated annual, domain wide EPM2.5 were 65 Gg yr 1 in 2004 and 454 Gg yr 1 in 2007. Annual WF emissions were a significant share of total emissions from non-WF sources (agriculture, dust, non-WF fire, fuel combustion, industrial processes, transportation, solvent, and miscellaneous) in the western United States as estimated in a national emission inventory. In the peak fire year of 2007, WF emissions were ~20% of total (WF + non-WF) CO emissions and ~39% of total PM2.5 emissions. During the months with the greatest fire activity, WF accounted for the majority of total CO and PM2.5 emitted across the study region. Uncertainties in annual, domain wide emissions was 28% to 51% for CO and 40% to 65% for PM2.5. Sensitivity of ECO and EPM2.5 to the emission model components depended on scale. At scales relevant to regional modeling applications (x = 10 km, t = 1 day) WFEI estimates |
| [23] | A threshold selection method from gray-level histograms OHTSU Nobuyuki IEEE Trans. Syst., Man, Cybern. SMC-9(1), 62-66, 1979 |
| [24] | The normalized difference vegetation index (NDVI) is the most widely used vegetation index for retrieval of vegetation canopy biophysical properties. Several studies have investigated the spatial scale dependencies of NDVI and the relationship between NDVI and fractional vegetation cover, but without any consensus on the two issues. The objectives of this paper are to analyze the spatial scale dependencies of NDVI and to analyze the relationship between NDVI and fractional vegetation cover at different resolutions based on linear spectral mixing models. Our results show strong spatial scale dependencies of NDVI over heterogeneous surfaces, indicating that NDVI values at different resolutions may not be comparable. The nonlinearity of NDVI over partially vegetated surfaces becomes prominent with darker soil backgrounds and with presence of shadow. Thus, the NDVI may not be suitable to infer vegetation fraction because of its nonlinearity and scale effects. We found that the scaled difference vegetation index (SDVI), a scale-invariant index based on linear spectral mixing of red and near-infrared reflectances, is a more suitable and robust approach for retrieval of vegetation fraction with remote sensing data, particularly over heterogeneous surfaces. The proposed method was validated with experimental field data, but further validation at the satellite level would be needed. |
| [25] | A modeling framework has been developed to examine the spatial and temporal aspects of biomass burning emissions from southern African savanna fires. The complexity of the fire emissions processes is described using a spatially and temporally explicit model that integrates recently published satellite-driven fuel load amounts, the GBA-2000 satellite burned area time series and empirically derived parameterizations of combustion completeness and emission factors (EFs). To represent fire behavior characteristics, land cover is classified into grasslands and woodlands using the MODIS percent tree cover product. The combustion completeness is modeled as a function of grass fuel moisture and the EFs as a function of grass fuel moisture in grasslands and fuel mixture in woodlands. Fuel moisture is derived from satellite vegetation index time series. The analysis at the regional scale shows that early burning in grasslands may lead to higher amounts of products of incomplete combustion, despite the lower amounts of fuel consumed, compared with late dry season burning. In contrast, early burning in woodlands results in lower emissions, in both products of complete and incomplete combustion, because less fuel is consumed than in the late dry season when the fuels are drier. Overall, burning in woodlands dominates the regional emission budgets. Emissions estimates for various atmospheric species, many of which are modeled for the first time, are reported. The modeled estimates for 2000 are (in Tg) 296 CO, 11.7 CO, 0.350 CH, 0.348 NMHC and 1.1 particulates (<2.5 m). Especially high is the previously undetermined contribution of oxygenated volatile organic compounds (0.915 Tg). A sensitivity analysis of fixed vs. seasonally variable EFs and combustion completeness demonstrates the importance of accounting for the seasonal variations of these two variables in emissions modeling. |
| [26] | This paper compares the results of emission estimates of trace gases from open vegetation fires in southern hemisphere Africa for the year 2000 using different data sets. The study employs several approaches, deriving carbon monoxide (CO) emissions from a variety of satellite information, measurement data sets, and empirically-based techniques to estimate burned areas (BA), fuel consumption (FC), and emission factors (EF). Three BA data sets are used: the Moderate Resolution Imaging Spectroradiomter (MODIS) burned area data set, the Global Burned Area data set for the year 2000 (GBA2000), and the Global Burn Scar Atlas (GLOBSCAR) in July and September, 2000. The estimated total BA in southern Africa varies significantly among data sets from 210,000 to 830,000 kmfor the sum of July and September. Temporal and spatial variations associated with CO emissions are analyzed using three different techniques for calculating the FC and EF. The first set of FC and EF extrapolates monthly variations in Zambia to southern Africa, the second extrapolates spatially resolved data for September to July, and the last includes monthly and spatial variations in both FC and EF. This analysis suggests the importance of accounting for the temporal and spatial variations in both FC and EF in order to determine the appropriate temporal and spatial variations of emissions from open vegetation fires. The CO emissions from open vegetation burning for the sum of July and September range from 18 to 31 Tg CO, using the MODIS BA data set and three different techniques for calculating FC and EF. The relative standard deviations (RSD) calculated from the three different methods are 58% for BA, 21% for FC, and 37% for EF. The best estimate of CO emissions from open biomass burning for the sum of the two months is 29 Tg CO, which may be compared to the estimates constrained by numerical models and measurements in 2000 which range from 22 to 39 Tg CO. |
| [27] | |
| [28] | does not introduce significant errors |
| [29] | . Biophysical inversion of remotely sensed data is constrained by the complexity of the remote sensing process. Variations in sensor response associated with solar and sensor geometries, surface directional reflectance, topography, atmospheric absorption and scattering, and sensor electrical-optical engineering interact in complex manners that are difficult to deconvolve and quantify in individual images or in time series of images. We have developed a model of the remote sensing process to allow systematic examination of these factors. The model is composed of three main components, including a ground scene model, an atmospheric model, and a sensor model, and may be used to simulate imagery produced by instruments such as the Landsat Thematic Mapper and the Advanced Very High Resolution Radiometer. Using this model, we examine the effect of subpixel variance in leaf area index (LAI) on relationships among LAI, the fraction of absorbed photosynthetically active radiation (FPAR), and the normalized difference vegetation index (NDVI). To do this, we use data from the first ISLSCP Field Experiment (FIFE) to parameterize ground scene properties within the model. Our results demonstrate interactions between sensor spatial resolution and spatial autocorrelation in ground scenes that produce a variety of effects in the relationship between both LAI and FPAR and NDVI. Specifically, sensor regularization, nonlinearity in the relationship between LAI and NDVI, and scaling the NDVI all influence the range, variance, and uncertainty associated with estimates of LAI and FPAR inverted from simulated NDVI data. These results have important implications for parameterization of land surface process models using biophysical variables such as LAI and FPAR estimated from remotely sensed data. |
| [30] | . 本文根据NDVI的定义及线性混合模型从数学形式上对NDVI的空间尺度问题进行了理论推导及分析，同时，提出了NDVI空间尺度问题纠正模型； 本文基于NDVI空间尺度效应提出了计算植被覆盖度的方法，考虑到实际象元不单纯由两种地物组成，所以对多于两种地物成份的混合象元进行了进一步推导，发现在一定空间分布下可以等同于两种地物混合的情况；同时对模型的敏感性从数学理论方面进行了分析； 本文通过模拟图像和实际图像对所提出方法进行了验证，结果表明，所提出的计算植被覆盖度方法在一定程度上可以消除植被覆盖度的尺度效应，从而得到更为准确的结果... 本文根据NDVI的定义及线性混合模型从数学形式上对NDVI的空间尺度问题进行了理论推导及分析，同时，提出了NDVI空间尺度问题纠正模型； 本文基于NDVI空间尺度效应提出了计算植被覆盖度的方法，考虑到实际象元不单纯由两种地物组成，所以对多于两种地物成份的混合象元进行了进一步推导，发现在一定空间分布下可以等同于两种地物混合的情况；同时对模型的敏感性从数学理论方面进行了分析； 本文通过模拟图像和实际图像对所提出方法进行了验证，结果表明，所提出的计算植被覆盖度方法在一定程度上可以消除植被覆盖度的尺度效应，从而得到更为准确的结果... |
| [31] | . 利用1km、4km和8km 3种空间分辨率的NOAA/AVHRR数字影像，对中国NECT样带西部地区进行了土地覆盖分类及其景观特征的比较研究。重点比较了几种空间分辨率遥感数据分类结果边界的一致性和空间差异，以及影像所记录的景观格局的差异。为进一步在不同尺度上研究景观变化过程以及尺度转换研究奠定了基础。研究表明：3种空间分辨率的遥感影像所反映的区域土地覆盖的宏观空间格局是一致的，但类型的边界、每一类型斑块的形状和数量均产生较大的差异；经过对反映景观空间结构的4种指标（分维数、破碎度、多样性、优势度）的比较显示出随着遥感影像空间分辨率的变化，影像所反映的景观结构发生了较大的变化。其中，各覆盖类型的分维数表现出最大差异，表征着空间分辨率的变化对斑块复杂程度的影响最大。 利用1km、4km和8km 3种空间分辨率的NOAA/AVHRR数字影像，对中国NECT样带西部地区进行了土地覆盖分类及其景观特征的比较研究。重点比较了几种空间分辨率遥感数据分类结果边界的一致性和空间差异，以及影像所记录的景观格局的差异。为进一步在不同尺度上研究景观变化过程以及尺度转换研究奠定了基础。研究表明：3种空间分辨率的遥感影像所反映的区域土地覆盖的宏观空间格局是一致的，但类型的边界、每一类型斑块的形状和数量均产生较大的差异；经过对反映景观空间结构的4种指标（分维数、破碎度、多样性、优势度）的比较显示出随着遥感影像空间分辨率的变化，影像所反映的景观结构发生了较大的变化。其中，各覆盖类型的分维数表现出最大差异，表征着空间分辨率的变化对斑块复杂程度的影响最大。 |
| [32] | . 土地覆盖是影响陆地生态系统机能的一个重要生物物理参量，作用于生物地球化学循环、水文过程以及地气交换，对于任何尺度的地球动力学研究都至关重要，是支撑多种科学研究的基础变量，可以为气候变化、地表能量交换、灾害监测、生态应用、农业估产以及自然资源利用等研究和应用提供直接的信息和相关参数。土地覆盖数据的质量直接影响各种相关研究和诸多应用结果的可靠性，但其获取需要花费很大代价，而且当前多数土地覆盖数据的质量都不尽如人意。遥感具有提供大范围内完整土地覆盖数据的潜能，而且，多时相的遥感影像可以对土地覆盖变化进行动态监测。因此，... 土地覆盖是影响陆地生态系统机能的一个重要生物物理参量，作用于生物地球化学循环、水文过程以及地气交换，对于任何尺度的地球动力学研究都至关重要，是支撑多种科学研究的基础变量，可以为气候变化、地表能量交换、灾害监测、生态应用、农业估产以及自然资源利用等研究和应用提供直接的信息和相关参数。土地覆盖数据的质量直接影响各种相关研究和诸多应用结果的可靠性，但其获取需要花费很大代价，而且当前多数土地覆盖数据的质量都不尽如人意。遥感具有提供大范围内完整土地覆盖数据的潜能，而且，多时相的遥感影像可以对土地覆盖变化进行动态监测。因此，... |
| [33] | . |
