基于高分一号影像光谱指数识别火烧迹地的决策树方法

森林火灾发生后,为及时、准确地掌握森林受灾情况,利用高分一号卫星(GF-1)16m宽幅影像各波段反射率信息,结合计算的归一化植被指数(NDVI)、过火区识别指数(BAI)、阴影植被指数(SVI)、归一化差异水体指数(NDWI)和全球环境监测指数(GEMI)等5种光谱指数,构建森林火烧迹地识别决策树模型(CART);在选取的研究区对该模型方法进行验证,并与最大似然监督分类法和非监督分类(ISODATA)方法所得到的结果精度进行了对比分析,结果表明:采用基于CART模型的决策树方法对火烧迹地识别结果精度较最大似然法总体分类精度提高了4.38%,Kappa系数提高了0.102 4,制图精度提高了14.96%,用户精度提高了8.50%;而采用ISODATA方法识别的火烧迹地的总体精度和Kappa系数都较低,制图精度和用户精度都没有达到1%。     

基于QUEST决策树的火烧迹地自动提取

以研究大尺度范围内森林火灾面积提取的可行性为目的。本文以昆明市为研究区,分别提取2005年和2006年两期Landsat TM影像的植被指数、水体指数、火烧迹地指数、纹理特征等,并选取适于大尺度范围的最佳特征波段;利用特征融合后的影像构建决策树,实现大尺度范围内火烧迹地的自动提取。研究表明,QUEST算法的分类总体精度达84.5%,在3种分类方法中精度最高;采用QUEST算法的决策树分类提取"3·29"火烧迹地面积占实际的97.9%,但大尺度范围内火烧迹地提取效果并不理想,仍需人工剔除提取产生的较小的斑块;最后用2015年"3·02"的火烧迹地验证了QUEST决策树方法的具有一定的普适性。该模式对大尺度范围内森林火灾面积统计的具有重要应用价值。     

基于EVI的大兴安岭火烧迹地植被恢复特征研究

[目的]选取Landsat数据源,基于dNBR和EVI指数开展火烧迹地识别和植被恢复特征研究.[方法]以1987年大兴安岭北部林区根河林业局金林林场森林火灾为研究背景,在地类划分的基础上,以dN-BR为基础数据,采用K-means方法识别并提取火烧迹地,并进行轻、中、重火烈度等级划分;基于火烧迹地的EVI(增强型植被指...     

基于规则面向对象分类法的贵州省山区火烧迹地提取

基于国产高分一号卫星宽幅（WFV）影像,对贵州省复杂地形山区火烧迹地的光谱特征、形状特征、纹理特征等进行了分析,建立了火烧迹地提取规则,完成了研究区基于规则面向对象的火烧迹地提取。利用高分二号卫星1 m分辨率影像对提取结果进行精度验证。结果表明,基于规则面向对象分类法的GF1WFV火烧迹地提取总体精度为92.67%,总Kappa系数为0.89,能较好地完成研究区的火烧迹地分类提取,分类质量达到极好水平,为贵州省山区火烧迹地提取提供了一定的参考和理论依据。     

呼中林区火烧迹地遥感提取及林火烈度的空间分析

【目的】利用Landsat TM影像,采用遥感指数构建决策树分类模型,提出一种识别火烧迹地面积与林火烈度分析的新方法,并结合坡度、坡向、海拔等地形因子对过火区域火烈度的空间分布进行科学系统的分析研究,为大兴安岭地区森林防火和林火管理提供一定的理论依据和数据支持。【方法】以大兴安岭地区呼中林区为研究区,以2010年9月火后TM影像以及2007年9月火前TM影像为基础数据,以DEM影像、林相图为辅助数据,利用NDVI、NDSWIR、MNDWI和dNBR等遥感指数构建决策树分类模型,对呼中林区2010年10场火烧迹地进行识别,根据dNBR阈值法将过火区域火烈度分为4级,并利用Arcgis软件将火烈度图分别与坡度、坡向、海拔图叠加分析。【结果】利用决策树分类模型所提取火烧迹地面积的分类总体精度和Kappa系数分别为97.97%和0.943 2,与平行六面体法和ISODATA法的分类的精度相比分别提高了7.56%和17.32%,Kappa系数也相应提高。决策树模型提取火烧迹地的制图精度和用户精度分别为97.51%和97.54%,而平行六面体分类法分别为90.43%和96.52%,ISODATA法分别为94.35%和95.68%。利用dNBR阈值法将已提取的过火区林火烈度分为:未过火、轻度火烧、中度火烧、重度火烧4个级别,其中中度火烧和重度火烧分别占总过火面积的46.6%和33.2%。叠加分析后,海拔在1 000~1 500 m的地区过火面积共4 177 hm~2,占总过火面积的64.4%。Ⅲ级坡(6°~15°)过火面积最大,占总过火面积的45.9%。南坡过火面积最大,为1 391 hm~2,约占总过火面积的21.4%。【结论】本文所使用的决策树分类模型能够准确地识别过火区域,在精度上相较平行六面体法与ISODATA法有显著提高,且过火面积也更接近目视解译判读所得到的过火面积,精度均达到82%以上。dNBR阈值法可将过火区域火烈度分为4个等级,结果表明过火区域中度火烧和重度火烧占总过火面积的比重较大,林火烈度与海拔、坡度、坡向之间存在一定相关关系。     

内蒙古毕拉河林场森林火灾受害程度遥感评价

为评价内蒙古毕拉河林场2017年5月2日发生的森林火灾损失情况,利用火灾发生前后的Landsat8卫星影像,计算得到两者间的差值归一化燃烧指数(d NBR),通过目视解译和数学统计相结合的方法构建森林火灾受害程度分级评价指标,对该火烧迹地受灾程度进行定量评价,并利用实地测量GPS数据和GF-2卫星数据验证受害程度的分级精度为86.39%。研究结果表明:内蒙古毕拉河林场火烧迹地的中度受害区域面积最大,为4 685.09hm~2,占火烧迹地总面积的40.35%;轻度受害面积次之,为4 213.1hm~2,占火烧迹地总面积的36.28%;重度受害面积为1 031.03hm~2,占总面积的8.88%;未受害区域的面积最小为906.57hm~2,占火烧迹地总面积的7.81%;火烧迹地内的受灾森林主要分布在中度受害区域和轻度受害区域,受灾草地主要分布在轻度受害区域。     

森林火灾火烧迹地遥感信息提取及应用

森林火灾是影响陆地生态系统、大气环境以及全球变化的重要因素之一。传统的森林火灾信息统计数据存在收集困难、难以空间化等方面的不足,而遥感技术的发展为火灾研究提供了一种新的技术手段。本文旨在对现有基于遥感的火烧迹地相关研究进行总结,从数据开发以及数据应用2个方面梳理现有研究成果,并对未来的研究提出展望,为森林火灾相关研究提供依据。在参阅近年来国内外相关文献的基础上,本文从2个方面对火烧迹地相关研究进行系统总结:1)现有火烧迹地数据产品集及其开发算法:从产品名称、遥感数据源、覆盖范围、时序范围、空间分辨率、时间分辨率等方面,对11种火烧迹地数据产品集进行介绍;随后,对不同数据产品的开发算法进行比较,包括基于热学异常的提取算法、基于光谱特征变化的提取算法以及两者相结合的提取算法;最后,对现有火烧迹地数据产品的精度特征进行总结。2)火烧迹地数据产品应用:对火烧迹地数据在火灾风险评估、大气污染物排放、全球植被动力学模型、火灾后生态系统监测4个方面的应用进行介绍。现有的火烧迹地数据产品还存在一定的不足,这些不足限制了其应用领域的扩展。未来的火烧迹地数据产品改进可以从以下3个方面展开:1)构建长时间序列火烧迹地数据产品集。2)开发高精度的火烧迹地数据产品集,充分考虑火灾发生时以及火灾发生前后的各种异常特性,真实反映火灾发生所带来的地表特征变化,才能建立更加精确、稳定的提取算法,提高火烧迹地数据产品精度。3)建立多样化的火烧迹地数据产品集,改良现有火烧迹地数据产品,增加新的监测内容,如火烧能量、火焰温度以及火灾规模等,这将进一步扩大火烧迹地数据产品的应用范围。     

遥感技术在火监测中的应用研究综述

火是地球表面频繁发生的异变现象,也是森林、灌丛、草地等生态系统中植被演替的主要动力。植被火烧释放出来的大量气溶胶颗粒物以及各种痕量气体,会引发大气环境和全球碳循环过程的深刻改变。遥感卫星能够监测和记录全球地表信息,尤其适用于地表异变现象特别强烈的火灾事件。遥感技术在野火识别和监测中的应用主要为了实现以下4个目的:1）火点精确地理位置判定;2）在燃火线的范围制图和火场发展趋势判定;3）火烧强度评价;4）火烧迹地植被恢复监测与评价。文中总结了遥感卫星技术和数据在火险评价与制图、火点识别、火烧面积统计以及火烧迹地植被恢复评价中的应用和贡献。     

不同火烧木管理方式火烧迹地林下物种多样性及地上生物量研究——以寒温带兴安落叶松林为例

以内蒙古大兴安岭兴安落叶松林火烧迹地为研究对象,采用分层取样的方法对群落中植被进行了调查,分析比较了不同火烧木管理方式下兴安落叶松林林下群落恢复过程中物种组成、多样性指数及地上生物量的差异。结果表明:火烧木择伐方式下火烧迹地与未火烧对照样地林下群落相似性最大。不同火烧木管理方式下,火烧迹地林下群落多个α多样性指数均表现为草本层灌木层的趋势。不同火烧木管理样地草本层Simpson和Shannon-Wiener指数均小于对照样地,而在灌木层都大于对照样地。3种不同类型火烧木管理方式火烧迹地恢复13年后,林下草本层生物量均小于对照样地,而灌木层生物量较对照样地相比都有不同程度的增加。     

青藏高原东北边缘云杉-冷杉林火烧迹地的灌木群落特征

【目的】探究青藏高原东北边缘云杉-冷杉林火烧迹地植被恢复与重建过程中的灌木群落特征动态，阐明火烧迹地灌木群落多样性格局和维持机制的生态过程，为在火烧迹地恢复与重建过程中对植被进行合理干预、尽快恢复生态效益提供合理依据。【方法】以空间代时间方法，分析恢复5年和15年火烧迹地灌木群落和未过火林地林下灌木层物种组成特征、α物种多样性和β物种多样性。【结果】在5年、15年火烧迹地灌木群落与未过火林地林下灌木层之间，植物种类组成存在差异，5年火烧迹地共调查到11科17属24种，15年火烧迹地共调查到8科10属17种，未过火林地共调查到12科19属35种。5年、15年火烧迹地和未过火林地的物种丰富度指数与Pielou均匀度指数，以及5年火烧迹地的物种丰富度指数与Simpson指数、物种丰富度指数与Shannon-Wiener指数的相关性均不强（P<0.1或P>0.1），其他指数间的相关性均较强（P<0.001或P<0.01）。物种丰富度指数为5年火烧迹地>15年火烧迹地>未过火林地，Shannon-Wiener指数和Simpson指数均为15年火烧迹地>未...     

两种热带森林覆盖变化检测方法比较研究

[目的]以热带森林复杂区域为对象,对两种缨帽变换“衍生数据”检测热带森林变化的方法的优劣进行比较.[方法]对两期数据进行缨帽变换,同时结合黑暗对象掩膜与局部直方图阈值提取等方法,获取亮度、绿度、湿度指数组合(MKT)和干扰指数(DI).采用MKT差值(MKT-D)、干扰指数差值(MDI-D)进行变化信息识别,然后,根据植被覆盖与亮度、绿度、湿度之间的变化关系,通过决策树分类提取变化信息,最后,对不同检测结果进行验证与比较分析.[结果]结果显示两种方法都能检测出森林内部的细微变化,但对小图斑变化,MKT-D检测优势明显,且MKT-D的kappa系数为0.763 0,MDI-D的kappa系数为0.655 9,两者相比,MKT-D方法优于MDI-D方法.[结论]MKT能够增强短波红外与近红外波段对森林变化信息的敏感性,有效地消除噪声等非目标信号,突出目标信号,此外,MKT-D为RGB彩色图像,更利于变化信息的目标提取与解译.     

Estimation of area burned by forest fires in Mediterranean countries: A remote sensing data mining perspective

Forest fires throughout the world result in tree mortality that can cause substantial timber and carbon losses. There is a critical need to map the areas burned by such fires to guide forest management decisions. Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery provides inexpensive and frequent coverage over large areas, facilitating forest health monitoring. In this study a MODIS post-fire image at a spatial resolution of 250 m serves as the starting point of an image mining based method. It involves three algorithms: modeling as a sum of Gaussian functions, kernel based smoothing, and adaptive thresholding. Adaptive thresholding serves as the reference to be compared to the image mining based method. Three spectral indices specifically designed for burned area identification have been used: the Burned Area Index (BAI), the Burned Area Index adapted to MODIS bands (BAIM), and the Normalized Burn Ratio (NBR). The κ statistic is applied to quantify the accuracy of the burned areas estimations by relating the estimated area with burned area perimeters measured on the ground by Global Positioning System (GPS). In addition, the κ statistic allows us to identify both the optimal spectral index and the optimal algorithms’ parameters. In this work, an accurate estimation (κ > 0.8) of areas burned by forest fires in Mediterranean countries is achieved, in particular if the BAIM index is used. The accuracy of these estimates is compared with the accuracy obtained by using the reference method by a McNemar’s test. Results show that our image mining based method allows a higher accuracy (the average increase of κ equals to 16%) than the reference method. We conclude that this method adequately maps burned areas, and that it may help management agencies to better understand of landscape-scale burn patterns.     

基于多源数据协同作业的森林信息提取研究进展

传统森林调查手段正逐渐被激光雷达、光学遥感及卫星遥感等新兴的遥感技术所代替。随着研究的不断深入,单一遥感数据源往往难以满足高精度的森林信息提取需求。因此,文中以激光雷达为研究主体,综述摄影测量、高光谱遥感及卫星遥感等多种遥感技术在森林信息提取中不同遥感技术协同作业的研究进展。相较于单一数据源,多源数据协同作业在森林结构参数提取、生物量估测以及树种识别等领域均有明显优势。结合目前的研究工作,从数据获取成本、获取方式、多源数据融合、配套软件及系统构建等方面展望了多源数据协同作业方式的发展趋势。     

无人机RGB影像中人工林单木位置的提取

从无人机RGB影像中提取单木位置时,由于树冠与非树冠植被的颜色相似,以及树冠之间存在粘连的问题,导致单木位置提取精度不高.针对这些问题,提出一种结合冠层高度模型(CHM)和形态学细化算法的人工林单木位置提取方法.首先根据无人机RGB影像生成数字正射影像(DOM)、数字高程模型(DEM)、数字表面模型(DSM),利用可见...     

Do logging,followed by burning,and wildfire differ in their decadal scale effects on tall open-forest bryophytes and vascular plants?

There has been only one time-controlled study, in Canada, comparing the occurrence of bryophyte species in forests regenerating after wildfire with that in those regenerating after logging and regeneration burning. Previous work on vascular plants in Tasmanian tall open-forest dominated by Eucalyptus showed that filmy ferns were less common in areas that had been logged and regeneration burned than in areas burned by wildfire two decades after the events, which suggested that hygrophilous species, such as bryophytes, might be vulnerable to this silvicultural system. Bryophytes, vascular plants structural and environmental data were collected from 50 sites, which had been burned in wildfires or clearfelled and burned by prescribed fires 31–39 years previously. Eighteen percent of the vascular plant species for which it was possible to develop a multiple regression model had logging/wildfire as a component, whereas the equivalent figure for bryophytes was 17%. The negative effects of logging were concentrated on the more hygrophilous species, and the positive effects were concentrated on the basal area of tree species and some of the mosses dependent on them. We conclude that wildfire and logging followed by regeneration burning result in vegetation differences that last more than three decades after disturbance, that these differences are no more pronounced for bryophytes than for vascular plants, and that hygrophilous taxa are favoured more by wildfire than logging.     

面向对象的决策树分类技术

以SPOT 5 HRG影像为遥感信息源,在湖南省资兴市天鹅山林场开展面向对象的决策树分类技术研究.利用最佳指数和联合熵2个指标确定最佳波段组合方式为1(R)4(G)3(B),在对遥感图像进行正射校正和融合处理的基础上,采用面向对象的决策树分类技术,以优势树种龄组作为分类对象.结果表明:优势树种龄组的分类精度达80%以上,取得了较高的分类精度与分类效果.     

基于TM影像的荒漠林光谱特征及提取模型研究

以TM为信息源,对TM多光谱遥感影像的辐射校正、几何校正、研究区实地样地的标定、区划,以及识别模型的提取验证,分析荒漠林树种在TM影像上的机理,分析荒漠林树种光谱特征,建立基于光谱知识的荒漠林树种信息提取的识别模型。结果表明:TM图像的红柳与梭梭的光谱特征明显的差异主要集中在TM4波段上,红柳在TM4的亮度值明显高于梭梭,即红柳亮度值具有在TM4最大的特征;经过对模型的验证,模型判对率为70%。     

Breeding bird community composition in different successional vegetation in the montane coniferous forests zone of Taiwan

To examine the relationship between forest succession following fire and the composition of bird communities, we investigated the vegetation structure, bird population density, foraging behavior and guild structure in bamboo grasslands (11 years since the last fire), pine savanna (41 years), pine woodland (58 years), old-growth hemlock forest (never burned), and old-growth spruce forest (never burned) in the Tatachia area of central Taiwan. Canopy height, total foliage cover, tree density, total basal area of tree, total basal area of snags, foliage height diversity, and tree species richness all increased with successional age. However, shrub cover peaked in intermediate successional stages. The vertical profile of foliage cover was more diverse in later successional forests, which had more breeding bird species and ecological guilds. All the breeding bird species recorded in early and intermediate stages were also found distributed in the late successional forests. Because Taiwan has high precipitation and humidity, and most forest fires in Taiwan are caused by human activities, forest fires and large areas of early successional vegetation were probably rare in the mountain areas of Taiwan prior to the arrival of humans. Therefore, bird species have not had enough time to adapt to areas with early or intermediate successional vegetation. Moreover, late successional forests host all the major plant species found in the early and intermediate stages and have higher foliage height diversity index, which was positively correlated with the bird species richness and bird species diversity index in this study. As a result, all breeding bird species and guilds in the area can be found in late successional forests. Efforts for conserving avian diversity in Taiwan should focus on protecting the remaining native old-growth forests.     

Estimation of stand attributes in <Emphasis Type="Italic">Cryptomeria japonica</Emphasis> and <Emphasis Type="Italic">Chamaecyparis obtusa</Emphasis> stands using QuickBird panchromatic data

This study aims to estimate stand density and stand volume in Cryptomeria japonica and Chamaecyparis obtusa stands from high-resolution satellite data and verify the reliability and uncertainty of the data. Sixty circular sample plots of 0.04 ha each were established. Their stand densities were estimated from the number of tree crowns derived from high-resolution satellite data using the watershed method. Stand densities derived from field surveys in the sample plots were compared with those obtained from high-resolution satellite data by stand age class. As a result, there was a positive correlation between them for sample plots of 41 years of age and over (R = 0.82); however, there was no correlation between them for sample plots of 40 years and under. Individual diameters at breast height (DBH) were estimated from crown areas obtained from high-resolution satellite data for the two species. Using the estimated DBH, individual tree heights were predicted from the height–diameter curves. Stand volumes were estimated from the sum of individual volumes, which were derived from volume formulas having two variables, i.e., DBH and height. Stand volumes derived from the field survey were compared with those obtained from high-resolution satellite data. The correlation coefficient between them for stands of 41 years of age and over was 0.78.