基于Sentinel-2的丛生竹林信息提取方法比较及分布特征

滇西南地区拥有丰富的丛生竹林景观和珍稀特有竹种资源,但竹资源分布储量不清、监测技术缺乏等问题很大程度限制了竹资源开发与利用。基于Sentinel-2A影像数据,采用反向传播神经网络、支持向量机、随机森林三种机器学习分类方法进行沧源县丛生竹林信息提取及精度评价,利用Google Earth影像和DEM数据对竹资源分布的空间和地形特征进行了分析。结果表明,随机森林分类精度优于支持向量机和反向传播神经网络,分类总体精度达90%,Kappa系数达0.87,竹林用户精度达81%。沧源县共有竹林138.07 km2,主要分布于城镇村庄、道路、水系和耕地周边,以四旁竹和防护竹林为主,采用Sentinel-2A10 m的分辨率很好地提取了空间上分布分散的丛生竹林。沧源县竹林主要分布在海拔900~2000 m,坡度范围大都位于缓坡和斜坡。研究结果可为沧源县竹资源开发利用提供数据支持,研究方法可作为大型丛生竹遥感监测的参考。     

Assessment of soil erosion risk using an integrated approach of GIS and Analytic Hierarchy Process (AHP) in Erzurum,Turkiye

Assessing soil erosion hazards and mapping the spatial distribution of soil erosion have an essential role in sustainable forest management. In this study, the potential soil erosion risk was evaluated through the Analytical Hierarchy Process (AHP) and Geographic Information Systems (GIS) in the Oltu forest planning unit, Erzurum. Seven erosion-related criteria, including slope, bedrock type, relative relief, drainage density and frequency, rainfall, and land use/land cover (LULC) were used for the present assessment. According to the AHP analysis, the slope was the most influential factor (21%) followed by bedrock type (19%), land cover (17%), and relative relief (14%) in the soil erosion process. The soil erosion risk in the study area was strongly influenced by the LULC where 59.46% is bare land with high erosion risk and 12.07%, with the lowest risk, is in an area with any forest cover. The estimated soil erosion risk was classified into five different classes namely very low, low, moderate, high, and very high. The results showed that this study area is highly prone to soil erosion. The larger proportion of the area (39.16%) is exposed to high to very high erosion, mainly determined by forest cover and geomorphology. To analyze the accuracy of the soil erosion risk map, 40 points were selected randomly in this study area. In these points, predicted values were compared to the real values obtained by Google Earth-colored images. The area under the ROC curve (AUC) method was applied to validate the efficiency of the AHP which showed a satisfactory accuracy of 81.00%. Findings presented that including the more influencing factors with a slope instead of including only the slope contributes to a more accurate erosion risk map. This study highlighted that GIS-based multi-criteria decision-making is a valuable and practical tool for decision-makers and land managers in creating soil erosion susceptibility maps and determining high-priority areas that require conservation measures for sustainable land use management by reducing the economic and ecological impacts of soil loss. Also, this approach can be practically applied in other planning units.     

基于NOAA-AVHRR数据的中国东部地区植被遥感分类研究

该文采用 19幅 (时间跨 8个月 ) 时间序列的NOAA_AVHRR的归一化植被指数 (NDVI) 最大值合成影像遥感数据, 经过主分量分析 (Principlecomponentanalysis, PCA) 处理后, 用非监督分类方法的ISODATA算法, 对中国东部地区的 (五省一市 ) 植被进行分类, 结果可以分出 2 8种土地覆盖类型, 除了两种类型为水体和城市或裸地外, 其余 2 6种类型均为植被类型, 根据中国植被分类系统, 这 2 6类可以归并为 6大植被类型 :1) 常绿阔叶林 ;2 ) 针叶林 ;3) 竹林 ;4 ) 灌草丛 ;5 ) 水生植被 ;6 ) 农业植被。用 1∶10 0 0 0 0 0数字化《中国植被图集》的植被类型检验遥感分类结果表明, 针叶林、灌草丛、常绿阔叶林和农业植被的分类具有较高的位置精度和面积精度, 位置精度分别为 79.2 %、91.3%、6 8.2 %和 95.9%, 面积精度分别达到 92.1%、95.9%、6 3.8%和 90.5 %。这 6大植被类型在地理空间上的分布规律与中国东部常绿阔叶林区植被的地带性分布基本一致。     

浙江省森林信息提取及其变化的空间分布

如何利用遥感技术提取森林信息是遥感应用的重要领域之一。以不同时相的Landsat TM/ETM+为数据源,采用面向对象和基于多级决策树的分类方法得到浙江省2000年、2005年以及2010年的森林植被覆被图。经实地采样点验证,2010年分类精度达到92.76%,精度满足要求。介绍了浙江森林信息的快速提取方法,即统计不同森林类型的Landsat TM影像原始波段和LBV变换值以及各种植被指数在各时相上的差异,经过C5决策树训练,选取合适的规则和阈值实现森林信息的提取。结果表明,面向对象分割与决策树算法结合可以作为森林信息提取的有效方法。最后,通过对3期森林专题图进行空间叠加分析,得到了森林资源动态变化的空间分布,并以此为基础对林地变化的类型及原因进行分析,结果显示浙江省森林资源变化主要分布在浙西北山区、浙中南山区以及沿海地带,这一结果可以为有关部门的决策提供依据。     

基于阈值法的山区森林常绿、落叶特征遥感自动识别方法——以贡嘎山地区为例

森林的常绿、落叶特征是土地覆被产品的重要属性。由于山区地形复杂,地表遥感辐射信号地形效应明显,导致山区森林常绿、落叶特征遥感自动识别一直是难点。提出了一种基于阈值法的山区森林常绿、落叶特征遥感自动识别简单实用方法。该方法利用多源、多时相遥感影像,选择归一化植被指数(NDVI)为指标,通过统计参考样本的NDVI在生长季和非生长季的差异,自动找出区分常绿、落叶特征的阈值,基于判别规则识别山区森林常绿、落叶特征。以贡嘎山地区为例,分别以多时相Landsat TM影像(简称TM)、多时相环境减灾卫星影像(简称HJ)为单源数据,多时相的HJ、TM组合影像为多源数据,验证该方法的有效性。实验结果表明,该方法能够有效识别山区森林常绿、落叶特征,总体精度达到93.87%,Kappa系数为0.87。该方法适用于山区大面积森林常绿、落叶特征遥感自动提取,已被成功应用于"生态十年"专项西南地区土地覆被数据的生产。     

Using multi-temporal tree inventory data in eucalypt forestry to benchmark global high-resolution canopy height models. A showcase in Mato Grosso,Brazil

The global monitoring of forest structure worldwide is increasingly being supported by refined and enhanced satellite mission datasets. Forest canopy height is a global metric to characterise and monitor dynamics in forest ecosystems worldwide. Satellite mapping missions as NASA's Global Ecosystem Dynamics Investigation (GEDI) are creating opportunities to refine global forest canopy height models adding forest structural information to time-series satellite imagery. A recent global canopy height model presented by Lang et al., (2022) using GEDI and 10-m Sentinel-2 and the map from Potapov et al., (2020) using GEDI and Landsat are both tested in this study using multi-temporal tree-level data collected over eucalypt plantations in Brazil. Our results at plot-level showed Lang et al., (2022)’s estimates of canopy height came short compared to 2020 maximum and mean tree height records in the plots, 7.6 and 3.6 m, respectively, but adding CHM standard deviation improves the agreement of ground records for maximum tree height. Higher errors were computed for the plots in 2019 using the Potapov's 30-m CHM: 14.2 and 9.5 m, respectively. Averaged stand values were more similar between the three sources tested. We report improvement from the 30-m CHM to the 10-m, but still height saturation problems were observed when accounting for height differences in tall eucalypt trees. As more global products for forest height and biomass are becoming available to users, more validation exercises as presented in this study are needed to assess the suitability of CHM products to forestry needs, and facilitate the uptake and actionability of the next generation of global height and biomass products. We provide recommendations and insights on the use of GEDI laser data for global mapping and on the potential of commercial forestry areas to benchmark the accuracy of satellite mapping missions focusing on tree height estimation in the tropics.     

Landscape‐scale lidar analysis of aboveground biomass distribution in secondary Brazilian Atlantic Forest

Secondary forests account for more than half of tropical forests and represent a growing carbon sink, but rates of biomass accumulation vary by a factor of two or more even among plots in the same landscape. To better understand the drivers of this variability, we used airborne lidar to measure forest canopy height and estimate biomass over 4529 ha at Serra do Conduru Park in Southern Bahia, Brazil. We measured trees in 30 georeferenced field plots (0.25‐ha each) to estimate biomass using allometry. Then we estimated aboveground biomass density (ABD) across the lidar study area using a statistical model developed from our field plots. This model related the 95th percentile of the distribution of lidar return heights to ABD. We overlaid this map of ABD on a Landsat‐derived forest age map to determine rates of biomass accumulation. We found rapid initial biomass regeneration (~6 Mg/ha yr), which slowed as forests aged. We also observed high variability in both height and biomass across the landscape within forests of similar age. Nevertheless, a regression model that accounted for spatial autocorrelation and included forest age, slope, and distance to roads or open areas explained 62 and 77 percent of the landscape variation in ABD and canopy height, respectively. Thus, while there is high spatial heterogeneity in forest recovery, and the drivers of this heterogeneity warrant further investigation, we suggest that a relatively simple set of predictor variables is sufficient to explain the majority of variance in both height and ABD in this landscape.     

Methods of Measuring the Proportions of Plant-Species Present in Forest and Their Effect on Estimates of Bird Preferences for Plant-Species

Stem density, basal area, vegetation cover and vegetation surface area were compared as measures of the proportions of plant species present in North Okarito Forest, South Westland, for use in determining bird preferences for plant species. In general, stem density estimates of the proportions of canopy species were about 10 times lower than basal area estimates. The converse was true for estimates of the proportions of sub- canopy and understorey species. The proportions estimated from vegetation cover and vegetation surface area were similar for most species, and were intermediate between the proportions estimated from stem density and basal area. However, in the upper forest tiers, vegetation cover gave lower estimates for the proportions of canopy species and higher estimates for the proportions of sub-canopy species than given by vegetation surface area. These differences affect calculation of bird preferences for plant species. We recommend vegetation surface area as a measure of the proportional availability of plant species to birds because it is appropriate to most birds in New Zealand forests, is likely to be more accurate than visual estimates of vegetation cover, and can be measured on the same plots separately for trunks, branches, foliage, and fruit.     

Burrowing seabirds affect forest regeneration, Rangatira Island, Chatham Islands, New Zealand

The forests of Rangatira Island (218 ha) in the Chatham Islands are a critical breeding site for a number of rare and threatened forest bird species, but are also home to more than three million seabirds, which could significantly affect forest regeneration processes. We surveyed the forests of Rangatira Island by establishing 40 permanent forest plots, estimated seabird density through burrow counts, and analysed soil properties. To determine if seabirds were impacting on forest regeneration, we established exclosures (0.25 m²) in 30 of the forest plots, and examined the role of canopy gaps in forest regeneration. The tallest current forest (c. 15 m), dominated by Plagianthus chathamicus, has mostly regenerated since stock were removed in 1959. Mean burrow density was estimated to be 1.19 per square metre, all soils were highly acidic (pH 3.36–5.18), and burrow density was positively correlated with soil phosphorus. Seedling density of woody species in seabird exclosures measured after 9, 24 and 33 months was significantly higher than in the adjacent non-gap plots, and seedling density was positively associated with reduced canopy cover. Seedling densities were also significantly higher in canopy gaps than in adjacent non-gap plots, but seabird burrow density was significantly lower in gaps. These results suggest that canopy gaps allow forest regeneration despite the negative impacts of seabird burrowing. However, the gap makers, largely senescing Olearia traversii, are slowly disappearing from the forests. The cohort of Plagianthus that has regenerated following farm abandonment may progressively collapse, allowing regeneration to continue in small openings, but there is also the potential for a catastrophic blowdown. This might have serious implications for forest-dwelling birds, invertebrates, and plants.     

基于MODIS混合像元分解的湖南省森林碳密度反演

随着遥感技术的快速发展,基于遥感影像和地面样地的方法成为目前森林碳密度估算的常用手段.然而由于混合像元的存在严重制约了区域森林碳密度反演精度的提高,特别是MODIS这种低空间分辨率影像.本研究以MODIS影像和固定样地为数据源,开展森林碳密度的反演研究.首先利用不带约束、带约束的线性分解和非线性分解3种方法进行混合像元分解,导出不同土地利用/覆盖类型的丰度图;然后采用结合和未结合丰度图的序列高斯协同模拟算法对湖南省森林碳密度进行反演.结果表明： 3种混合像元分解模型中,带约束线性分解估计的地物丰度精度最高（平均均方根误差0.002）,明显优于不带约束线性分解和非线性分解模型;通过将混合像元分解模型和序列高斯协同模拟算法结合,森林碳密度反演精度从74.1%提高到81.5%,均方根误差从7.26减小到5.18;2009年湖南省森林碳密度的平均值为30.06 t·hm^-2,变化范围介于0.00～67.35 t·hm^-2之间.这表明混合像元分解在提高区域和全球尺度森林碳密度反演精度方面显示出巨大的潜力.     

Traits and growth of liana regeneration in primary and secondary forests of Central Amazonia

Question: Do traits of liana regeneration differ among secondary forest types of varying land‐use history and primary forest? Location: Eighty kilometers north of Manaus, Brazil. Methods: We compared plant functional traits and growth rates of liana regeneration (<1.7‐m length) among two secondary forest types and primary forest. Secondary forest types were: Vismia (on land formerly clear‐cut, used for pasture and intensively burned) and Cecropia (no pasture usage or intensive fires after clear‐cut). Results: A principal components analysis indicated that most of the primary forest species exhibited a similar habit and were characterized by short shoots and small, round leaves with low specific leaf area, whereas secondary forest species had a broad range of trait values. At the plot level, primary and secondary forest communities were separated mainly by plant length and leaf size. Plant size varied more within secondary than within primary forest plots. The two secondary forest types could not be separated based on the traits of liana regeneration. Relative growth rate (RGR) did not correlate significantly with any measured plant trait, except for a negative relation to initial length. RGR increased with decreasing canopy cover and was highest in Vismia forest plots. Conclusion: Plant functional traits of liana regeneration were more similar in the primary forest and differed substantially from secondary forests, yet canopy cover only partly explained the observed differences.     

A new forest cover map of continental southeast Asia derived from SPOT‐VEGETATION satellite imagery

Question: Can recent satellite imagery of coarse spatial resolution support forest cover assessment and mapping at the regional level? Location: Continental southeast Asia. Methods: Forest cover mapping was based on digital classification of SPOT4‐VEGETATION satellite images of 1 km spatial resolution from the dry seasons 1998/1999 and 1999/2000. Following a geographical stratification, the spectral clusters were visually assigned to land cover classes. The forest classes were validated by an independent set of maps, derived from interpretation of satellite imagery of high spatial resolution (Landsat TM, 30 m). Forest area estimates from the regional forest cover map were compared to the forest figures of the FAO database. Results: The regional forest cover map displays 12 forest and land cover classes. The mapping of the region's deciduous and fragmented forest cover remained challenging. A high correlation was found between forest area estimates obtained from this map and from the Landsat TM derived maps. The regional and sub‐regional forest area estimates were close to those reported by FAO. Conclusion: SPOT4‐VEGETATION satellite imagery can be used for mapping consistently and uniformly the extent and distribution of the broad forest cover types at the regional scale. The new map can be considered as an update and improvement on existing regional forest cover maps.     

A Practical and Automated Approach to Large Area Forest Disturbance Mapping with Remote Sensing

In this paper, I describe a set of procedures that automate forest disturbance mapping using a pair of Landsat images. The approach is built on the traditional pair-wise change detection method, but is designed to extract training data without user interaction and uses a robust classification algorithm capable of handling incorrectly labeled training data. The steps in this procedure include: i) creating masks for water, non-forested areas, clouds, and cloud shadows; ii) identifying training pixels whose value is above or below a threshold defined by the number of standard deviations from the mean value of the histograms generated from local windows in the short-wave infrared (SWIR) difference image; iii) filtering the original training data through a number of classification algorithms using an n-fold cross validation to eliminate mislabeled training samples; and finally, iv) mapping forest disturbance using a supervised classification algorithm. When applied to 17 Landsat footprints across the U.S. at five-year intervals between 1985 and 2010, the proposed approach produced forest disturbance maps with 80 to 95% overall accuracy, comparable to those obtained from traditional approaches to forest change detection. The primary sources of mis-classification errors included inaccurate identification of forests (errors of commission), issues related to the land/water mask, and clouds and cloud shadows missed during image screening. The approach requires images from the peak growing season, at least for the deciduous forest sites, and cannot readily distinguish forest harvest from natural disturbances or other types of land cover change. The accuracy of detecting forest disturbance diminishes with the number of years between the images that make up the image pair. Nevertheless, the relatively high accuracies, little or no user input needed for processing, speed of map production, and simplicity of the approach make the new method especially practical for forest cover change analysis over very large regions.     

发展NECT土地覆盖特征数据集的原理、方法和应用

着重探讨了建立中国东北样带 (NortheastChinatransect, NECT) 土地覆盖特征数据集的原理、方法及其在全球变化研究方面的重要应用。NECT土地覆盖特征数据集是以多时相的 1km分辨率的NOAA/AVHRR归一化植被指数NDVI (Normalizeddifferencevegetationindex) 数字影像为基础, 同时采用高程、气候、土壤、植被、土地利用、土地资源、生态区域、行政边界、经济、社会等多源数据作为数据源, 并经过标准化处理 (如数字化、空间插值、几何配准、投影转换 ) 集成而成。在土地覆盖特征数据集的主要应用方面, 如 :1) 利用多时相、1km分辨率的NOAA/AVHRR影像完成了中国东北样带土地覆盖分类图。一级分类系统包括森林、草原、荒漠和沙地、灌丛、农田、混合覆盖 类型、城镇和水体等 8类, 二级分类体系包括 12类。经过地面采样进行精度检验, 分类精度达到 81.6 1%。 2 ) 对主要植被类型的植物生长季变化进行的研究。利用多时相的遥感影像构造了能够反映植被年际、季节生长变化的遥感植被指数ND VImax、NDVI变幅xam以及NDVI的标准偏差x′s 等, 分析这 3个参数 1983~ 1999年的 17年中的变化情况。该数据集的建立是研究该样带土地覆盖特征及其变化规律的基础, 对基于样带的全球变化研究有重要的意义。     

The Utility of Spectral Indices from Landsat ETM+ for Measuring the Structure and Composition of Tropical Dry Forests1

There is a growing emphasis on developing methods for quantifying the structure and composition of tropical forests that can be applied over large landscapes, especially for tropical dry forests that are severely fragmented and have a high conservation priority. This study investigates the relationships between various measures of forest structure (annual woody increment, canopy closure, stand density, stand basal area) and composition (tree species diversity, tree community composition) measured in semi‐deciduous tropical dry forests on islands in Lago Guri, Venezuela and three spectral indices derived from Landsat ETM+: Normalized Difference Vegetation Index (NDVI), Infrared Index (IRI), and Mid‐Infrared Index (MIRI). Even though there were significant autocorrelations among spectral indices, there were significant differences in the relationships between spectral indices and forest attributes. IRI was not significantly correlated with any of the structural variables while MIRI was correlated with canopy closure and NDVI values were correlated with canopy closure as well as annual woody increment. NDVI and MIRI were both related to relative tree diversity and all three indices were associated with aspects of tree species composition. Based on the results of this study, it appears that spectral indices, and in particular NDVI, may be useful indicators of forest attributes in tropical dry forest habitats. Further research needs to be undertaken to identify if the results of this study can be applied to other tropical dry forests at a global spatial scale.     

基于MODIS数据的中国三种主要土地类型变化的空间特征分析

人类活动使得土地利用和植被覆盖发生了巨大变化,直接影响着全球气候。本研究通过从2000—2013年对中国三种主要土地利用类型的NDVI变化特征进行了分析,结果表明:(1)14年来,中国三种主要土地利用类型NDVI平均值均有增强的趋势。(2)三种主要土地利用类型中除耕地中的水田,林地中的有林地和草地中的高覆盖草地增长速率不显著外,其他土地类型增长速率均显著。(3)三种土地利用类型均以改善面积大于退化面积,耕地中改善面积占总耕地的64.21%,退化的区域占18.50%;林地改善的区域占总林地的54.21%,退化的区域占20.13%;草地改善的区域占55.53%,退化的区域占18.23%。三种土地类型均有所改善且改善明显的区域主要集中在甘肃以南,陕西以北和东北部分地区。     

Successional Change and Resilience of a Very Dry Tropical Deciduous Forest Following Shifting Agriculture

We analyzed successional patterns in a very dry tropical deciduous forest by using 15 plots differing in age after abandonment and contrasted them to secondary successions elsewhere in the tropics. We used multivariate ordination and nonlinear models to examine changes in composition and structure and to estimate forest recovery rates and resilience. A shrub phase characterized early succession (0–3 yr); afterwards, the tree Mimosa acantholoba became dominant. Below its canopy, sprouts and seed-regenerated individuals of mature forest species slowly accumulated. Canopy height, plant density, and crown cover stabilized in less than 15 yr, whereas species richness, diversity, and basal area continued to increase. The pioneer species group has very low diversity and the long-lived pioneer phase typical of humid forests is absent; species composition may therefore recover soon as suggested by convergence toward mature forest species composition. The time trend of plant density also differed from humid forests for it lacked its characteristic density decline, presumably because of differences in regeneration mechanisms between very dry and other less water-stressed forest types. As opposed to the prevailing hypothesis, resilience was not higher than in moister forests, and thus factors other than structure relative simplicity must be accounted for when assessing resilience.     

Determinants of woody species richness in Scot pine and beech forests: climate, forest patch size and forest structure

We analysed patterns of woody species richness in Pinus sylvestris and Fagus sylvatica forests in Catalonia (NE Spain) from forestry inventory databank in relation to climate and landscape structure. Both types of forests are found within the same climatic range, although they have been managed following somewhat different goals. Overall, woody species richness significantly increased when conditions get closer to the Mediterranean ones, with milder temperatures. Differences between the two types of forests arose when comparing the relationship between richness and forest patch size. Woody species richness increased in pine forests with patch size, while the opposite trend was observed in beech forests. This pattern is explained by the different behaviour of structural canopy properties, since leaf area index and canopy cover showed a steeper increase with increasing forest patch size in Fagus forests than in Pinus ones. Accordingly, richness decreased with canopy cover in Fagus plots, but not in Pinus ones. We suggest that these differences would be related to management history, which may have enhanced the preservation of beech stands in larger forest landscape units.     

大兴安岭呼中林业局森林景观格局变化及其驱动力

以大兴安岭呼中林业局为研究区,利用TM数据、森林资源清查数据和采伐统计资料,结合野外调查,以GIS为技术手段,剖析研究区1989～2000年景观变化,结果表明该区主要景观类型为针叶林、针阔混交林和阔叶林,这3种类型面积总比例达90%以上,其他景观类型面积比例小于10%,主要有荒草地、火烧迹地、采伐迹地、沼泽、居民点和道路.10a间景观由原有的大面积连续的针叶林为基质、小面积的阔叶林和针阔混交林斑块散布其中的格局,转变成大面积针阔混交林为基质的格局.研究区1989年是成过熟林、高公顷蓄积量、高郁闭度占优势的景观,2000年中幼龄林、低公顷蓄积量和低郁闭度森林面积增大,并趋于连续,形成低质量的相对均质化的景观.最后,结合历史资料,通过比较分析火烧、人工更新、土地利用和人为采伐活动这几种主要驱动力的特征,论证了采伐活动在该地区景观变化中所发挥的主导作用.