Tree species identification within an extensive forest area with diverse management regimes using airborne hyperspectral data

Information on tree species composition is crucial in forest management and can be obtained using remote sensing. While the topic has been addressed frequently over the last years, the remote sensing-based identification of tree species across wide and complex forest areas is still sparse in the literature. Our study presents a tree species classification of a large fraction of the Białowieża Forest in Poland covering 62 000 ha and being subject to diverse management regimes. Key objectives were to obtain an accurate tree species map and to examine if the prevalent management strategy influences the classification results. Tree species classification was conducted based on airborne hyperspectral HySpex data. We applied an iterative Support Vector Machine classification and obtained a thematic map of 7 individual tree species (birch, oak, hornbeam, lime, alder, pine, spruce) and an additional class containing other broadleaves. Generally, the more heterogeneous the area was, the more errors we observed in the classification results. Managed forests were classified more accurately than reserves. Our findings indicate that mapping dominant tree species with airborne hyperspectral data can be accomplished also over large areas and that forest management and its effects on forest structure has an influence on classification accuracies and should be actively considered when progressing towards operational mapping of tree species composition.     

Moving from plot-based to hillslope-scale assessments of savanna vegetation structure with long-range terrestrial laser scanning (LR-TLS)

Reliable quantification of savanna vegetation structure is critical for accurate carbon accounting and biodiversity assessment under changing climate and land-use conditions. Inventories of fine-scale vegetation structural attributes are typically conducted from field-based plots or transects, while large-area monitoring relies on a combination of airborne and satellite remote sensing. Both of these approaches have their strengths and limitations, but terrestrial laser scanning (TLS) has emerged as the benchmark for vegetation structural parameterization – recording and quantifying 3D structural detail that is not possible from manual field-based or airborne/spaceborne methods. However, traditional TLS approaches suffer from similar spatial constraints as field-based inventories. Given their small areal coverage, standard TLS plots may fail to capture the heterogeneity of landscapes in which they are embedded. Here we test the potential of long-range (>2000 m) terrestrial laser scanning (LR-TLS) to provide rapid and robust assessment of savanna vegetation 3D structure at hillslope scales. We used LR-TLS to sample entire savanna hillslopes from topographic vantage points and collected coincident plot-scale (1 ha) TLS scans at increasing distances from the LR-TLS station. We merged multiple TLS scans at the plot scale to provide the reference structure, and evaluated how 3D metrics derived from LR-TLS deviated from this baseline with increasing distance. Our results show that despite diluted point density and increased beam divergence with distance, LR-TLS can reliably characterize tree height (RMSE = 0.25–1.45 m) and canopy cover (RMSE = 5.67–15.91%) at distances of up to 500 m in open savanna woodlands. When aggregated to the same sampling grain as leading spaceborne vegetation products (10–30 m), our findings show potential for LR-TLS to play a key role in constraining satellite-based structural estimates in savannas over larger areas than traditional TLS sampling can provide.     

Update and spatial extension of strategic forest inventories using time series remote sensing and modeling

Up-to-date forest inventory information relating the characteristics of managed and natural forests is fundamental to sustainable forest management and required to inform conservation of biodiversity and assess climate change impacts and mitigation opportunities. Strategic forest inventories are difficult to compile over large areas and are often quickly outdated or spatially incomplete as a function of their long production cycle. As a consequence, automated approaches supported by remotely sensed data are increasingly sought to provide exhaustive spatial coverage for a set of core attributes in a timely fashion. The objective of this study was to demonstrate the integration of current remotely-sensed data products and pre-existing jurisdictional inventory data to map four forest attributes of interest (stand age, dominant species, site index, and stem density) for a 55 Mha study region in British Columbia, Canada. First, via image segmentation, spectrally homogenous objects were derived from Landsat surface-reflectance pixel composites. Second, a suite of Landsat-based predictors (e.g., spectral indices, disturbance history, and forest structure) and ancillary variables (e.g., geographic, topographic, and climatic) were derived for these units and used to develop predictive models of target attributes. For the often difficult classification of dominant species, two modelling approaches were compared: (a) a global Random Forests model calibrated with training samples collected over the entire study area, and (b) an ensemble of local models, each calibrated with spatially constrained local samples. Accuracy assessment based upon independent validation samples revealed that the ensemble of local models was more accurate and efficient for species classification, achieving an overall accuracy of 72% for the species which dominate 80% of the forested areas in the province. Results indicated that site index had the highest agreement between predicted and reference (R² = 0.74, %RMSE = 23.1%), followed by stand age (R² = 0.62, %RMSE = 35.6%), and stem density (R² = 0.33, %RMSE = 65.2%). Inventory attributes mapped at the image-derived unit level captured much finer details than traditional polygon-based inventory, yet can be readily reassembled into these larger units for strategic forest planning purposes. Based upon this work, we conclude that in a multi-source forest monitoring program, spatially localized and detailed characterizations enabled by time series of Landsat observations in conjunction with ancillary data can be used to support strategic inventory activities over large areas.     

Inter-comparison of remote sensing platforms for height estimation of mango and avocado tree crowns

To support the adoption of precision agricultural practices in horticultural tree crops, prior research has investigated the relationship between crop vigour (height, canopy density, health) as measured by remote sensing technologies, to fruit quality, yield and pruning requirements. However, few studies have compared the accuracy of different remote sensing technologies for the estimation of tree height. In this study, we evaluated the accuracy, flexibility, aerial coverage and limitations of five techniques to measure the height of two types of horticultural tree crops, mango and avocado trees. Canopy height estimates from Terrestrial Laser Scanning (TLS) were used as a reference dataset against height estimates from Airborne Laser Scanning (ALS) data, WorldView-3 (WV-3) stereo imagery, Unmanned Aerial Vehicle (UAV) based RGB and multi-spectral imagery, and field measurements. Overall, imagery obtained from the UAV platform were found to provide tree height measurement comparable to that from the TLS (R² = 0.89, RMSE = 0.19 m and rRMSE = 5.37 % for mango trees; R² = 0.81, RMSE = 0.42 m and rRMSE = 4.75 % for avocado trees), although coverage area is limited to 1–10 km² due to battery life and line-of-sight flight regulations. The ALS data also achieved reasonable accuracy for both mango and avocado trees (R² = 0.67, RMSE = 0.24 m and rRMSE = 7.39 % for mango trees; R² = 0.63, RMSE = 0.43 m and rRMSE = 5.04 % for avocado trees), providing both optimal point density and flight altitude, and therefore offers an effective platform for large areas (10 km²–100 km²). However, cost and availability of ALS data is a consideration. WV-3 stereo imagery produced the lowest accuracies for both tree crops (R² = 0.50, RMSE = 0.84 m and rRMSE = 32.64 % for mango trees; R² = 0.45, RMSE = 0.74 m and rRMSE = 8.51 % for avocado trees) when compared to other remote sensing platforms, but may still present a viable option due to cost and commercial availability when large area coverage is required. This research provides industries and growers with valuable information on how to select the most appropriate approach and the optimal parameters for each remote sensing platform to assess canopy height for mango and avocado trees.     

Satellite image analysis using crowdsourcing data for collaborative mapping: current and opportunities

ABSTRACT

Researchers are continually finding new applications of satellite images because of the growing number of high-resolution images with wide spatial coverage. However, the cost of these images is sometimes high, and their temporal resolution is relatively coarse. Crowdsourcing is an increasingly common source of data that takes advantage of local stakeholder knowledge and that provides a higher frequency of data. The complementarity of these two data sources suggests there is great potential for mutually beneficial integration. Unfortunately, there are still important gaps in crowdsourced satellite image analysis by means of crowdsourcing in areas such as land cover classification and emergency management. In this paper, we summarize recent efforts, and discuss the challenges and prospects of satellite image analysis for geospatial applications using crowdsourcing. Crowdsourcing can be used to improve satellite image analysis and satellite images can be used to organize crowdsourced efforts for collaborative mapping.     

面向遥感大数据的地学知识图谱构想

由于地球表面的时空异质性与复杂性,传统从遥感影像具有的信息特征出发,构建智能解译算法解决遥感地学认知的思路在应对面向全球的海量遥感大数据分析时,其精度和地学实用性已触及瓶颈.为此,本文从地学知识为核心的角度出发,结合当前知识图谱理论的发展,提出一种新的面向遥感大数据分析的地学思维构想——地学知识图谱.本构想将地学知识的...     

1961—2017年新疆积雪期时空变化特征及其与气象因子的关系

利用新疆89个地面站逐日积雪深度观测资料,研究探讨了1961—2017年新疆区域积雪期、积雪初日、积雪终日的时空变化规律,分析了北疆和天山山区积雪期的年代际和周期变化特征及其与气温、降水的关系.结果表明:新疆各地积雪期、积雪初日和终日存在明显的差异,积雪期以天山为界北多南少;从空间分布看,天山山区和新疆北部阿勒泰、塔城...     

四川盆地干旱灾害统计特征

利用四川盆地1980—2009年17市103个县(市)实测逐日降水资料,按照四川省气象局制定的四川盆地的干旱地方标准DB51/T581—2007,对四川盆地近30年干旱灾害进行统计分析。结果表明:四川盆地夏旱出现的频率最高,夏旱的高发区集中在盆地西北部的成都、德阳和绵阳,伏旱的发生频率最低,强度最强。春、夏、伏旱的空间分布高发区依次从盆地西北部向东南转移。干旱发生的频率整体呈增长趋势,且严重干旱发生频次增长明显,与20世纪80年代相比90年代增幅达到110.3%,21世纪00年代在90年代的基础上又递增20.0%,21世纪00年代发生的严重干旱频次为80年代的2.5倍。     

2001-2015年中国植被覆盖人为影响的时空格局

基于MODIS-NDVI和气温、降水数据,使用基于变异系数的人为影响模型定量计算了2001-2015年中国植被覆盖人为影响,辅以趋势分析、Hurst指数等方法探讨了中国植被覆盖人为影响的时空变化特征及未来演变趋势。研究发现：① 2001-2015年,中国植被覆盖人为影响由南向北空间分异愈发明显,年均值为-0.0102,植被覆盖在人类活动影响下轻微减少,负影响面积占51.59%,略大于正影响面积。② 中国植被覆盖人为影响年际变化特征明显,整体呈负影响波动减少趋势,降速为0.5%/10a,其中正影响、负影响均呈增大趋势,正影响增速（0.3%/10a）远大于负影响（0.02%/10a）。③ 2001-2015年间,中国植被覆盖人为正影响重心向东北方向移动,负影响重心向西南方向移动,东北部植被覆盖在人为影响下不断改善,西南部人类活动对植被破坏程度不断增大。④ 中国植被覆盖人为影响主要呈负影响减少和正影响增大趋势,面积占比分别为28.14%和25.21%,生态环境趋于改善。⑤ Hurst指数分析表明,中国植被覆盖人为影响未来演变趋势的反向特征强于正向特征,主要呈人为负影响先减少后增大趋势,面积占比15.59%。