Search

This article presents and technically describes a new field spectro-goniometer system for the ground-based characterization of the surface reflectance anisotropy under natural illumination conditions developed at the Alfred Wegener Institute (AWI). The spectro-goniometer consists of a Manual Transportable Instrument platform for ground-based Spectro-directional observations (ManTIS), and a hyperspectral sensor system. The presented measurement strategy shows that the AWI ManTIS field spectro-goniometer can deliver high quality hemispherical conical reflectance factor (HCRF) measurements with a pointing accuracy of ± 6 cm within the constant observation center. The sampling of a ManTIS hemisphere (up to 30° viewing zenith, 360° viewing azimuth) needs approx. 18 min. The developed data processing chain in combination with the software used for the semi-automatic control provides a reliable method to reduce temporal effects during the measurements. The presented visualization and analysis approaches of the HCRF data of an Arctic low growing vegetation showcase prove the high quality of spectro-goniometer measurements. The patented low-cost and lightweight ManTIS instrument platform can be customized for various research needs and is available for purchase. [ABSTRACT FROM AUTHOR]

Remote sensing has become a valuable tool in monitoring arctic environments. The aim of this paper is ground-based hyperspectral characterization of Low Arctic Alaskan tundra communities along four environmental gradients (regional climate, soil pH, toposequence, and soil moisture) that all vary in ground cover, biomass, and dominating plant communities. Field spectroscopy in connection with vegetation analysis was carried out in summer 2012, along the North American Arctic Transect (NAAT). Spectral metrics were extracted, including the averaged reflectance and absorption-related metrics such as absorption depths and area of continuum removal. The spectral metrics were investigated with respect to "greenness", biomass, vegetation height, and soil moisture regimes. The results show that the surface reflectances of all sites are similar in shape with a reduced near-infrared (NIR) reflectance that is specific for low-growing biomes. The main spectro-radiometric findings are: (i) Southern sites along the climate gradient have taller shrubs and greater overall vegetation biomass, which leads to higher reflectance in the NIR. (ii) Vegetation height and surface wetness are two antagonists that balance each other out with respect to the NIR reflectance along the toposequence and soil moisture gradients. (iii) Moist acidic tundra (MAT) sites have "greener" species, more leaf biomass, and green-colored moss species that lead to higher pigment absorption compared to moist non-acidic tundra (MNT) sites. (iv) MAT and MNT plant community separation via narrowband Normalized Difference Vegetation Index (NDVI) shows the potential of hyperspectral remote sensing applications in the tundra. [ABSTRACT FROM AUTHOR]

Historical land cover maps are of high importance for scientists and policy makers studying the dynamic character of land cover change in the Sudano-Sahel, including anthropogenic and climatological drivers. Despite its relevance, an accurate high resolution record of historical land cover maps is currently lacking over the Sudano-Sahel. In this study, 30 m resolution historically consistent land cover and cover fraction maps are provided over the Sudano-Sahel for the period 1986–2015. These land cover/cover fraction maps are achieved based on the Landsat archive preprocessed on Google Earth Engine and a random forest classification/regression model, while historical consistency is achieved using the hidden Markov model. Using these historical maps, a multitude of variability in the dynamic Sudano-Sahel region over the past 30 years is revealed. On the one hand, Sahel-wide cropland expansion and the re-greening of the Sahel is observed in the discrete land cover classification. On the other hand, subtle changes such as forest degradation are detected based on the cover fraction maps. Additionally, exploiting the 30 m spatial resolution, fine-scale changes, such as smallholder or subsistence farming, can be detected. The historical land cover/cover fraction maps presented in this study are made available via an open-access platform. [ABSTRACT FROM AUTHOR]

In May 2019, Collection 2 of the Copernicus Global Land Cover layers was released. Next to a global discrete land cover map at 100 m resolution, a set of cover fraction layers is provided depicting the percentual cover of the main land cover types in a pixel. This additional continuous classification scheme represents areas of heterogeneous land cover better than the standard discrete classification scheme. Overall, 20 layers are provided which allow customization of land cover maps to specific user needs or applications (e.g., forest monitoring, crop monitoring, biodiversity and conservation, climate modeling, etc.). However, Collection 2 was not just a global up-scaling, but also includes major improvements in the map quality, reaching around 80% or more overall accuracy. The processing system went into operational status allowing annual updates on a global scale with an additional implemented training and validation data collection system. In this paper, we provide an overview of the major changes in the production of the land cover maps, that have led to this increased accuracy, including aligning with the Sentinel 2 satellite system in the grid and coordinate system, improving the metric extraction, adding better auxiliary data, improving the biome delineations, as well as enhancing the expert rules. An independent validation exercise confirmed the improved classification results. In addition to the methodological improvements, this paper also provides an overview of where the different resources can be found, including access channels to the product layer as well as the detailed peer-review product documentation. [ABSTRACT FROM AUTHOR]

Warming in the Arctic has resulted in changes in the distribution and composition of vegetation communities. Many of these changes are occurring at fine spatial scales and at the level of individual species. Broad-band, coarse-scale remote sensing methods are commonly used to assess vegetation changes in the Arctic, and may not be appropriate for detecting these fine-scale changes; however, the use of hyperspectral, high resolution data for assessing vegetation dynamics remains scarce. The aim of this paper is to assess the ability of field spectroscopy to differentiate among four vegetation communities in the Low Arctic of Alaska. Primary data were collected from the North Slope site of Ivotuk, Alaska (68.49°N, 155.74°W) and analyzed using spectrally resampled hyperspectral narrowbands (HNBs). A two-step sparse partial least squares (SPLS) and linear discriminant analysis (LDA) was used for community separation. Results from Ivotuk were then used to predict community membership at five other sites along the Dalton Highway in Arctic Alaska. Overall classification accuracy at Ivotuk ranged from 84%-94% and from 55%-91% for the Dalton Highway test sites. The results of this study suggest that hyperspectral data acquired at the field level, along with the SPLS and LDA methodology, can be used to successfully discriminate among Arctic tundra vegetation communities in Alaska, and present an improvement over broad-band, coarse-scale methods for community classification. [ABSTRACT FROM AUTHOR]

Aims: An Arctic Vegetation Classification (AVC) is needed to address issues related to rapid Arctic-wide changes to climate, land-use, and biodiversity. Location: The 7.1 million km2 Arctic tundra biome. Approach and conclusions: The purpose, scope and conceptual framework for an Arctic Vegetation Archive (AVA) and Classification (AVC) were developed during numerous workshops starting in 1992. The AVA and AVC are modeled after the European vegetation archive (EVA) and classification (EVC). The AVA will use Turboveg for data management. The AVC will use a Braun-Blanquet (Br.-Bl.) classification approach. There are approximately 31,000 Arctic plots that could be included in the AVA. An Alaska AVA (AVA-AK, 24 datasets, 3026 plots) is a prototype for archives in other parts of the Arctic. The plan is to eventually merge data from other regions of the Arctic into a single Turboveg v3 database. We present the pros and cons of using the Br.-Bl. classification approach compared to the EcoVeg (US) and Biogeoclimatic Ecological Classification (Canada) approaches. The main advantages are that the Br.-Bl. approach already has been widely used in all regions of the Arctic, and many described, well-accepted vegetation classes have a pan-Arctic distribution. A crosswalk comparison of Dryas octopetala communities described according to the EcoVeg and the Braun-Blanquet approaches indicates that the non-parallel hierarchies of the two approaches make crosswalks difficult above the plant- community level. A preliminary Arctic prodromus contains a list of typical Arctic habitat types with associated described syntaxa from Europe, Greenland, western North America, and Alaska. Numerical clustering methods are used to provide an overview of the variability of habitat types across the range of datasets and to determine their relationship to previously described Braun-Blanquet syntaxa. We emphasize the need for continued maintenance of the Pan-Arctic Species List, and additional plot data to fully sample the variability across bio- climatic subzones, phytogeographic regions, and habitats in the Arctic. This will require standardized methods of plot-data collection, inclusion of physiogonomic information in the numeric analysis approaches to create formal definitions for vegetation units, and new methods of data sharing between the AVA and national vegetation-plot databases. [ABSTRACT FROM AUTHOR]

Widespread degradation of ice wedges has been observed during the last decades in numerous areas within the continuous permafrost zone of Eurasia and North America. To study ice-wedge degradation, we performed field investigations at Prudhoe Bay and Barrow in northern Alaska during 2011–2016. In each study area, a 250-m transect was established with plots representing different stages of ice-wedge degradation/stabilization. Field work included surveying ground- and water-surface elevations, thaw-depth measurements, permafrost coring, vegetation sampling, and ground-based LiDAR scanning. We described cryostratigraphy of frozen soils and stable isotope composition, analyzed environmental characteristics associated with ice-wedge degradation and stabilization, evaluated the vulnerability and resilience of ice wedges to climate change and disturbances, and developed new conceptual models of ice-wedge dynamics that identify the main factors affecting ice-wedge degradation and stabilization and the main stages of this quasi-cyclic process. We found significant differences in the patterns of ice-wedge degradation and stabilization between the two areas, and the patterns were more complex than those previously described because of the interactions of changing topography, water redistribution, and vegetation/soil responses that can interrupt or reinforce degradation. Degradation of ice wedges is usually triggered by an increase in the active-layer thickness during exceptionally warm and wet summers or as a result of flooding or disturbance. Vulnerability of ice wedges to thermokarst is controlled by the thickness of the intermediate layer of the upper permafrost, which overlies ice wedges and protects them from thawing. In the continuous permafrost zone, degradation of ice wedges rarely leads to their complete melting; and in most cases wedges eventually stabilize and can then resume growing, indicating a somewhat cyclic and reversible process. Stabilization of ice wedges after their partial degradation makes them better protected than before degradation because the intermediate layer is usually 2 to 3 times thicker on top of stabilized ice wedges than on top of initial ice wedges in undisturbed conditions. As a result, the likelihood of formation of large thaw lakes in the continuous permafrost zone triggered by ice-wedge degradation alone is very low. [ABSTRACT FROM AUTHOR]

Currently most global land cover maps are produced with discrete classes, which express the dominant land cover class in each pixel, or a combination of several classes at a predetermined ratio. In contrast, land cover fraction mapping enables expressing the proportion of each pure class in each pixel, which increases precision and reduces legend complexity. To map land cover fractions, regression rather than classification algorithms are needed, and multiple approaches are available for this task. A major challenge for land cover fraction mapping models is data sparsity. Land cover fraction data is by its nature zero-inflated due to how common the 0% fraction is. As regression favours the mean, 0% and 100% fractions are difficult for regression models to predict accurately. We proposed a new solution by combining three models: a binary model determines whether a pixel is pure; if so, it is processed using a classification model; otherwise with a regression model. We compared multiple regression algorithms and implemented our proposed three-step model on the algorithm with the lowest RMSE. We further evaluated the spatial and per-class accuracy of the model and demonstrated a wall-to-wall prediction of seven land cover fractions over the globe. The models were trained on over 138,000 points and validated on a separate dataset of over 20,000 points, provided by the CGLS-LC100 project. Both datasets are global and aligned with the PROBA-V 100 m UTM grid. Results showed that the random forest regression model reached the lowest RMSE of 17.3%. Lowest MAE (7.9%) and highest overall accuracy (72% ± 2%) was achieved using random forest with our proposed three-model approach and median vote. This research proves that machine learning algorithms can be applied globally to map a wide variety of land cover fractions. Fraction mapping expresses land cover more precisely, and empowers users to create their own discrete maps using user-defined thresholds and rules, which enables customising the result for a diverse range of uses. The three-step approach is useful for addressing the zero-inflation issue and mapping 0% and 100% fractions more accurately, and thus has already been taken up in the operational production of global land cover fraction layers within the CGLS-LC100 project. Furthermore, this study contributes to the accuracy assessment of land cover fraction maps both thematically and spatially, and these methods could be taken up by future land cover fraction mapping efforts. • 7 land cover fractions derived using machine learning models and validated globally. • Multi-step approach proposed for dealing with zero inflation in fraction data. • Independent spatial accuracy assessment based on over 20,000 reference points. • Multi-step approach gives a best guess and improves 0% and 100% fraction estimation. • Multi-fraction land cover mapping shown as an alternative to discrete classification. [ABSTRACT FROM AUTHOR]

A pilot study for mapping the Arctic wetlands was conducted in the Yukon Flats National Wildlife Refuge (Refuge), Alaska. It included commissioning the HySpex VNIR-1800 and the HySpex SWIR-384 imaging spectrometers in a single-engine Found Bush Hawk aircraft, planning the flight times, direction, and speed to minimize the strong bidirectional reflectance distribution function (BRDF) effects present at high latitudes and establishing improved data processing workflows for the high-latitude environments. Hyperspectral images were acquired on two clear-sky days in early September, 2018, over three pilot study areas that together represented a wide variety of vegetation and wetland environments. Steps to further minimize BRDF effects and achieve a higher geometric accuracy were added to adapt and improve the Hyspex data processing workflow, developed by the German Aerospace Center (DLR), for high-latitude environments. One-meter spatial resolution hyperspectral images, that included a subset of only 120 selected spectral bands, were used for wetland mapping. A six-category legend was established based on previous U.S. Geological Survey (USGS) and U.S. Fish and Wildlife Service (USFWS) information and maps, and three different classification methods—hybrid classification, spectral angle mapper, and maximum likelihood—were used at two selected sites. The best classification performance occurred when using the maximum likelihood classifier with an averaged Kappa index of 0.95; followed by the spectral angle mapper (SAM) classifier with a Kappa index of 0.62; and, lastly, by the hybrid classifier showing lower performance with a Kappa index of 0.51. Recommendations for improvements of future work include the concurrent acquisition of LiDAR or RGB photo-derived digital surface models as well as detailed spectra collection for Alaska wetland cover to improve classification efforts. [ABSTRACT FROM AUTHOR]