Surface composition of debris-covered glaciers across the Himalaya using linear spectral unmixing of Landsat 8 OLI imagery

The Himalaya mountain range is characterized by highly glacierized, complex, dynamic topography. The ablation area of Himalayan glaciers often features a highly heterogeneous debris mantle comprising ponds, steep and shallow slopes of various aspects, variable debris thickness, and exposed ice cliffs associated with differing ice ablation rates. Understanding the composition of the supraglacial debris cover is essential for a proper understanding of glacier hydrology and glacier-related hazards. Until recently, efforts to map debris-covered glaciers from remote sensing focused primarily on glacier extent rather than surface characteristics and relied on traditional whole-pixel image classification techniques. Spectral unmixing routines, rarely used for debris-covered glaciers, allow decomposition of a pixel into constituting materials, providing a more realistic representation of glacier surfaces. Here we use linear spectral unmixing of Landsat 8 Operational Land Imager (OLI) images (30 m) to obtain fractional abundance maps of the various supraglacial surfaces (debris material, clean ice, supraglacial ponds and vegetation) across the Himalaya around the year 2015. We focus on the debris-covered glacier extents as defined in the database of global distribution of supraglacial debris cover. The spectrally unmixed surfaces are subsequently classified to obtain maps of composition of debris-covered glaciers across sample regions. We test the unmixing approach in the Khumbu region of the central Himalaya, and we evaluate its performance for supraglacial ponds by comparison with independently mapped ponds from high-resolution Pléiades (2 m) and PlanetScope imagery (3 m) for sample glaciers in two other regions with differing topo-climatic conditions. Spectral unmixing applied over the entire Himalaya mountain range (a supraglacial debris cover area of 2254 km2) indicates that at the end of the ablation season, debris-covered glacier zones comprised 60.9 % light debris, 23.8 % dark debris, 5.6 % clean ice, 4.5 % supraglacial vegetation, 2.1 % supraglacial ponds, and small amounts of cloud cover (2 %), with 1.2 % unclassified areas. The spectral unmixing performed satisfactorily for the supraglacial pond and vegetation classes (an F score of ∼0.9 for both classes) and reasonably for the debris classes (F score of 0.7). Supraglacial ponds were more prevalent in the monsoon-influenced central-eastern Himalaya (up to 4 % of the debris-covered area) compared to the monsoon-dry transition zone (only 0.3 %) and in regions with lower glacier elevations. Climatic controls (higher average temperatures and more abundant precipitation), coupled with higher glacier thinning rates and lower average glacier velocities, further favour pond incidence and the development of supraglacial vegetation. With continued advances in satellite data and further method refinements, the approach presented here provides avenues towards achieving large-scale, repeated mapping of supraglacial features.

Combining Optical and Radar Satellite Imagery to Investigate the Surface Properties and Evolution of the Lordsburg Playa, New Mexico, USA

Driven by erodible soil, hydrological stresses, land use/land cover (LULC) changes, and meteorological parameters, windblown dust events initiated from Lordsburg Playa, New Mexico, United States, threaten public safety and health through low visibility and exposure to dust emissions. Combining optical and radar satellite imagery products can provide invaluable benefits in characterizing surface properties of desert playas—a potent landform for wind erosion. The optical images provide a long-term data record, while radar images can observe land surface irrespective of clouds, darkness, and precipitation. As a home for optical and radar imagery, powerful algorithms, cloud computing infrastructure, and application programming interface applications, Google Earth Engine (GEE) is an invaluable resource facilitating acquisition, processing, and analysis. In this study, the fractional abundance of soil, vegetation, and water endmembers were determined from pixel mixtures using the linear spectral unmixing model in GEE for Lordsburg Playa. For this approach, Landsat 5 and 8 images at 30 m spatial resolution and Sentinel-2 images at 10–20 m spatial resolution were used. Employing the Interferometric Synthetic Aperture Radar (InSAR) techniques, the playa’s land surface changes and possible sinks for sediment loading from the surrounding catchment area were identified. In this data recipe, a pair of Sentinel-1 images bracketing a monsoon day with high rainfall and a pair of images representing spring (dry, windy) and monsoon seasons were used. The combination of optical and radar images significantly improved the effort to identify long-term changes in the playa and locations within the playa susceptible to hydrological stresses and LULC changes. The linear spectral unmixing algorithm addressed the limitation of Landsat and Sentinel-2 images related to their moderate spatial resolutions. The application of GEE facilitated the study by minimizing the time required for acquisition, processing, and analysis of images, and storage required for the big satellite data.

Mapping Mangrove Zonation Changes in Senegal with Landsat Imagery Using an OBIA Approach Combined with Linear Spectral Unmixing

The mangrove areas in Senegal have fluctuated considerably over the last few decades, and it is therefore important to monitor the evolution of forest cover in order to orient and optimise forestry policies. This study presents a method for mapping plant formations to monitor and study changes in zonation within the mangroves of Senegal. Using Landsat ETM+ and Landsat 8 OLI images merged to a 15-m resolution with a pansharpening method, a processing chain that combines an OBIA approach and linear spectral unmixing was developed to detect changes in mangrove zonation through a diachronic analysis. The accuracy of the discriminations was evaluated with kappa indices, which were 0.8 for the Saloum delta and 0.83 for the Casamance estuary. Over the last 20 years, the mangroves of Senegal have increased in surface area. However, the dynamics of zonation differ between the two main mangrove hydrosystems of Senegal. In Casamance, a colonisation process is underway. In the Saloum, Rhizophora mangle is undergoing a process of densification in mangroves and appears to reproduce well in both regions. Furthermore, this study confirms, on a regional scale, observations in the literature noting the lack of Avicennia germinans reproduction on a local scale. In the long term, these regeneration gaps may prevent the mangrove from colonising the upper tidal zones in the Saloum. Therefore, it would be appropriate to redirect conservation policies towards reforestation efforts in the Saloum rather than in Casamance and to focus these actions on the perpetuation of Avicennia germinans rather than Rhizophora mangle, which has no difficulty in reproducing. From this perspective, it is necessary to gain a more in-depth understanding of the specific factors that promote the success of Avicennia germinans seeding.