A5041252131- Land Use and Ecosystem Services
- Arctic and Antarctic ice dynamics
- Remote-Sensing Image Classification
- Distributed and Parallel Computing Systems
- Remote Sensing in Agriculture
- Species Distribution and Climate Change
- Cryospheric studies and observations
- Remote Sensing and Land Use
- Atmospheric and Environmental Gas Dynamics
- Advanced Vision and Imaging
- Real-Time Systems Scheduling
- Plant Water Relations and Carbon Dynamics
- Fire effects on ecosystems
- Scientific Computing and Data Management
- Climate variability and models
- Computational Physics and Python Applications
- Photovoltaic System Optimization Techniques
- Infrared Target Detection Methodologies
- Image Enhancement Techniques
- Energy and Environment Impacts
- Environmental Changes in China
- Opportunistic and Delay-Tolerant Networks
- Human Pose and Action Recognition
- Data Visualization and Analytics
- Advanced Computational Techniques and Applications
Leipzig University
2022-2025
Center For Remote Sensing (United States)
2024
Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR)
2020-2022
Humboldt-Universität zu Berlin
2020-2022
Dienstleistungszentrum Ländlicher Raum
2022
Shandong University
2016-2018
Institute of Space Science - INFLPR Subsidiary
2016
Abstract Climate extremes are on the rise. Impacts of extreme climate and weather events ecosystem services ultimately human well‐being can be partially attenuated by organismic, structural, functional diversity affected land surface. However, ongoing transformation terrestrial ecosystems through intensified exploitation management may put this buffering capacity at risk. Here, we summarize evidence that reductions in biodiversity destabilize functioning facing extremes. We then explore if...
Over the past decades, solar panels have been widely used to harvest energy owing decreased cost of silicon-based photovoltaic (PV) modules, and therefore it is essential remotely map monitor presence PV modules. Many studies explored on module detection based color aerial photography manual photo interpretation. Imaging spectroscopy data are capable providing detailed spectral information identify features PV, thus potentially become a promising resource for automated operational detection....
Abstract With climate extremes’ rising frequency and intensity, robust analytical tools are crucial to predict their impacts on terrestrial ecosystems. Machine learning techniques show promise but require well-structured, high-quality, curated analysis-ready datasets. Earth observation datasets comprehensively monitor ecosystem dynamics responses climatic extremes, yet the data complexity can challenge effectiveness of machine models. Despite recent progress in deep monitoring, there is a...
Advancements in Earth system science have seen a surge diverse datasets. System Data Cubes (ESDCs) been introduced to efficiently handle this influx of high-dimensional data. ESDCs offer structured, intuitive framework for data analysis, organising information within spatio-temporal grids. The structured nature unlocks significant opportunities Artificial Intelligence (AI) applications. By providing well-organised data, are ideally suited wide range sophisticated AI-driven tasks. An...
Abstract Recent advancements in Earth system science have been marked by the exponential increase availability of diverse, multivariate datasets characterised moderate to high spatio-temporal resolutions. System Data Cubes (ESDCs) emerged as one suitable solution for transforming this flood data into a simple yet robust structure. ESDCs achieve organising an analysis-ready format aligned with grid, facilitating user-friendly analysis and diminishing need extensive technical processing...
Progress in Earth system science is accelerating rapidly, due to the increasing availability of multivariate datasets, often global, with moderate high spatio-temporal resolutions. Turning these data into knowledge presents interoperability, technical, analytical, and other challenges. System Data Cubes (ESDCs) have surfaced as essential tools, offering analysis-ready, cloud-optimised solutions. Coupled advancements Artificial Intelligence (AI), solutions potential release a wealth...
Antarctic blue ice areas are exposed due to erosion and sublimation of snow. At the same time, surface melt can form types that spectrally similar ice, especially at low elevations. These termed melt-induced areas. Both sensitive indicators climate change. Satellite remote sensing is a powerful technique retrieve spatial extent their variation in time. Yet, existing satellite-derived area products either mono-temporal for entire sheet, or multi-temporal limited area. Here, we present FABIAN,...
Terrestrial surface processes exhibit distinctive spectral signatures captured by optical satellites. Despite the development of over two hundred indices (SIs), current studies often narrow their focus to individual SIs, overlooking broader context land processes. This project seeks understand holistic features Sentinel-2 based SIs and relationships with human impact overall dynamics. To address this, we propose an AI-driven approach that synthesises derived from Sentinel data through...
Abstract. The intensification of climate extremes is one the most immediate effects global change. Heatwaves and droughts have uneven impacts on ecosystems that can be exacerbated in case compound events. To comprehensively study these events, e.g. with local high-resolution remote sensing or in-situ data, a catalogue such events essential. Here, we propose workflow to build database large-scale dry hot extreme based data from ERA5 reanalysis. Drought indicators are constructed...
With climate change-related extreme events on the rise, high dimensional Earth observation data presents a unique opportunity for forecasting and understanding impacts ecosystems. This is, however, impeded by complexity of processing, visualizing, modeling, explaining this data. To showcase how challenge can be met, here we train convolutional long short-term memory-based architecture novel DeepExtremeCubes dataset. includes around 40,000 long-term Sentinel-2 minicubes (January 2016-October...
One of the most basic classification tasks in remote sensing is to distinguish between water bodies and other surface types. Although there are numerous techniques for extracting from satellite imagery, still a need research more accurately identify with view efficient maintenance future. Delineation accuracy limited by varying amounts suspended matter different background land covers, especially those low albedo. Therefore, objective this study was develop an advanced index that improves...
Climate extremes are on the rise. Impacts of extreme climate and weather events ecosystem services ultimately human well-being can be partially attenuated by organismic, structural, functional diversity affected land surface. However, ongoing transformation terrestrial ecosystems through intensified exploitation management may put this buffering capacity at risk. Here, we summarise evidence that reductions in biodiversity destabilise functioning facing extremes. We then explore if impaired...
With climate extremes' rising frequency and intensity, robust analytical tools are crucial to predict their impacts on terrestrial ecosystems. Machine learning techniques show promise but require well-structured, high-quality, curated analysis-ready datasets. Earth observation datasets comprehensively monitor ecosystem dynamics responses climatic extremes, yet the data complexity can challenge effectiveness of machine models. Despite recent progress in deep monitoring, there is a need for...
Recent advancements in Earth system science have been marked by the exponential increase availability of diverse, multivariate datasets characterised moderate to high spatio-temporal resolutions. System Data Cubes (ESDCs) emerged as one suitable solution for transforming this flood data into a simple yet robust structure. ESDCs achieve organising an analysis-ready format aligned with grid, facilitating user-friendly analysis and diminishing need extensive technical processing knowledge....
Urban areas contain a complex mixture of surface materials resulting in mixed pixels that are challenging to handle with conventional mapping approaches. In particular, for spaceborne hyperspectral images (HSIs) sufficient spectral resolution differentiate urban materials, the spatial 30 m (e.g. EnMAP HSIs) makes it difficult find spectrally pure required detailed materials. Gradient analysis, which is commonly used ecology map natural vegetation consisting species, therefore promising and...
In this paper, we propose a new fast and bad data detection identification technology model as the research goal, to achieve accurate efficient identification. This paper analyzes influence of single, multiple multi-correlation on state estimation results in actual power grid. Based characteristics its status methods at home abroad, according distribution, The object grid rapid partition method, further proposed two-layer fine solution method.
Abstract. Solar photovoltaic power plants are in rapid expansion throughout the world, with total area occupied by panels being linked to electrical produced. This paper considers this case as an instance of generic problem estimating a class interest spaceborne hyperspectral images. As spatial resolution characterizing these sensors is too coarse, spectral unmixing techniques identify contribution specific material spectrum related single image element. Final results obtained summing all...
Many studies analyzing spaceborne hyperspectral images (HSIs) have so far struggled to deal with a lack of pure pixels due complex mixtures urban surface materials. Recently, an alternative concept gradients in material composition has been proposed and successfully applied map cities HSIs without the requirement for previous determination pixels. The gradient treats all as mixed aims describe quantify gradual transitions cover fractions This presents promising approach tackle mapping using...
Intensive transformation of landscape has taken place in Weihai, Shandong province. This gave us the idea to investigate urban development this area by use archived satellite data, available for last 30 years. We utilized advanced processing schemes geometrically correct and spectrally calibrate about 60 cloud free frames Landsat data from 1984 2015. used different classifiers extract model relevant parameters provide information layers that display changes trends incl. robust statistics...
The Weihai area in Shandong province has undergone intensive changes of the landscape over past years. city expanded to a high percentage but is still characterized by huge number lakes and surrounding coastal waters (case-II). In study, we investigated temperature those zones as well some features at land 20 We were able measure subtle trends considering background global warming issues 1.4 degree within this time span. First all, advanced processing schemes are utilized conduct necessary...
Extreme events are on the rise. The 2022 compound heatwave and drought event caused significant vegetation mortality serious ecosystem destruction in Europe that urgently need to be investigated. In this study, we used climate data (ERA5-Land air temperature at 2 m precipitation) remote sensing products (kNDVI derived from MODIS ESA CCI Land Cover product) investigate dynamics of extreme responses. Furthermore, compared effects year other normal as well abnormal years Europe. We propose a...
Compound heat waves and drought events draw our particular attention as they become more frequent. Co-occurring extreme often exacerbate impacts on ecosystems can induce a cascade of detrimental consequences. However, the research to understand these is still in its infancy. DeepExtremes project funded by European Space Agency (https://rsc4earth.de/project/deepextremes/) aiming at using deep learning gain insight into Earth surface under climate conditions. Specifically, goal forecast...