Refine
Has Fulltext
- yes (26)
Is part of the Bibliography
- yes (26)
Year of publication
- 2020 (26) (remove)
Document Type
- Journal article (23)
- Doctoral Thesis (3)
Language
- English (26) (remove)
Keywords
- machine learning (4)
- remote sensing (4)
- AI (2)
- CNN (2)
- Earth Observation (2)
- Earth observation (2)
- Landsat (2)
- SAR (2)
- artificial intelligence (2)
- convolutional neural networks (2)
Institute
- Institut für Geographie und Geologie (26) (remove)
Worldwide, cold regions are undergoing significant alterations due to climate change. Snow, the most widely distributed cold region component, is highly sensitive to climate change. At the same time, snow itself profoundly impacts the Earth’s energy budget, biodiversity, and natural hazards, as well as hydropower management, freshwater management, and winter tourism/sports. Large parts of the cold regions in Europe are mountain areas, which are densely populated because of the various ecosystem services and socioeconomic well-being in mountains. At present, severe consequences caused by climate change have been observed in European mountains and their surrounding areas. Yet, large knowledge gaps hinder the development of effective regional and local adaptation strategies. Long-term and evidence-based regional studies are urgently needed to enhance the comprehension of regional responses to climate change.
Earth Observation (EO) provides long-term consistent records of the Earth’s surface. It is a great alternative and/or supplement to conventional in-situ measurements which are usually time-consuming, cost-intensive and logistically demanding, particularly for the poor accessibility of cold regions. With the assistance of EO, land surface dynamics in cold regions can be observed in an objective, repeated, synoptic and consistent way. Thanks to free and open data policies, long-term archives such as Landsat Archive and Sentinel Archive can be accessed free-of-charge. The high- to medium-resolution remote sensing imagery from these freely accessible archives gives EO-based time series datasets the capability to depict snow dynamics in European mountains from the 1980s to the present. In order to compile such a dataset, it is necessary to investigate the spatiotemporal availability of EO data, and develop a spatiotemporally transferable framework from which one can investigate snow dynamics.
Among the available EO image archives, the Landsat Archive has the longest uninterrupted records of the Earth’s land surface. Furthermore, its 30 m spatial resolution fulfils the requirements for snow monitoring in complex terrains. Landsat data can yield a time series of snow dynamics in mountainous areas from 1984 to the present. However, severe Landsat data gaps have occurred across certain regions of Europe. Moreover, the Landsat Level 1 Precision and Terrain (L1TP) data is scarcer (up to 50% less) in high-latitude mountainous areas than in low-latitude mountainous areas. Given the abovementioned facts, the Regional Snowline Elevation (RSE) is selected to characterize the snow dynamics in mountainous areas, as it can handle cloud obstructions in the optical images. In this thesis, I present a five-step framework to derive and densify RSE time series in European mountains, i.e. (1) pre-processing, (2) snow detection, (3) RSE retrieval, (4) time series densification, and (5) Regional Snowline Retreat Curve (RSRC) production.
The results of the intra-annual RSE variations show a uniquely high variation in the beginning of the ablation seasons in the Alpine catchment Tagliamento, mainly toward higher elevation. As for inter-annual variations of RSE, median RSE increases in all selected catchments, with an average speed of around 4.66 m ∙ a−1 (median) and 5.87 m ∙ a−1 (at the beginning of the ablation season). The fastest significant retreat is observed in the catchment Drac (10.66 m ∙ a−1, at the beginning of the ablation season), and the slowest significant retreat is observed in the catchment Uzh (1.74 m ∙ a−1, at the beginning of the ablation season). The increase of RSEs at the beginning of the ablation season is faster than the median RSEs, whose average difference is nearly 1.21 m ∙ a−1, particularly in the catchment Drac (3.72 m ∙ a−1). The results of the RSRCs show a significant rise in RSEs at the beginning of the ablation season, except for the Alpine catchment Alpenrhein and Var, and the Pyrenean catchment Ariege. It indicates that 11.8 and 3.97 degrees Celsius less per year are needed for the regional snowlines to reach the middle point of the RSRC in the Tagliamento and Tysa, respectively. The variation of air temperature is regarded as an example of a potential climate driver in this thesis. The retrieved monthly mean RSEs are highly correlated (mean correlation coefficient "R" ̅ = 0.7) with the monthly temperature anomalies, which are more significant in months with extremely low/high temperature. Another case study that investigates the correlation between river discharges and RSEs is carried out to demonstrate the potential consequences of the derived snowline dynamics. The correlation analysis shows a good correlation between river discharges and RSEs (correlation coefficient, R=0.52).
In this thesis, the developed framework signifies a better understanding of the snow dynamics in mountain areas, as well as their potential triggers and consequences. Nonetheless, an urgent need persists for: (1) validation data to assess long-term snow-related observations based on high-resolution EO data; (2) further studies to reveal interactions between snow and its ambient environment; and (3) regional and local adaptation-strategies coping with climate change. Further studies exploring the above-mentioned research gaps are urgently needed in the future.
Forecasting spatio-temporal dynamics on the land surface using Earth Observation data — a review
(2020)
Reliable forecasts on the impacts of global change on the land surface are vital to inform the actions of policy and decision makers to mitigate consequences and secure livelihoods. Geospatial Earth Observation (EO) data from remote sensing satellites has been collected continuously for 40 years and has the potential to facilitate the spatio-temporal forecasting of land surface dynamics. In this review we compiled 143 papers on EO-based forecasting of all aspects of the land surface published in 16 high-ranking remote sensing journals within the past decade. We analyzed the literature regarding research focus, the spatial scope of the study, the forecasting method applied, as well as the temporal and technical properties of the input data. We categorized the identified forecasting methods according to their temporal forecasting mechanism and the type of input data. Time-lagged regressions which are predominantly used for crop yield forecasting and approaches based on Markov Chains for future land use and land cover simulation are the most established methods. The use of external climate projections allows the forecasting of numerical land surface parameters up to one hundred years into the future, while auto-regressive time series modeling can account for intra-annual variances. Machine learning methods have been increasingly used in all categories and multivariate modeling that integrates multiple data sources appears to be more popular than univariate auto-regressive modeling despite the availability of continuously expanding time series data. Regardless of the method, reliable EO-based forecasting requires high-level remote sensing data products and the resulting computational demand appears to be the main reason that most forecasts are conducted only on a local scale. In the upcoming years, however, we expect this to change with further advances in the field of machine learning, the publication of new global datasets, and the further establishment of cloud computing for data processing.
Mapping buried paleogeographical features of the Nile Delta (Egypt) using the Landsat archive
(2020)
The contribution highlights the use of Landsat spectral-temporal metrics (STMs) for the detection of surface anomalies that are potentially related to buried near-surface paleogeomorphological deposits in the Nile Delta (Egypt), in particular for a buried river branch close to Buto. The processing was completed in the Google Earth Engine (GEE) for the entire Nile Delta and for selected seasons of the year (summer/winter) using Landsat data from 1985 to 2019. We derived the STMs of the tasseled cap transformation (TC), the Normalized Difference Wetness Index (NDWI), and the Normalized Difference Vegetation Index (NDVI). These features were compared to historical topographic maps of the Survey of Egypt, CORONA imagery, the digital elevation model of the TanDEM-X mission, and modern high-resolution satellite imagery. The results suggest that the extent of channels is best revealed when differencing the median NDWI between summer (July/August) and winter (January/February) seasons (ΔNDWI). The observed difference is likely due to lower soil/plant moisture during summer, which is potentially caused by coarser-grained deposits and the morphology of the former levee. Similar anomalies were found in the immediate surroundings of several Pleistocene sand hills (“geziras”) and settlement mounds (“tells”) of the eastern delta, which allowed some mapping of the potential near-surface continuation. Such anomalies were not observed for the surroundings of tells of the western Nile Delta. Additional linear and meandering ΔNDWI anomalies were found in the eastern Nile Delta in the immediate surroundings of the ancient site of Bubastis (Tell Basta), as well as several kilometers north of Zagazig. These anomalies might indicate former courses of Nile river branches. However, the ΔNDWI does not provide an unambiguous delineation.
Visualizing movement data is challenging: While traditional spatial data can be sufficiently displayed as two‐dimensional plots or maps, movement trajectories require the representation of time in a third dimension. To address this, we present moveVis, an R package, which provides tools to animate movement trajectories, overlaying simultaneous uni‐ or multi‐temporal raster imagery or vector data.
moveVis automates the processing of movement and environmental data to turn such into an animation. This includes (a) the regularization of movement trajectories enforcing uniform time instances and intervals across all trajectories, (b) the frame‐wise mapping of movement trajectories onto temporally static or dynamic environmental layers, (c) the addition of customizations, for example, map elements or colour scales and (d) the rendering of frames into an animation encoded as GIF or video file.
moveVis is designed to display interactions and concurrencies of animal movement and environmental data. We present examples and use cases, ranging from data exploration to visualizing scientific findings.
Static spatial plots of movement data disregard the temporal dimension that distinguishes movement from other spatial data. In contrast, animations allow to display relocation in both time and space. We deem animations a powerful way to visually explore movement data, frame analytical findings and display potential interactions with spatially continuous and temporally dynamic environmental covariates.
The new ellipsocephaloid trilobite species Kingaspidoides spinirecurvatus has a spectacular morphology because of a unique set of two long and anteriorly recurved spines on the occipital ring and the axial ring of thoracic segment 8. Together with the long genal spines this whimsical dorsally directed spine arrangement is thought to act as a non-standard protective device against predators. This is illustrated by the body posture during different stages of enrolment, contrasting with the more sophisticated spinosities seen in later trilobites, which are discussed in brief. Kingaspidoides spinirecurvatus from the lower–middle Cambrian boundary interval of the eastern Anti-Atlas in Morocco has been known for about two decades, with specimens handled as precious objects on the fossil market. Similar, but far less spectacular, spine arrangements on the thoracic axial rings are known from other ellipsocephaloid trilobites from the Anti-Atlas of Morocco and the Franconian Forest region of Germany. This suggests that an experimental phase of spine development took place within the Kingaspi-doides clade during the early–middle Cambrian boundary interval.
Grasslands cover one third of the earth’s terrestrial surface and are mainly used for livestock production. The usage type, use intensity and condition of grasslands are often unclear. Remote sensing enables the analysis of grassland production and management on large spatial scales and with high temporal resolution. Despite growing numbers of studies in the field, remote sensing applications in grassland biomes are underrepresented in literature and less streamlined compared to other vegetation types. By reviewing articles within research on satellite-based remote sensing of grassland production traits and management, we describe and evaluate methods and results and reveal spatial and temporal patterns of existing work. In addition, we highlight research gaps and suggest research opportunities. The focus is on managed grasslands and pastures and special emphasize is given to the assessment of studies on grazing intensity and mowing detection based on earth observation data. Grazing and mowing highly influence the production and ecology of grassland and are major grassland management types. In total, 253 research articles were reviewed. The majority of these studies focused on grassland production traits and only 80 articles were about grassland management and use intensity. While the remote sensing-based analysis of grassland production heavily relied on empirical relationships between ground-truth and satellite data or radiation transfer models, the used methods to detect and investigate grassland management differed. In addition, this review identified that studies on grassland production traits with satellite data often lacked including spatial management information into the analyses. Studies focusing on grassland management and use intensity mostly investigated rather small study areas with homogeneous intensity levels among the grassland parcels. Combining grassland production estimations with management information, while accounting for the variability among grasslands, is recommended to facilitate the development of large-scale continuous monitoring and remote sensing grassland products, which have been rare thus far.
This study compares the performance of the five widely used crop growth models (CGMs): World Food Studies (WOFOST), Coalition for Environmentally Responsible Economies (CERES)-Wheat, AquaCrop, cropping systems simulation model (CropSyst), and the semi-empiric light use efficiency approach (LUE) for the prediction of winter wheat biomass on the Durable Environmental Multidisciplinary Monitoring Information Network (DEMMIN) test site, Germany. The study focuses on the use of remote sensing (RS) data, acquired in 2015, in CGMs, as they offer spatial information on the actual conditions of the vegetation. Along with this, the study investigates the data fusion of Landsat (30 m) and Moderate Resolution Imaging Spectroradiometer (MODIS) (500 m) data using the spatial and temporal reflectance adaptive reflectance fusion model (STARFM) fusion algorithm. These synthetic RS data offer a 30-m spatial and one-day temporal resolution. The dataset therefore provides the necessary information to run CGMs and it is possible to examine the fine-scale spatial and temporal changes in crop phenology for specific fields, or sub sections of them, and to monitor crop growth daily, considering the impact of daily climate variability. The analysis includes a detailed comparison of the simulated and measured crop biomass. The modelled crop biomass using synthetic RS data is compared to the model outputs using the original MODIS time series as well. On comparison with the MODIS product, the study finds the performance of CGMs more reliable, precise, and significant with synthetic time series. Using synthetic RS data, the models AquaCrop and LUE, in contrast to other models, simulate the winter wheat biomass best, with an output of high R2 (>0.82), low RMSE (<600 g/m\(^2\)) and significant p-value (<0.05) during the study period. However, inputting MODIS data makes the models underperform, with low R2 (<0.68) and high RMSE (>600 g/m\(^2\)). The study shows that the models requiring fewer input parameters (AquaCrop and LUE) to simulate crop biomass are highly applicable and precise. At the same time, they are easier to implement than models, which need more input parameters (WOFOST and CERES-Wheat).
Information about land use/land cover (LULC) and their changes is useful for different stakeholders to assess future pathways of sustainable land use for food production as well as for nature conservation. In this study, we assess LULC changes in the Kilombero catchment in Tanzania, an important area of recent development in East Africa. LULC change is assessed in two ways: first, post-classification comparison (PCC) which allows us to directly assess changes from one LULC class to another, and second, spectral change detection. We perform LULC classification by applying random forests (RF) on sets of multitemporal metrics that account for seasonal within-class dynamics. For the spectral change detection, we make use of the robust change vector analysis (RCVA) and determine those changes that do not necessarily lead to another class. The combination of the two approaches enables us to distinguish areas that show (a) only PCC changes, (b) only spectral changes that do not affect the classification of a pixel, (c) both types of change, or (d) no changes at all. Our results reveal that only one-quarter of the catchment has not experienced any change. One-third shows both, spectral changes and LULC conversion. Changes detected with both methods predominantly occur in two major regions, one in the West of the catchment, one in the Kilombero floodplain. Both regions are important areas of food production and economic development in Tanzania. The Kilombero floodplain is a Ramsar protected area, half of which was converted to agricultural land in the past decades. Therefore, LULC monitoring is required to support sustainable land management. Relatively poor classification performances revealed several challenges during the classification process. The combined approach of PCC and RCVA allows us to detect spatial patterns of LULC change at distinct dimensions and intensities. With the assessment of additional classifier output, namely class-specific per-pixel classification probabilities and derived parameters, we account for classification uncertainty across space. We overlay the LULC change results and the spatial assessment of classification reliability to provide a thorough picture of the LULC changes taking place in the Kilombero catchment.
The Kunduz River is one of the main tributaries of the Amu Darya Basin in North Afghanistan. Many communities live in the Kunduz River Basin (KRB), and its water resources have been the basis of their livelihoods for many generations. This study investigates climate change impacts on the KRB catchment. Rare station data are, for the first time, used to analyze systematic trends in temperature, precipitation, and river discharge over the past few decades, while using Mann–Kendall and Theil–Sen trend statistics. The trends show that the hydrology of the basin changed significantly over the last decades. A comparison of landcover data of the river basin from 1992 and 2019 shows significant changes that have additional impact on the basin hydrology, which are used to interpret the trend analysis. There is considerable uncertainty due to the data scarcity and gaps in the data, but all results indicate a strong tendency towards drier conditions. An extreme warming trend, partly above 2 °C since the 1960s in combination with a dramatic precipitation decrease by more than −30% lead to a strong decrease in river discharge. The increasing glacier melt compensates the decreases and leads to an increase in runoff only in the highland parts of the upper catchment. The reduction of water availability and the additional stress on the land leads to a strong increase of barren land and a reduction of vegetation cover. The detected trends and changes in the basin hydrology demand an active management of the already scarce water resources in order to sustain water supply for agriculture and ecosystems in the KRB.
The monitoring of land cover and land use change is critical for assessing the provision of ecosystem services. One of the sources for long-term land cover change quantification is through the classification of historical and/or current maps. Little research has been done on historical maps using Object-Based Image Analysis (OBIA). This study applied an object-based classification using eCognition tool for analyzing the land cover based on historical maps in the Main river catchment, Upper Franconia, Germany. This allowed land use change analysis between the 1850s and 2015, a time span which covers the phase of industrialization of landscapes in central Europe. The results show a strong increase in urban area by 2600%, a severe loss of cropland (−24%), a moderate reduction in meadows (−4%), and a small gain in forests (+4%). The method proved useful for the application on historical maps due to the ability of the software to create semantic objects. The confusion matrix shows an overall accuracy of 82% for the automatic classification compared to manual reclassification considering all 17 sample tiles. The minimum overall accuracy was 65% for historical maps of poor quality and the maximum was 91% for very high-quality ones. Although accuracy is between high and moderate, coarse land cover patterns in the past and trends in land cover change can be analyzed. We conclude that such long-term analysis of land cover is a prerequisite for quantifying long-term changes in ecosystem services.
Availability of water and desiccation of important water reservoirs is a vital challenge in semi-arid to arid climates with growing economy and population. Low quantities of precipitation and high evaporation rates leave the water supply vulnerable to human activity and climatic variations. Endorheic basins of Northern Iran were hydrologically landlocked within geological timescales and thus bear evidence of past variations of water resources in generations of water related landforms, like abandoned lake level shorelines, alluvial fans and stream terraces. Understanding the development of these landforms reveals crucial information about past water reservoirs and landscape history.
This study offers a comprehensive approach on understanding the geomorphological development of the landscape throughout Late Pleistocene and Holocene times. It integrates remote sensing and geographic information system analysis, with geomorphological and stratigraphical mapping fieldwork and detailed sedimentological investigations.
The work shows the importance of analytical geomorphological mapping for delineating stratigraphic units of the Iranian Quaternary. Thus, several phases of drying and lake level retreat were identified in parallel geoarchives and could be dated to a time span from today to Late Pleistocene. The findings link the fate of the citizens of the ancient city of "Tepe Hissar" to their access to water and to the power of geomorphological processes, which started changing their environment.
Supraglacial lakes can have considerable impact on ice sheet mass balance and global sea-level-rise through ice shelf fracturing and subsequent glacier speedup. In Antarctica, the distribution and temporal development of supraglacial lakes as well as their potential contribution to increased ice mass loss remains largely unknown, requiring a detailed mapping of the Antarctic surface hydrological network. In this study, we employ a Machine Learning algorithm trained on Sentinel-2 and auxiliary TanDEM-X topographic data for automated mapping of Antarctic supraglacial lakes. To ensure the spatio-temporal transferability of our method, a Random Forest was trained on 14 training regions and applied over eight spatially independent test regions distributed across the whole Antarctic continent. In addition, we employed our workflow for large-scale application over Amery Ice Shelf where we calculated interannual supraglacial lake dynamics between 2017 and 2020 at full ice shelf coverage. To validate our supraglacial lake detection algorithm, we randomly created point samples over our classification results and compared them to Sentinel-2 imagery. The point comparisons were evaluated using a confusion matrix for calculation of selected accuracy metrics. Our analysis revealed wide-spread supraglacial lake occurrence in all three Antarctic regions. For the first time, we identified supraglacial meltwater features on Abbott, Hull and Cosgrove Ice Shelves in West Antarctica as well as for the entire Amery Ice Shelf for years 2017–2020. Over Amery Ice Shelf, maximum lake extent varied strongly between the years with the 2019 melt season characterized by the largest areal coverage of supraglacial lakes (~763 km\(^2\)). The accuracy assessment over the test regions revealed an average Kappa coefficient of 0.86 where the largest value of Kappa reached 0.98 over George VI Ice Shelf. Future developments will involve the generation of circum-Antarctic supraglacial lake mapping products as well as their use for further methodological developments using Sentinel-1 SAR data in order to characterize intraannual supraglacial meltwater dynamics also during polar night and independent of meteorological conditions. In summary, the implementation of the Random Forest classifier enabled the development of the first automated mapping method applied to Sentinel-2 data distributed across all three Antarctic regions.
Protection and recovery of natural resource and biodiversity requires accurate monitoring at multiple scales. Airborne Laser Scanning (ALS) provides high-resolution imagery that is valuable for monitoring structural changes to vegetation, providing a reliable reference for ecological analyses and comparison purposes, especially if used in conjunction with other remote-sensing and field products. However, the potential of ALS data has not been fully exploited, due to limits in data availability and validation. To bridge this gap, the global network for airborne laser scanner data (GlobALS) has been established as a worldwide network of ALS data providers that aims at linking those interested in research and applications related to natural resources and biodiversity monitoring. The network does not collect data itself but collects metadata and facilitates networking and collaborative research amongst the end-users and data providers. This letter describes this facility, with the aim of broadening participation in GlobALS.
In China, freshwater is an increasingly scarce resource and wetlands are under great pressure. This study focuses on China's second largest freshwater lake in the middle reaches of the Yangtze River — the Dongting Lake — and its surrounding wetlands, which are declared a protected Ramsar site. The Dongting Lake area is also a research region of focus within the Sino-European Dragon Programme, aiming for the international collaboration of Earth Observation researchers. ESA's Copernicus Programme enables comprehensive monitoring with area-wide coverage, which is especially advantageous for large wetlands that are difficult to access during floods. The first year completely covered by Sentinel-1 SAR satellite data was 2016, which is used here to focus on Dongting Lake's wetland dynamics. The well-established, threshold-based approach and the high spatio-temporal resolution of Sentinel-1 imagery enabled the generation of monthly surface water maps and the analysis of the inundation frequency at a 10 m resolution. The maximum extent of the Dongting Lake derived from Sentinel-1 occurred in July 2016, at 2465 km\(^2\), indicating an extreme flood year. The minimum size of the lake was detected in October, at 1331 km\(^2\). Time series analysis reveals detailed inundation patterns and small-scale structures within the lake that were not known from previous studies. Sentinel-1 also proves to be capable of mapping the wetland management practices for Dongting Lake polders and dykes. For validation, the lake extent and inundation duration derived from the Sentinel-1 data were compared with excerpts from the Global WaterPack (frequently derived by the German Aerospace Center, DLR), high-resolution optical data, and in situ water level data, which showed very good agreement for the period studied. The mean monthly extent of the lake in 2016 from Sentinel-1 was 1798 km\(^2\), which is consistent with the Global WaterPack, deviating by only 4%. In summary, the presented analysis of the complete annual time series of the Sentinel-1 data provides information on the monthly behavior of water expansion, which is of interest and relevance to local authorities involved in water resource management tasks in the region, as well as to wetland conservationists concerned with the Ramsar site wetlands of Dongting Lake and to local researchers.
Illegal small-scale mining (galamsey) in South-Western Ghana has grown tremendously in the last decade and caused significant environmental degradation. Excessive cloud cover in the area has limited the use of optical remote sensing data to map and monitor the extent of these activities. This study investigated the use of annual time-series Sentinel-1 data to map and monitor illegal mining activities along major rivers in South-Western Ghana between 2015 and 2019. A change detection approach, based on three time-series features — minimum, mean, maximum — was used to compute a backscatter threshold value suitable to identify/detect mining-induced land cover changes in the study area. Compared to the mean and maximum, the minimum time-series feature (in both VH and VV polarization) was found to be more sensitive to changes in backscattering within the period of investigation. Our approach permitted the detection of new illegal mining areas on an annual basis. A backscatter threshold value of +1.65 dB was found suitable for detecting illegal mining activities in the study area. Application of this threshold revealed illegal mining area extents of 102 km\(^2\), 60 km\(^2\) and 33 km\(^2\) for periods 2015/2016–2016/2017, 2016/2017–2017/2018 and 2017/2018–2018/2019, respectively. The observed decreasing trend in new illegal mining areas suggests that efforts at stopping illegal mining yielded positive results in the period investigated. Despite the advantages of Synthetic Aperture Radar data in monitoring phenomena in cloud-prone areas, our analysis revealed that about 25% of the Sentinel-1 data, mostly acquired in March and October (beginning and end of rainy season respectively), were unusable due to atmospheric effects from high intensity rainfall events. Further investigation in other geographies and climatic regions is needed to ascertain the susceptibility of Sentinel-1 data to atmospheric conditions.
In Earth observation (EO), large-scale land-surface dynamics are traditionally analyzed by investigating aggregated classes. The increase in data with a very high spatial resolution enables investigations on a fine-grained feature level which can help us to better understand the dynamics of land surfaces by taking object dynamics into account. To extract fine-grained features and objects, the most popular deep-learning model for image analysis is commonly used: the convolutional neural network (CNN). In this review, we provide a comprehensive overview of the impact of deep learning on EO applications by reviewing 429 studies on image segmentation and object detection with CNNs. We extensively examine the spatial distribution of study sites, employed sensors, used datasets and CNN architectures, and give a thorough overview of applications in EO which used CNNs. Our main finding is that CNNs are in an advanced transition phase from computer vision to EO. Upon this, we argue that in the near future, investigations which analyze object dynamics with CNNs will have a significant impact on EO research. With a focus on EO applications in this Part II, we complete the methodological review provided in Part I.
Deep learning (DL) has great influence on large parts of science and increasingly established itself as an adaptive method for new challenges in the field of Earth observation (EO). Nevertheless, the entry barriers for EO researchers are high due to the dense and rapidly developing field mainly driven by advances in computer vision (CV). To lower the barriers for researchers in EO, this review gives an overview of the evolution of DL with a focus on image segmentation and object detection in convolutional neural networks (CNN). The survey starts in 2012, when a CNN set new standards in image recognition, and lasts until late 2019. Thereby, we highlight the connections between the most important CNN architectures and cornerstones coming from CV in order to alleviate the evaluation of modern DL models. Furthermore, we briefly outline the evolution of the most popular DL frameworks and provide a summary of datasets in EO. By discussing well performing DL architectures on these datasets as well as reflecting on advances made in CV and their impact on future research in EO, we narrow the gap between the reviewed, theoretical concepts from CV and practical application in EO.
Land surface temperature (LST) is a fundamental parameter within the system of the Earth’s surface and atmosphere, which can be used to describe the inherent physical processes of energy and water exchange. The need for LST has been increasingly recognised in agriculture, as it affects the growth phases of crops and crop yields. However, challenges in overcoming the large discrepancies between the retrieved LST and ground truth data still exist. Precise LST measurement depends mainly on accurately deriving the surface emissivity, which is very dynamic due to changing states of land cover and plant development. In this study, we present an LST retrieval algorithm for the combined use of multispectral optical and thermal UAV images, which has been optimised for operational applications in agriculture to map the heterogeneous and diverse agricultural crop systems of a research campus in Germany (April 2018). We constrain the emissivity using certain NDVI thresholds to distinguish different land surface types. The algorithm includes atmospheric corrections and environmental thermal emissions to minimise the uncertainties. In the analysis, we emphasise that the omission of crucial meteorological parameters and inaccurately determined emissivities can lead to a considerably underestimated LST; however, if the emissivity is underestimated, the LST can be overestimated. The retrieved LST is validated by reference temperatures from nearby ponds and weather stations. The validation of the thermal measurements indicates a mean absolute error of about 0.5 K. The novelty of the dual sensor system is that it simultaneously captures highly spatially resolved optical and thermal images, in order to construct the precise LST ortho-mosaics required to monitor plant diseases and drought stress and validate airborne and satellite data.
Forests in Germany cover around 11.4 million hectares and, thus, a share of 32% of Germany's surface area. Therefore, forests shape the character of the country's cultural landscape. Germany's forests fulfil a variety of functions for nature and society, and also play an important role in the context of climate levelling. Climate change, manifested via rising temperatures and current weather extremes, has a negative impact on the health and development of forests. Within the last five years, severe storms, extreme drought, and heat waves, and the subsequent mass reproduction of bark beetles have all seriously affected Germany’s forests. Facing the current dramatic extent of forest damage and the emerging long-term consequences, the effort to preserve forests in Germany, along with their diversity and productivity, is an indispensable task for the government. Several German ministries have and plan to initiate measures supporting forest health. Quantitative data is one means for sound decision-making to ensure the monitoring of the forest and to improve the monitoring of forest damage. In addition to existing forest monitoring systems, such as the federal forest inventory, the national crown condition survey, and the national forest soil inventory, systematic surveys of forest condition and vulnerability at the national scale can be expanded with the help of a satellite-based earth observation. In this review, we analysed and categorized all research studies published in the last 20 years that focus on the remote sensing of forests in Germany. For this study, 166 citation indexed research publications have been thoroughly analysed with respect to publication frequency, location of studies undertaken, spatial and temporal scale, coverage of the studies, satellite sensors employed, thematic foci of the studies, and overall outcomes, allowing us to identify major research and geoinformation product gaps.
Water crises are becoming severe in recent times, further fueled by population increase and climate change. They result in complex and unsustainable water management. Spatial estimation of consumptive water use is vital for performance assessment of the irrigation system using Remote Sensing (RS). For this study, its estimation is done using the Soil Energy Balance Algorithm for Land (SEBAL) approach. Performance indicators including equity, adequacy, and reliability were worked out at various spatiotemporal scales. Moreover, optimization and sustainable use of water resources are not possible without knowing the factors mainly influencing consumptive water use of major crops. For that purpose, random forest regression modelling was employed using various sets of factors for site-specific, proximity, and cropping system. The results show that the system is underperforming both for Kharif (i.e., summer) and Rabi (i.e., winter) seasons. Performance indicators highlight poor water distribution in the system, a shortage of water supply, and unreliability. The results are relatively good for Rabi as compared to Kharif, with an overall poor situation for both seasons. Factors importance varies for different crops. Overall, distance from canal, road density, canal density, and farm approachability are the most important factors for explaining consumptive water use. Auditing of consumptive water use shows the potential for resource optimization through on-farm water management by the targeted approach. The results are based on the present situation without considering future changes in canal water supply and consumptive water use under climate change.