Search

Lake ice is an important feature for many limnic ecosystems. Periodic ice cover influences biological, chemical and physical processes in lakes during the cold climate period. Additionally, ice cover also affects processes outside the ice period, for instance, lake water temperature, timing of spring bloom, primary productivity and mixing regimes. In the face of climate change, many regions experience shifting lake ice cover. The observed and projected loss significantly affects lake ecology but also cultural ecosystem services. In the alpine region of Germany, people associate personal memories, sportive activities and many other aspects with lake ice. Lake ice is connected to society and also strongly affected by climate change. Therefore, it well suits as an indicator to communicate climate change. The region, however, lacks systematic measurements and data on lake ice cover.In our project, we therefore aimed at developing a remote sensing approach to create a comparable data basis for a climate change indicator on lake ice. Our case study analysed six lakes between 700 and 2000 m AMSL. We generated a data set on lake surface characteristics (water, ice, snow, transparent ice etc.) using public webcam imagery as independent source. The data set was used to train and validate random forest classifiers for Setinel-1 A/B, Sentinel-2 A/B and Landsat 8 imagery. We excluded Sentinel-1 data, which were acquired at wind speeds > 1 m/s (ERA5-LAND). Thus, we prevented erroneous classification of rough waters. The validation revealed very high accuracies with balanced overall accuracies around 0.99, which is misleading. The high accuracies result from how we designed the ground the data since we only used data with labelling under high certainty. In this region, mapping lake ice faces the challenge of multiple freezing in thawing processes within the ice period. We therefore, implemented an air temperature (ERA5-LAND) filter to check the plausibility of classification results.The final classification results differentiated binary between ice and no-ice pixels. From this data, we defined ice-days with at least 80 % ice cover on a lake. To build the indicator, we divided the monthly sum of ice days by the number of valid image acquisitions. Thus, the indicator also accounts for varyingly available satellite data.With covering currently seven ice periods, the time series is relatively short for a climate change indicator. The approach may also be transferred to archived imagery whereas lacking ground truth data remain challenging. The small size of (0.2 - 3 km²) complicates the usage of large scale sensors such as MODIS. Thus, combining data from five satellites resolving at 10 – 30 m allowed to generate comparable and spatially explicit data on ice cover of these lakes for the first time.

This is the most common answer to a common question at the beginning of the semester. Teaching remote sensing skills at university often is associated in physical geography but also geoscience studies. Thus, the topics with which we teach remote sensing skills are often related to these subjects. In undergraduate courses, the thematic interests strongly vary among students. In advanced master courses, students with various thematic and technical backgrounds (geo sciences, computing science, economics, ecology, law, politics etc.) may join remote sensing classes. Additionally, the number of students increases. From the teaching perspective, we aim to address the varying student needs and backgrounds and enable them to further develop their technical skills and have to cope with these challenges. In this presentation, we want to present two practical formats of currently taught remote sensing classes.In all classes, we work with freely available satellite data (Sentinels, Landsat, MERIS, MODIS etc.) and software (SNAP, QGIS, GoogleEarth, Sentinel-Hub and other browser-based tools). The first class is designed for undergraduate students (geography and related subjects, e.g. geosciences) who have a theoretical remote sensing background (lecture). After completing the class, the students should be able to independently conduct and document a remote sensing processing routine and evaluate results. To this end, the class is split in a part with instructions and a second part with independent work. First, the students work in groups through a modular online implemented course for ten weeks. The modules chronologically follow a basic routine to finally classify land use/ land cover in a study area. The modules contain theoretic background, prepared data, short videos on software usage and broad instructions. To assure the learning process, the students conduct self-tests after completed modules and participate in a weekly on-site tutorial. After completing all modules, they have to independently assess a flood event without detailed instructions and write a fictious report for a catastrophe response unit. Students positively evaluate the split structure, free division of work, videos and self-tests with feedback. Otherwise, they wish more time for asking questions and discuss issues of understanding in the on-site tutorials.The advanced master course “Remote Sensing Applications” is open for students with a basic, practical remote sensing knowledge coming from different master programs. After completing the class, the students should be able to independently process, analyse and discuss remote sensing data and combine them with additional data to work on a geo-/study-related topic (geology, coast, socio-economic, climatic etc.). To this end, we selected New Zealand as study area. Within on-site classes, the students work on the topics geothermics, urban heat islands, droughts, forestry and cloud computing with non-prepared satellite and other data. For the final project, they select a research topic on their own and present their analyses and results in a storymap. Students highly appreciated choosing an own topic for the examination and discussing them in the whole group. Here, we aim to reflect the presented classes with the community to further improve our current “solutions” for challenges in teaching remote sensing.

Seagrass meadows cover large benthic areas of the Baltic Sea, but eutrophication and climate change imply declining seagrass coverage. Apart from acoustic methods and traditional diver mappings, optical remote sensing techniques allow for mapping seagrass. Optical satellite analyses of seagrass mapping may supplement acoustic methods in shallow coastal waters with observations that are more frequent and have a larger spatial coverage.In the clear Greek Mediterranean Sea, Sentinel-2 was already applied successfully to detect bathymetry and seagrass meadows. We are now testing whether Sentinel-2 data are also suitable for analysing the sublittoral in the turbid waters of the Baltic Sea. We focus on an extensive shallow water area near Kiel/Germany. Based on Sentinel-2 data, we analyse water depth and differentiate between seagrass covered and bare sandy ground. We derive these parameters using empirical and process-based models. First results show that Sentinel-2 allows to determine water depths up to 4 m (RMSE ~ 0.2 m). Comparisons with LiDAR water depths show that inaccuracies increase in overgrown areas. Our study also shows that the atmospheric correction algorithm influences sublittoral ground mappings with Sentinel-2 data. For instance, the absolute water depths of the process-based modelling differ up to 2.5 m on average depending on the atmospheric correction algorithm (ACOLITE, Sen2Cor, iCOR).Comparing Sentinel-2 seagrass classifications with diver mappings and aerial imagery emphasises that empiric approaches provide plausible sublittoral ground classifications up to approximately 4 m water depth. Combining these results with seagrass mappings based on acoustic measurements (deeper than 4 m water) provides a synthesised sublittoral classification map of the study area up to the present growth limit of seagrass (~ 7 m in the study area).The Baltic Sea is considered as a very turbid environment, nevertheless we show that satellite-based remote sensing has a great potential for shedding light into the "white ribbon". The spatial coverage and temporal resolution of the analysed Sentinel-2 data increases the knowledge about the occurrence of seagrass and its spatio-temporal dynamics. Nevertheless, the influence of the selected atmospheric correction approach on the results shows that further research in remote sensing is necessary to assess seagrass meadows reliably.