SDSebastian Drost

Sebastian Drost

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Sebastian Drost has a total of 11 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Real-time Delivery of Sensor Data Streams using IoT and OGC Standards

    by Simon Jirka, Christian Autermann, Sebastian Drost

    In the past, many projects have evaluated and demonstrated the use of the Sensor Web Enablement (SWE) Standards of the Open Geospatial Consortium (OGC) in order to publish sensor data. Advantages of these standards included the provision of a domain-independent approach for ensuring interoperability of interfaces, data, and metadata. However, in most cases, the developed infrastructures were limited to pull-based data retrieval patterns. This means that data consumers regularly query servers for data updates which may result in high server loads due to a high-frequency of update requests or increased latencies until a consumer receives new sensor data.Although there were relevant specifications such as the OGC Publish/Subscribe standard as well as discussion papers, the OGC SWE framework never included a widely accepted solution to handle an active push-based delivery of observation data. With the adaptation of the SensorThings API standard of the OGC in conjunction with mainstream Internet of Things protocols such as the Message Queuing Telemetry Transport (MQTT) protocol this has changed in recent years.In 2020 we have already presented at the EGU an approach on how to use these technologies to enable the efficient collection of sensor observation data in hydrological application by bridging between sensors and data management servers (Drost et al., 2020).As part of this contribution, we will discuss the applicability of these technologies, OGC SensorThings API as well as MQTT, to also cover the delivery of data to consumers in addition to the previously described data transmission from sensor devices to a data sink. We will put special emphasis on experiences gathered from the deployment in marine environments (e.g., live underway data and event metadata streams of research vessels), as part of the EMODnet Ingestion II project. Special consideration will be given to a discussion of potential advantages of push-based communication patterns as well as identified challenges for future work (e.g., metadata about push-based data streams, standardization of payloads, access control, best practices on how to structure provided data streams).Furthermore, we will address the development of data visualization tools for such interoperable real-time data streams and will discuss the opportunities to transfer these technologies to further application domains such as hydrology.ReferencesDrost, S., Speckamp, J., Hollmann, C., Malewski, C., Rieke, J., & Jirka, S. (2020). Internet of Things Technologies for the Efficient Collection of Hydrological Measurement Data. EGU General Assembly 2020, Online. https://doi.org/10.5194/egusphere-egu2020-10452

  • WaCoDiS: Automated Earth Observation data processing within an event-driven architecture for water monitoring

    by Sebastian Drost, Arne Vogt, Christian Danowski-Buhren, Simon Jirka, Verena Kirstein, Kian Pakzad, Matthes Rieke


  • Open Access

    WaCoDiS: Automated Earth Observation Data Processing within an Event-Driven Architecture for Water Monitoring

    by Sebastian Drost, Arne Vogt, Christian Danowski-Buhren, Simon Jirka, Verena Kirstein, Kian Pakzad, Matthes Rieke

    To ensure an efficient and environmentally friendly water resource management, water management associations need means for efficient water monitoring as well as novel strategies to reduce the pollution of surface and ground water. Traditionally, water management associations operate large sensor networks to suffice their needs for hydrological and meteorological measurement data to monitor and model physical processes within catchments. Implementing a comprehensive monitoring system often suffers from sparse coverage of in-situ data. Due to the evolvement of the Copernicus satellite platforms, the broader availability of satellite data provides a great potential for deriving complementary information from Earth Observation data. Although the number of satellite data platforms that provide online processing environments is growing, it is still a big challenge to integrate those platforms into traditional workflows of users from environmental domains such as hydrology. Thus, in this paper, we introduce a software architecture to facilitate the generation of Earth Observation information targeted towards hydrology. The presented WaCoDiS System comprises several microservices as well standardized interfaces that enable a platform-independent processing of satellite data. First, we discuss the contribution of Earth Observation data to water monitoring and derive several challenges regarding the facilitation of satellite data processing. We then describe our system design with a brief overview about the different system components which form an automated processing pipeline. The suitability of our system is proven as part of a pre-operational deployment for a German water management association. In addition, we demonstrate how our system is capable of integrating satellite data platforms, using the Copernicus Data and Exploitation Platform - Deutschland (CODE-DE) as a reference example.

  • Open Access

    The Impact of Land Cover Data on Rainfall-Runoff Prediction Using an Entity-Aware-LSTM

    by Sebastian Drost, Fabian Netzel, Andreas Wytzisk-Ahrens, Christoph Mudersbach

    The application of Deep Learning methods for modelling rainfall-runoff have reached great advances in the last years. Especially, long short-term memory (LSTM) networks have gained enhanced attention for time-series prediction. The architecture of this special kind of recurrent neural network is optimized for learning long-term dependencies from large time-series datasets. Thus, different studies proved the applicability of LSTM networks for rainfall-runoff predictions and showed, that they are capable of outperforming other types of neural networks (Hu et al., 2018).Understanding the impact of land-cover changes on rainfall-runoff dynamics is an important task. Such a hydrological modelling problem typically is solved with process-based models by varying model-parameters related to land-cover-incidents at different points in time. Kratzert et al. (2019) proposed an adaption of the standard LSTM architecture, called Entity-Aware-LSTM (EA-LSTM), which can take static catchment attributes as input features to overcome the regional modelling problem and provides a promising approach for similar use cases. Hence, our contribution aims to analyse the suitability of EA-LSTM for assessing the effect of land-cover changes.In different experimental setups, we train standard LSTM and EA-LSTM networks for multiple small subbasins, that are associated to the Wupper region in Germany. Gridded daily precipitation data from the REGNIE dataset (Rauthe et al., 2013), provided by the German Weather Service (DWD), is used as model input to predict the daily discharge for each subbasin. For training the EA-LSTM we use land cover information from the European CORINE Land Cover (CLC) inventory as static input features. The CLC inventory includes Europe-wide timeseries of land cover in 44 classes as well as land cover changes for different time periods (Büttner, 2014). The percentage proportion of each land cover class within a subbasin serves as static input features. To evaluate the impact of land cover data on rainfall-runoff prediction, we compare the results of the EA-LSTM with those of the standard LSTM considering different statistical measures as well as the Nash–Sutcliffe efficiency (NSE).In addition, we test the ability of the EA-LSTM to outperform physical process-based models. For this purpose, we utilize existing and calibrated hydrological models within the Wupper basin to simulate discharge for each subbasin. Finally, performance metrics of the calibrated model are used as benchmarks for assessing the performance of the EA-LSTM model.ReferencesBüttner, G. (2014). CORINE Land Cover and Land Cover Change Products. In: Manakos & M. Braun (Hrsg.), Land Use and Land Cover Mapping in Europe (Bd. 18, S. 55–74). Springer Netherlands. https://doi.org/10.1007/978-94-007-7969-3_5Hu, C., Wu, Q., Li, H., Jian, S., Li, N., & Lou, Z. (2018). Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation. Water, 10(11), 1543. https://doi.org/10.3390/w10111543Kratzert, F., Klotz, D., Shalev, G., Klambauer, G., Hochreiter, S., & Nearing, G. (2019). Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets. Hydrology and Earth System Sciences, 23(12), 5089–5110. https://doi.org/10.5194/hess-23-5089-2019Rauthe, M, Steiner, H, Riediger, U, Mazurkiewicz, A &Gratzki, A (2013): A Central European precipitation climatology – Part I: Generation and validation of a high-resolution gridded daily data set (HYRAS), Meteorologische Zeitschrift, Vol 22, No 3, 235–256. https://doi.org/10.1127/0941-2948/2013/0436