
Dr. Benedikt Schulz
-
by appointment
-
Kollegiengebäude Mathematik (20.30)
2.012
+49 721 608 45261
-
benedikt.schulz2@kit.edu
-
Karlsruhe Institute of Technology (KIT)
Institute of Stochastics
Englerstraße 2
76131 Karlsruhe
Within project C5 of the Transregional Collaborative Research Center Waves to Weather („Dynamical feature-based ensemble postprocessing of wind gusts within European winter storms“) I am working on methods for statistical postprocessing of weather forecasts. We focus on ensemble forecasts of wind gusts and apply modern machine learning methods like neural networks.
Further information regarding our research can be found on the following websites:
- Press release of the KIT on forecasting wind gusts with AI (15/03/22).
- Interview in the "Campus-Report" series on forecasting wind gusts with AI (05/07/22; German only).
- Article on the homepage of the IMK-TRO on the distinction of high-wind areas in winter storms (08/08/22).
Statistical Postprocessing of Real-Time Forecasts
The IMK-TRO (Institute of Meteorology and Climate Research, Department Troposphere Research) displays various meteorological forecasts on the KIT-Weather portal, including operational (near) real-time forecasts of several meteorological variables in many European cities, for instance Karlsruhe. In collaboration with the Working Group “Atmospheric Dynamics“, we statistically postprocess the real-time forecasts for selected German cities. The statistically postprocessed forecasts can be found here.
Semester | Titel | Typ |
---|---|---|
Summer Semester 2023 | Forecasting: Theory and Practice II | Lecture |
Winter Semester 2022/23 | Forecasting: Theory and Praxis | Lecture |
Summer Semester 2022 | Time Series Analysis | Lecture |
Winter Semester 2021/22 | Einführung in die Stochastik | Lecture |
Preprints
- Höhlein, K., Schulz, B., Westermann, R. and Lerch, S. (2023): Postprocessing of Ensemble Weather Forecasts Using Permutation-invariant Neural Networks. Preprint, available at arXiv:2309.04452.
- Schulz, B. and Lerch, S. (2022): Aggregating distribution forecasts from deep ensembles. Preprint, available at arXiv:2204.02291.
Publications
- Eisenstein, L., Schulz, B., Pinto, J. G., and Knippertz, P. (2023): Identification of high-wind features within extratropical cyclones using a probabilistic random forest - Part 2: Climatology over Europe, Weather and Climate Dynamics, 4, 981-999. doi.org/10.5194/wcd-4-981-2023.
- Ageet, S., Fink, A., Maranan, M. and Schulz, B. (2023): Predictability of Rainfall over Equatorial East Africa in the ECMWF Ensemble Reforecasts on short to medium-range time scales. Weather and Forecasting, in press. doi:10.1175/WAF-D-23-0093.1.
- Gneiting, T., Lerch, S. and Schulz, B. (2023): Probabilistic solar forecasting: Benchmarks, post-processing, verification. Solar Energy, 252, 72-80. doi:10.1016/j.solener.2022.12.054.
- Eisenstein, L., Schulz, B., Qadir, G. A., Pinto, J. G., and Knippertz, P. (2022): Identification of high-wind features within extratropical cyclones using a probabilistic random forest - Part 1: Method and case studies, Weather and Climate Dynamics, 3, 1157-1182. doi.org/10.5194/wcd-3-1157-2022.
- Schulz, B. and Lerch, S. (2022): Machine learning methods for postprocessing ensemble forecasts of wind gusts: A systematic comparison. Monthly Weather Review, 150 (1), 235-257. doi:10.1175/MWR-D-21-0150.1.
- Maier-Gerber, M., Fink, A., Riemer, M., Schoemer, E., Fischer, C. and Schulz, B. (2021): Statistical-Dynamical Forecasting of Sub-Seasonal North Atlantic Tropical Cyclone Occurrence. Weather and Forecasting, 36 (6), 2127-2142. doi:10.1175/WAF-D-21-0020.1.
- Schulz, B., El Ayari, M., Lerch, S. and Baran, S. (2021): Post-processing numerical weather prediction ensembles for probabilistic solar irradiance forecasting. Solar Energy, 220, 1016-1031. doi:10.1016/j.solener.2021.03.023.
Conferences and workshops
- MathSEE Symposium 2023, Karlsruhe, talk: Aggregating distribution forecasts from deep ensembles.
- VALPRED 4 (2023), Aussois (FR), mini-course: Statistical and Machine Learning Methods for Postprocessing Ensemble Weather Forecasts.
- EMS Annual Meeting 2022, Bonn, talk: Machine learning for postprocessing ensemble forecasts of wind gusts with a focus on European winter storms. doi.org/10.5194/ems2022-271.
- 42nd International Symposium on Forecasting (ISF 2022), Oxford (UK), talk: Aggregating distribution forecasts from deep ensembles. Youtube (session recording).
- EGU General Assembly 2022, Vienna (AT), talk: Machine learning for postprocessing ensemble forecasts of wind gusts with a focus on European winter storms. doi.org/10.5194/egusphere-egu22-869.
- 9th HKMetrics Workshop (2022), Mannheim, talk: Aggregating distribution forecasts from deep ensembles.
- ESA-ECMWF workshop 2021, online, poster: Machine Learning Methods for Postprocessing Ensemble Forecasts of Wind Gusts: A Systematic Comparison.
- VALPRED 3 (2021), Aussois (FR), talk: Machine learning methods for postprocessing ensemble forecasts of wind gusts.
- 3rd NOAA Workshop on Leveraging AI in Environmental Sciences (2021), online, talk: Machine learning methods for postprocessing ensemble forecasts of wind gusts: A systematic comparison. Youtube (session recording).
- International Symposium on Forecasting (ISF) 2021, online, talk: Calibrating and combining probability forecasts.
- EGU General Assembly 2021, online, vPICO: Statistical and machine learning methods for postprocessing ensemble forecasts of wind gusts. doi.org/10.5194/egusphere-egu21-1326.
- ICCARUS 2021, online, talk: Statistical and machine learning methods for postprocessing ensemble forecasts of wind gusts.
- Practical Operational implementation of Statistical Post-Processing for ensemble forecasts (2020), online, poster: Statistical post-processing of near-real-time ICON ensemble forecasts.
- VALPRED 2 (2020), Aussois (FR), talk: Calibrating and combining probability forecasts.