Context: Earth’s climate is changing due to human activity. Therefore, the ability to predict future climate change is of utmost importance if we want to reduce human interference with the climate system (climate change mitigation) and our vulnerability to the harmful effects of climate change (climate change adaptation). The general idea of the course is to give students the knowledge and expertise to build their own (simple) climate model – starting from scratch, the students will go step by step toward the full climate model during the semester and learn everything necessary to achieve this goal.
Content: The course focuses on the so-called energy balance model on a 2D spatial grid (in spherical coordinates) that models the earth’s surface temperature throughout the year for a given CO2 value of the atmosphere. It depends on the geography (land, ocean, ice, snow), the albedo, the turbulent diffusion, the heat capacity, and the solar forcing throughout the year from the sun. It is discretized with finite difference methods in space, backward Euler in time, and solved until annual energy equilibrium is reached. The recommended programming language of the course is Python, however, if the students have experience with the Julia programming language, this is an option as well. The content of the course is split into milestones, where for each milestone lectures on the theory (physics, mathematics, software) are followed by a “do-it-yourself” implementation phase, which will be guided and supervised by the lecturers and the tutors of the course. Reference implementations of all milestones are available and will be provided to the students after each milestone. Once the full climate model is implemented and tested, the last step in the course is to investigate the behavior of the model (e.g. regarding CO2, or ice-snow cover, maybe cloud coverage, and its effect on the albedo) and “go wild” with an extension/application free of choice in the very last milestone and present the findings to the rest of the course participants at the end of the semester.
Learning Objective: The students will get a broad education on the different ingredients necessary to make a (simple) climate model: from physics, modeling assumptions, numerical methods, and algorithms to the software and programming aspects. The students will understand the assumptions made to derive the model and thus the limits of the model. They will get a climate model tool, where they know every single code line and thus can experiment and investigate its behavior and use it to study climate evolution/change on their own.
Target Audience: In general, all students that are interested to learn the technical details of a (simple) climate model are welcome. Working as a group is strongly encouraged. The complexity of the course content aims at students in their early Masters or late Bachelor studies. Our aim of the course is to provide all relevant ideas/information/background theory/algorithms and give guidance/help with the software implementation throughout the semester. While the course is open for students from all subjects, some knowledge/strong interest in basic physics, and/or some knowledge/strong interest in computational physics/mathematics/numerical methods and software development with some experience in programming is helpful. This experience might be obtained for instance in courses from Bachelor studies of Mathematics (e.g. second subject in Physics), Computer Science, Physics, Geophysics, Meteorology or in Teachers’ program on Mathematics (e.g. with second subject in Physics).
Literature/Material/Software: All relevant information will be provided as course material during the lecture. The content of the course is partially based on the publication “A NetCDF version of the two-dimensional energy balance model based on the full multigrid algorithm” by Zhuang et al. (Software X, 2017). As a reference for the course, we have implemented the full climate model in our Julia code Klimakoffer.jl (https://github.com/klimakoffer/Klimakoffer.jl), where a short description can be found at https://www.mi.uni-koeln.de/NumSim/2021/09/30/snapshot-numerical-simulations-of-earths-climate/. However, we have re-structured the software for the course into several milestones to make it easier to digest in a step-by-step approach during the semester. These Python (Julia) reference solutions will be provided after each milestone.
Organisation: The course itself will be fully organized via ILIAS, with all relevant information made accessible. After pre-registration, all students will be invited to the course through ILIAS.
Bitte beachten Sie, dass alle weiteren Informationen zu dieser Veranstaltung über ILIAS bereitgestellt werden.
Veranstaltungsort und -zeit:
Mittwoch 12.00-13.30 Uhr, Hörsaal Mathematik (Raum 203)
Donnerstag 12.00-13.30 Uhr, Hörsaal Mathematik (Raum 203)
Kontakt bitte nur über die Gruppen-Email: numsim-group@uni-koeln.de