Climate projection and weather forecasting use the same equations but differ in their approach. Weather forecasting relies on data assimilation and aims for accuracy over a 10-14 day period, while climate projection is free-running, focusing on long-term statistics over hundreds of years, such as average temperatures or extreme weather events.
The 'window of chaos' refers to the phenomenon where tiny differences in initial conditions, like rounding errors, cause models to diverge significantly over time. In atmospheric models, these errors double every 5-6 days, making accurate predictions beyond 10-14 days nearly impossible.
Climate models typically use grid boxes of about 100 by 100 kilometers. These large scales are necessary due to computational constraints, though finer resolutions would be ideal for capturing smaller-scale processes like local wind patterns.
The 30-minute time step is a balance between computational feasibility and numerical stability. Shorter time steps would be more accurate but computationally expensive, while longer steps could cause the model to become unstable and produce inaccurate results.
Key parameters include density, temperature, pressure, wind speeds in three directions (U, V, W), water vapor content, and cloud fraction. These variables are essential for modeling atmospheric dynamics and energy conservation.
Oceans are often represented using a 'mixed layer' model, which simulates the top 50-100 meters, or by specifying sea surface temperatures. Fully coupling the deep ocean to the atmosphere is computationally expensive and typically not done for long-term climate projections.
Atmospheric gravity waves are generated by various atmospheric processes and propagate across large distances. They contribute significantly to the momentum budget of the jet stream and can influence weather patterns, such as the polar vortex and storm tracks.
Research on the polar vortex focuses on its breakup events, which can shift weather patterns and lead to extreme winter conditions. For tropical cyclones, studies are exploring the relationship between the Intertropical Convergence Zone (ITCZ) and cyclone frequency, suggesting that a more northern ITCZ leads to more cyclones.
Google's balloon project, originally intended for internet access, has provided free, global-scale atmospheric data. This data is being used to study atmospheric gravity waves and other sub-grid scale processes, offering new insights into atmospheric dynamics without the high cost of traditional balloon campaigns.
Geoengineering, such as injecting sulfate aerosols into the stratosphere, requires extensive modeling to understand potential side effects. The atmosphere's multiscale and non-local nature means that interventions could have unforeseen consequences elsewhere, necessitating careful simulation before any real-world implementation.
Climate modeler Aditi Sheshadri) says that while weather forecasting and climate projection are based on similar science, they are very different disciplines. Forecasting is about looking at next week, while projection is about looking at the next century. Sheshadri tells host Russ Altman) how new data and techniques, like low-cost high-altitude balloons and AI, are reshaping the future of climate projection on this episode of Stanford Engineering’s The Future of Everything podcast.
Have a question for Russ? Send it our way in writing or via voice memo, and it might be featured on an upcoming episode. Please introduce yourself, let us know where you're listening from, and share your quest. You can send questions to [email protected]).
Episode Reference Links:
Connect With Us:
Chapters:
(00:00:00) Introduction
Russ Altman introduces guest Aditi Sheshadri, a professor of Earth systems science at Stanford University.
(00:02:58) Climate Projection vs. Weather Forecasting
The differences between climate projection and weather forecasting.
(00:04:58) The Window of Chaos
The concept of the "window of chaos" in climate modeling.
(00:06:11) Scale of Climate Models
The limitations and scale of climate model boxes.
(00:08:19) Computational Constraints
Computational limitations on grid size and time steps in climate modeling.
(00:10:56) Parameters in Climate Modeling
Essential parameters measured, such as density, temperature, and water vapor.
(00:12:18) Oceans in Climate Models
The role of oceans in climate modeling and their integration into projections.
(00:14:35) Atmospheric Gravity Waves
Atmospheric gravity waves and their impact on weather patterns.
(00:18:51) Polar Vortex and Cyclones
Research on the polar vortex and on tropical cyclone frequency.
(00:21:53) Climate Research and Public Awareness
Communicating climate model findings to relevant audiences.
(00:23:33) New Data Sources
How unexpected data from a Google project aids climate research,
(00:25:09) Geoengineering Considerations
Geoengineering and the need for thorough modeling before intervention.
(00:28:19) Conclusion
Connect With Us:
Episode Transcripts >>> The Future of Everything Website)
Connect with Russ >>> Threads) / Bluesky) / Mastodon)
Connect with School of Engineering >>>Twitter/X) / Instagram) / LinkedIn) / Facebook)