Global Climate Models (GCMs)

1. Whether ALL the physical processes associated with Conceptual and Mathematical Modelling of GCMs remain based on ‘Physical Laws’; and NO more ‘approximations’ are involved?

2. Whether GCMs simulate Earth’s Climate System using Fine-Grained (finer resolution) Simulation?

3. Whether the current GCMs with Coarser Grid Resolution are able to consider those processes that occur on scales finer than their Grid Resolution (including Cloud Processes)?

4. Feasible to simulate the complex coupling between (A) Earth’s Atmosphere; (B) Ocean [Oceanic Circulation]; (C) Land Surface; and (D) Sea Ice & Glaciers?

5. Feasible to solve such highly non-linear equations with turbulent effects for a strong coupling between Atmosphere, Cryosphere, Hydrosphere & Land-Surface System?

6. Supposed to consider ‘Bio-Geo-Chemical Processes’ as well, on top of ‘Physical Processes’?

7. Are these Simulations fulfill the accuracy and precision required by Key Variables towards Simulating Future Climate Change (where, each equation remains solved @ a fixed time interval between 10 – 30 minutes, towards its evolution from ICs)?

8. What are the consequences, if the response time remains to be much longer for Atmosphere than Ocean?

9. Could these Simulations describe the precise controlling factors that remain responsible to Climate Forcing?

10. In the absence of knowing, how exactly the society will develop in the future, how is it possible to deduce various scenarios of how exactly, the respective radiative properties keep changing, how do we deduce the required information on the radiative properties of the Climate System?

11. Feasible to dynamically down-scale the DATA from GCMs to a higher resolution over the ‘Specific/Limited Area of Interest’ using ‘Regional Circulating Models’?

12. How successful are the Climate Models Simulating El Nino/Southern Oscillation and the North Atlantic Overturning Circulation?

13. How exactly are we validating the Model Results towards checking the accuracy of the model’s representation of the real-world climate system?

14. Whether the Conceptualization on (a) the way the heat gets transported, and the way, the ocean exchanges heat and moisture with the atmosphere (associated with the oceanic circulation model); and (b) the way, the vegetation, soil and snow-cover exchanges energy and moisture with the atmosphere (associated with the land surface model) - are uniformly getting translated into its equivalent Mathematical Model (from Conceptual Model)?

15. With real processes, either being poorly conceptualized or remaining too intricate to consider in the conceptual modeling stage itself, and in turn, the respective Mathematical Model itself remaining approximate, then, what is the very purpose of having sophisticated Numerical Models?

16. In the context of Climate Modelling, Can we expect GCMs to represent processes ‘realistically’, having the resolution of 100 km in the horizontal direction; 1 km in the vertical direction; with a time-stepping of 10 minutes? How about those clouds and rainfall that remain associated with scales, which are smaller than the model resolution (apart from approximation with parameterizations related with clouds and precipitation)? Can we afford to apply the trade-off between model resolution and model complexity here?

17. Do we have ALL the required DATA associated with the processes of atmospheric variations at longer scales (including radiation and air-land and water exchanges) towards Atmospheric and Oceanic Global Climate Modeling that dictate the resulting Spatial and Temporal distributions of Atmospheric conditions, by considering (a) the atmospheric momentum conservation using Newton’s 2nd Law of Motion; (b) the Heat Energy conservation using 1stLaw of Thermodynamics; and (c) the conservation laws for air and water masses?

18. Whether GCMs have clearly reproduced the past climate state that distinguishes the responsible and sensitive factors attributed to climate change remaining anthropogenic; and that resulting from natural causes including solar variations or slow circulations in the ocean or from volcanic eruptions?

19. If GCMs could simulate the past climate in terms of spatial variations of climate conditions of the past (of course, averaged values), why can’t it predict the future climate conditions in the climate system at the end of 1 month, or, 1 year? Why does it remain challenging to simulate over a smaller period? Also, why don’t have a particular (one) climate model that best represents the reality; and why do we have a bunch of models with varying conceptualizations, varying resolutions, range of complexities, range of non-linearities and having a range of parameterizations? If a model cannot include multiple facets (such as considering carbon-cycle, non-linear interactions between clouds and aerosol particles, ocean circulations) simultaneously, then, would it remain to forecast climate accurately (in the absence of deducing the best model so far)?

20. Well, how could we reduce the uncertainties resulting from simplification of conceptual model, simplification of mathematical model, model input parameters, initial and boundary conditions, and parameterizations?

21. Can we take it for granted that the models will always provide ‘the correct answers for the right reasons’, just because, model simulations and data agreed well?

22. Can we completely rely on GCMs, apart from its projection to better understand the climate system?

Suresh Kumar Govindarajan, Professor [HAG]

IIT Madras 14-Dec-2024