How about a more fine resolution Regional Climate Models (RCMs)?

Miyuru Bandara Gunathilake , RCMs are derived from GCMs, right! I'm willing to know any procedures or techniques that aren't related to GCMs.

To respond to your questions, starting with the first question the answer is no, and the answer of your second question is no

GCMs are the only tools currently available to predict future evolution of global climate. However, their coarse spatial resolution (1000km by 1000km or 100km by 100km) limit the application of their output in impact evaluation of local or regional climate. Therefore GCMs output need to be downscaled to obtain high resolution climate simulation

Dear Philbert Modest Luhunga , the answer to my first question is

1. Based on analogies with different climatic zones or historical time periods

2. From GCMs using simple manipulation of current climate observations(E.g.: Change Factor Methodology)

3. From GCMs using more sophisticated statistical and dynamical downscaling methodologies.

This is what I have in mind. I could not get a broader idea of what other explanation I can get for the question I've asked. Are you able to understand what I'm trying to convey?

P.S: Thanks for responding!

1. It is not realistic to forecast/predict the climate of location by considering the linear relationship between historical and future climate conditions. If you do that you neglect changes in the main climate components which is not realistic. GCMs consider the future evolution of components of climate system. By the way how would you know the analogies without the use of GCMs forecast. Therefore the GCM remain the only tool available to diagnose the future climate. We use analogues year in seasonal forecasting but some times we miss to have seasons that are exactly the same

2. Yes you can do that and technically is called statistical downscaling-compare large atmospheric field with local atmospheric condition to construct some transfer function

3. yes possible to use sophisticated statistical downscaling/dynamical downscaling

1. I agree that it's not realistic and can't be considered as a reliable analysis. It's just a rough estimate method.

What I feel is that we keep ending up at GCMs, RCMs, downscaling and nothing more. It's hard to digest that I'm unable to come up with methods apart from three mentioned in the former line.

No, but machine learning methods have been applied to weather forecasting in test cases with some success: see https://www.geosci-model-dev.net/12/2797/2019/gmd-12-2797-2019.pdf

Charles Turner neural network is another type of statistical downscaling

Dear Hitesh,

The GCM are the most modern, and, which is more important, physically consistent way to describe the atmosphere - in the past, present and the future. Applying RCMs as downscaling tool with driver GCM (i.e. for the IC and BC) is dynamically coherent way to obtain regional details.

However, other ways exists also: Statistical downscaling. See for example

https://www.nature.com/articles/nclimate3393#access

and the citations therein.

The following link shows also the product of the implementation of such technique

http://chelsa-climate.org/

Regards: Hristo

You can retrieve the Regional Climate Model (RegCMs) datasets for past,present and future events. They also posses local climate variables for climatic studies.

Weather Forecast Models can be used as alternatives to GCMs. But they have limitations as well.

This paper on "Representative general circulation models selection and downscaling of climate data" might be helpful:

https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/joc.6447

Plz follow these websites

https://cordex.org/2019/

http://cccr.tropmet.res.in/home/cordexsa_datasets.jsp

Malay Kumar Pramanik, thank you for sharing those links. They are useful but not quite answer my question.