How to improve the resolution of meteorological data when calculating large-scale spatial Tmrt？

[I-T-H]

I am trying to calculate Tmrt and human thermal comfort at the city scale, but it seems that SOWEIG cannot input different meteorological data based on geographical location and calculate it at a time. It seems that SOWEIG can only input meteorological data for specific points, which means you have to calculate the whole city based on the data of one point. Is there any way to improve the resolution of it?

[biglimp]

As you seem to have plans to calculate Tmrt for a whole city, you need to divide your calculations in sub-domains. Then each sub-domain can use specific input forcing. See the paper for inspiration: https://link.springer.com/article/10.1007/s00484-025-03030-2

[I-T-H]

Thanks! It seems that this article divided the area into 500 m x 500 m sub-grid domains.
However, I plan to simulate the Tmrt of the whole city for about a year, which means it will generate alot of data. So I think python is a must.
One another question: In the metdata processor, I chose the "Keep year from EPW-file" with no effect. The year is still 1985. Is there anyway to fix it? It is quite confusing if large amount data is processed.

[I-T-H]

However, this article didnt clarify the resolution of the meteorological data it used.

[biglimp]

The reason I recommended this article was to demonstrate that you need to divide your data into sub-domain if you want to calculate Tmrt for a whole city. If you try to do it for a whole city, your computer will end up with memory issues. When you calculate in sub-domains, you can use different forcing data for each domain.

As EPW files are retrieved from different years, the year column in a EPW file consist of a variation of years. Hence, we just set on year, in this case 1985. You can change that to any year you like.