# UTCI deviations

It seems that UTCI calculation has some weird deviations happen in the range of Ta(44-46C). Look these examples:
Ta------Tmrt------air velocity-----RH
46-------14--------------6-----------100 = Strongly Cold!!
45-------14--------------6-----------100 = Comfort !!!
44-------14--------------6-----------100 = Strongly Hot!!!
I suggest to omit these from the calculations since these effect on the total hot weather simulations.
Would be great to see other idea here as well.

Hi @Navid, would you be able to point out which component you were using to determine these results?

Looking at the online UTCI calculator I get the following results:

Ta Tmrt air velocity RH UTCI (C°) Comfort
46 14 6 100 179.1 Strongly Hot!!
45 14 6 100 169.4 Strongly Hot!!
44 14 6 100 155.5 Strongly Hot!!

Also, what’s the environmental context where you get 100% RH at 40+C°?!

Well you should change the input Tmrt to delta Tmrt (Tmrt - Ta)

@Navid ,
As @AntoineDao , points out, what component are you using and how are you using it?

The code for the polynomial in Ladybug is the same as the online calculator at utci.org that you cite. It’s just that the component automatically does the calculation from MRT to dmrt for you.

@chris
I mean the main UTCI calculation has some weird deviations happen in the range of Ta(44-46C). Following this, the LB component (Outdoor comfort calculation) does the same.

Hi @Navid

UTCI is a fitted (6th degree) polynomial. As all fitted polynomials it has weird behavior close to its bounds/limits. I know, that it is stated that UTCI works from -50 to +50 C, but in my experience it starts breaking around +45C

Furthermore if your relative humidity is 100% at 45C you might be approaching the limit for vapour pressure as well it might fuck it up even more.

If you want to compute the comfort those extreme environments, you should use a different model. You can either use PET or the original UTCI model.

Currently I am preparing a comparison between PET and UTCI. For this I have looked at the calculation for UTCI in Ladybug “Outdoor Comfort Calculator”. I noticed that the wind speed 10 m in Ladybug has a different range of values (0.05 m/s) than the original code (Range original 0.5 - 17 m/s). Do you know if there are other deviations in the code and how big the influence on the results are?

My second question can certainly be answered by Chris. Is the calculation of PET updated by these guys https://github.com/ladybug-tools/ladybug-comfort/issues/4 ?