Questions on simulations using the full forcing possibility

Discussions about the Version 4 concerning the core model simulations
ADrebs
Posts: 22
Joined: Sun Sep 04, 2016 7:42 pm
Contact:

Questions on simulations using the full forcing possibility

Postby ADrebs » Sun Aug 11, 2019 8:01 pm

Hi ENVI-met team,
I am making research on bio-thermal outdoor condition during July, 31st until August, 2nd 2018 in Vantaa, Finland (60° 17', 25° 3'), 72 hours. For the meteorological forcing observation I use the radiation data (SWDir, SWDiff and LWin from the Finnish Meteorological Institute headquarter about 10 kilometres to the south (quality controlled) beside in-situ observation of air temperature, relative humidity, wind speed and direction. The creation of the .fox-file was a successfully straight-through process. The data appearance in the .fox-file was checked.
After the 72-hour-simulation run the ENVI-met Bio-Met module was used to calculate the human thermal indices (PMV, PET, UTCI, and SET) at 1.4 m height.
By exploring the results I do not understand the high SWin values presented in figure 1, timestamp 1.8.2018, 20:00. According the forcing file the local sunset was approximately around this time (20:00 - 20:30, figure 2). I made a time series of one grid in the area of interest for SWin shown in figure 3. Using these results to calculate for example the PET thermal index is shown in figure 4. The PET values are fare beyond a bio-meteorological definition of 'extreme thermal heat stress'.
What went wrong in the calculation procedure or my way to produce the data?

A-)
Attachments
Figure_4_PET_Heureka_FF_Sommer_2018.png
Figure_3_Leonardo_SWin_Heureka_FF_Sommer_2018.png
Figure_2_FF_01082018_Heureka_Sommer_2018.png
Figure_1_SWDir_Heureka_FF_Sommer_2018.png

Tim
Posts: 618
Joined: Mon May 15, 2017 3:46 pm

Re: Questions on simulations using the full forcing possibility

Postby Tim » Mon Aug 12, 2019 7:52 am

Hi,

I think there are two things about the radiation:
1. The longwave radiation values are too high. They shall represent the downward longwave radiation coming from the sky. Values around 250-300 W/m² are expected.
2. The SW radiation however is not affected by that. Since the input SW radiation values look good, there seems to be a wrong recalculation of the values during the simulation. I'd thus suggest that the parameter whether the values are measured horizontally is not set correctly. You may fix that and see if the output radiation values are then correct.

Best regards,
Tim

ADrebs
Posts: 22
Joined: Sun Sep 04, 2016 7:42 pm
Contact:

Re: Questions on simulations using the full forcing possibility

Postby ADrebs » Mon Aug 12, 2019 8:28 am

Hi Tim,

the LWin radiation I used is measured at the FMI main building and quality checked, that I should be sure, that they are right. In the attachment a screen shot from less than half an hour ago from the site I make the research on. LWin values >350W/m2 are more the rule than the exception, times are UTC.

I run the simulation for horizontal and inclined SWin, the results are the same.

As I cannot look into the software, but see the shape of the outcome graph for SWin, it came into my mind, is there an algebraic sign error in the calculation?

A-)
Attachments
iscape_heureka_12082019.png

Tim
Posts: 618
Joined: Mon May 15, 2017 3:46 pm

Re: Questions on simulations using the full forcing possibility

Postby Tim » Mon Aug 12, 2019 8:48 am

Hi,

Ah, I didn't check the legend for longwave radiation but for shortwave radiation. That's why I thought the longwave values to be too high (at 700). With values up to 400 they are definitely fine, yes.

I will check that issue with some simulations.

Best regards,
Tim

Tim
Posts: 618
Joined: Mon May 15, 2017 3:46 pm

Re: Questions on simulations using the full forcing possibility

Postby Tim » Mon Aug 12, 2019 9:13 am

Hi,

Could you zip and upload the csv or fox file?

Best regards,
Tim

ADrebs
Posts: 22
Joined: Sun Sep 04, 2016 7:42 pm
Contact:

Re: Questions on simulations using the full forcing possibility

Postby ADrebs » Mon Aug 12, 2019 9:21 am

Hi Tim,

here the zipped .fox-file,

A-)
Attachments
ForcingFileHeurekaJuliAugust2018.zip
(137.35 KiB) Downloaded 4 times

helge
Posts: 101
Joined: Mon May 15, 2017 12:27 pm

Re: Questions on simulations using the full forcing possibility

Postby helge » Mon Aug 12, 2019 10:57 am

Dear A-)

We had a look at your FOX-Files: the 20:00:00 and 20:30:00 values from 02th August were missing.
We strongly believe your values are in the "lambert" format. The radiation data is not relative to a horizontal surface but rather to a surface that is always perpendicular to the sun.

Furthermore we found a bug where the information about the reference surface for the radiation is only read from the 1st Jan 00:00:00 - we will fix this bug in the next regular release.

We adjusted your forcing file (see attachment) so it should now yield the correct values.

Also, please make sure that you have the correct reference timezone longitude (I believe 30° E in your case) and georeference in the INX -File. Like all models ENVI-met does not use the day light saving time.

Best
Helge
Attachments
HEUREKA_lambert.zip
(118.03 KiB) Downloaded 5 times

ADrebs
Posts: 22
Joined: Sun Sep 04, 2016 7:42 pm
Contact:

Re: Questions on simulations using the full forcing possibility

Postby ADrebs » Mon Aug 12, 2019 12:17 pm

Hi Helge,

my original .csv-data file contains the two missing rows and I have no explanation for that why they dropped out. So, thank you for the correction.

Do I have now to wait for a new release?

And yes, the time zone is set to CET/ UTC+2 with the reference longitude to 30.

A-)

Tim
Posts: 618
Joined: Mon May 15, 2017 3:46 pm

Re: Questions on simulations using the full forcing possibility

Postby Tim » Mon Aug 12, 2019 12:40 pm

Hi,

Okay!
No, you do not have to wait since all timesteps (even the invalid first timestep which is read unfortunately) hold now the lambert information and will be treated correctly.

Best regards,
Tim

ADrebs
Posts: 22
Joined: Sun Sep 04, 2016 7:42 pm
Contact:

Re: Questions on simulations using the full forcing possibility

Postby ADrebs » Mon Aug 12, 2019 9:00 pm

Hi, Tim, Helge,

I continued to test my simulations with the forcing file you provided, and with the standard non-forced option for clear-sky condition. Due to my restricted hardware speed (I don't complain about that), I just finished the 8-hour morning simulations. I started to compare SWin outcomes for forced (Figure 1) and unforced (Figure 2) conditions. As you can see the patterns in colour and shape for both are quite similar. The SWin values instead show still a great difference to the disadvantage of the full-forced simulation. Calculating the PET index based on these values is shown in Figure 3 and 4. The difference in the PET value distribution in these two figures is due to the constant wind speed (5m/s) and direction (225°) in the non-forced simulation. Nevertheless, the non-forced simulation produces still more realistic PET values that the full-forced.

Can you explain me why?

A-)
Attachments
Figure_4_PET_Heureka_notFF_Sommer_2018_0800.png
Figure_3_PET_Heureka_FF_Sommer_2018_0800.png
Figure_2_SWDir_Heureka_notFF_Sommer_2018_0800.png
Figure_1_SWDir_Heureka_FF_Sommer_2018_0800.png

helge
Posts: 101
Joined: Mon May 15, 2017 12:27 pm

Re: Questions on simulations using the full forcing possibility

Postby helge » Tue Aug 13, 2019 1:40 pm

Dear A-),

Thank you for the feedback we will have a look into it by simulating a small test area and will get back to you.

Best Helge


Return to “ENVI-met 4 (Main Simulation Model)”

Who is online

Users browsing this forum: No registered users and 1 guest