trajectory runs get stuck for ERA-interim meteo files

Topics about the HYSPLIT trajectory model.
alicec
Posts: 418
Joined: February 8th, 2016, 12:56 pm
Registered HYSPLIT User: Yes

Re: trajectory runs get stuck for ERA-interim meteo files

Post by alicec »

If you are retrieving ERA-INTERIM through the website, then there is an option to select a date interval at the top of the site.
http://apps.ecmwf.int/datasets/data/int ... vtype=sfc/

I would recommend that you retrieve no more than a week at a time. This will keep your grib files from being over 2GB.
It is the grib files which are input into the conversion program which must be smaller than 2GB.

If you are going to be needing a lot of ERA-INTERIM files then I suggest learning to use their web-api.
https://software.ecmwf.int/wiki/display ... m+examples
ryanxia
Posts: 19
Joined: July 1st, 2016, 1:05 pm
Registered HYSPLIT User: No

Re: trajectory runs get stuck for ERA-interim meteo files

Post by ryanxia »

Hi Alice,

I downloaded ERA files in weekly interval and not it works in trajectory generation!!! Thank you !!!

Also, I have another question, which it looks like you answered before. If I include total precipitation (forecast field) when downloading 2D files, then if I checked meteorology output along trajectory, does the HYSPLIT could include the precipitation data and plot them during trajectory generation? I notice you mentioned this year that now we can download 2D forecast and analysis fields in one file.

In addition, I think I should check step 0 3 6 and time 0 6 12 18; is that correct? My understanding is step 3 and 6 could forecast the precipitation in 3 and 6 h and if I include time 0 6 12 18, then the trajectory output is able to include precipitation rate (mm/hr) for that whole day? Let me know if I am wrong. I notice you once said For retrieving 2d forecast fields I use a time of 00/12 and step of 6/12.

Great thanks,
Ryan
ryanxia
Posts: 19
Joined: July 1st, 2016, 1:05 pm
Registered HYSPLIT User: No

Re: trajectory runs get stuck for ERA-interim meteo files

Post by ryanxia »

Here are something I tried:

I downloaded 2D data (2m Temp, 10m U V wind, and total precipitation): Time 0h 6h 12h 18h and Step 0 3 6. I converted to ARL meteo files. However, I could not run the trajectory generation model. The error information is below:

Model started ...
HYSPLIT4 - Initialization
Last Changed Rev: 854
Last Changed Date: 2017-01-12 10:47:54 -0500 (Thu, 12 Jan 2017)

NOTICE: using namelist file
SETUP.CFG

ERROR metset: meteorological data time interval varies
Changed from 360 min to -3960 min
At day/hr 29 0

It looks like the error message shown and discussed at viewtopic.php?f=14&t=202&p=2779#p2779 is still present. I did not know the reason why I failed. I displayed the data text and showed that the data are in 12h interval.

METEOROLOGICAL PROFILE LISTING ...
Meteorological Profile: ERA.nov16.w4
File start time : 16 11 22 0 0
File ending time: 16 11 28 18 0

___________________________________________________
Profile Time: 16 11 22 0 0
Used Nearest Grid Point ( 181, 91) to Lat: 0.00, Lon: 179.50
2D Fields
SHGT U10M V10M T02M WDIR WSPD
m m/s m/s oC deg m/s
1013 0 -5.5 -1.4 27.2 76.2 5.7
3D Fields
HGTS TEMP UWND VWND WWND RELH TPOT WDIR WSPD
m oC m/s m/s mb/h % oK deg m/s
1000 64.0 25.1 -6.0 -1.4 -0.51 72.9 298.3 76.6 6.1
975 286 22.8 -6.3 -1.5 0.34 79.9 298.2 77.0 6.5
950 513 20.8 -6.5 -1.4 0.77 88.8 298.3 77.7 6.7
925 744 19.7 -6.6 -1.2 1.3 88.1 299.5 79.9 6.7
900 981 18.4 -6.3 -1.2 1.4 90.1 300.4 79.3 6.5
875 1223 18.5 -5.4 -2.0 0.66 61.6 303.0 70.0 5.7
850 1472 19.6 -4.4 -3.2 0.76 34.6 306.7 53.8 5.4
825 1728 19.4 -3.3 -3.7 0.78 25.8 309.1 42.4 5.0
800 1992 18.1 -3.0 -3.4 0.94 22.7 310.5 41.6 4.5
775 2263 16.8 -2.6 -3.0 0.71 22.9 311.9 40.8 4.0
750 2541 15.3 -2.0 -2.9 0 22.3 313.2 35.3 3.5
700 3122 11.9 -0.85 -2.9 0 26.1 315.7 16.3 3.0
650 3737 7.9 0.66 -2.9 0 30.8 317.9 347.1 3.0
600 4393 3.9 1.5 -2.1 0 33.3 320.7 324.3 2.6
550 5095 0.37 1.6 -2.0 0 26.9 324.5 321.4 2.5
500 5853 -3.9 1.4 -1.2 1.8 27.0 328.3 309.9 1.9
450 6677 -8.8 0.98 3.9 1.5 32.5 332.2 194.1 4.0
400 7582 -13.2 2.0 8.2 2.0 15.9 337.8 193.9 8.5
350 8587 -19.2 2.4 8.5 4.2 14.3 342.9 195.7 8.9
300 9715 -27.8 1.7 9.3 2.7 11.5 346.3 190.6 9.4
250 10995 -38.8 2.6 8.9 0.18 21.0 348.4 196.2 9.3
225 11708 -45.4 3.8 7.9 0.34 43.6 349.0 205.4 8.8
200 12483 -52.5 4.5 7.6 0.67 68.9 349.7 210.8 8.9
175 13330 -60.1 5.4 9.3 0.17 93.0 350.8 210.2 10.7
150 14273 -68.2 6.2 12.3 -0.48 98.2 352.6 206.7 13.8

___________________________________________________
Profile Time: 16 11 22 12 0
Used Nearest Grid Point ( 181, 91) to Lat: 0.00, Lon: 179.50
2D Fields
SHGT U10M V10M T02M WDIR WSPD
m m/s m/s oC deg m/s
1013 0 -7.0 -0.18 26.6 88.5 7.0
3D Fields
HGTS TEMP UWND VWND WWND RELH TPOT WDIR WSPD
m oC m/s m/s mb/h % oK deg m/s
1000 74.0 25.8 -7.7 -0.26 0 67.4 299.0 88.1 7.7
975 297 23.7 -8.0 0 0.40 74.2 299.0 90.0 8.0
950 523 21.5 -8.0 0 0.92 80.5 299.0 90.0 8.0
925 755 19.7 -8.0 0.28 1.1 82.4 299.5 92.0 8.0
900 991 18.1 -7.3 0.29 0.73 85.8 300.2 92.2 7.3
875 1232 17.0 -6.8 0.28 0 78.4 301.4 92.3 6.8
850 1481 18.2 -6.4 0.38 -1.8 46.7 305.2 93.4 6.4
825 1736 19.1 -5.1 1.1 -2.6 22.9 308.8 102.6 5.2
800 2000 18.3 -4.1 0.66 -2.3 21.9 310.6 99.3 4.1
775 2271 17.0 -3.5 0 -3.2 22.5 312.1 90.0 3.5
750 2549 15.5 -2.8 -0.62 -2.6 23.5 313.4 77.6 2.9
700 3130 12.0 -1.7 -2.1 -3.1 27.3 315.8 39.0 2.8
650 3746 8.1 -0.17 -3.0 -3.4 33.0 318.1 3.3 3.0
600 4402 3.9 1.3 -2.2 -3.6 32.1 320.7 329.0 2.5
550 5103 -4E-2 3.0 -1.7 -4.0 26.0 324.0 300.1 3.4
500 5860 -4.6 4.2 -1.6 -3.6 25.2 327.5 290.8 4.5
450 6680 -10.1 4.1 0.91 -2.0 38.0 330.6 257.5 4.2
400 7580 -14.7 4.9 4.9 -1.2 32.0 335.9 224.8 6.9
350 8580 -20.4 3.7 7.7 -0.96 24.6 341.3 205.5 8.5
300 9702 -28.9 1.3 13.2 -1.4 31.8 344.7 185.6 13.3
250 10979 -39.4 4.7 11.5 -1.2 69.0 347.6 202.1 12.4
225 11691 -45.6 5.5 8.2 -0.15 78.6 348.7 213.7 9.9
200 12466 -52.4 4.1 8.6 0.77 58.9 349.9 205.6 9.5
175 13314 -59.7 3.1 11.8 1.7 50.0 351.4 194.9 12.2
150 14258 -67.9 5.8 14.7 2.3 84.1 353.2 201.7 15.8


I also tried downloading time 0h 6h 12h 18h and step 0 and 12. However, it still could not work with the same error information. Any caveat or thread?

Thanks,
Ryan
alicec
Posts: 418
Joined: February 8th, 2016, 12:56 pm
Registered HYSPLIT User: Yes

Re: trajectory runs get stuck for ERA-interim meteo files

Post by alicec »

You need to make sure the times for the 2D fields match those of your pressure fields.
For pressure fields (if you are downloading from the web) you can get analysis fields every 6 hours (00,06,12,18)

For the 2D analysis fields use a time of 00/06/12/18 and step of 0.

For the 2D forecast fields use time of 00/12 and step of 6 and 12. This means that
for the time of 00 you get forecast fields at 06 and 12 and
for the time of 12 you get forecast fields at 18 and 00.


Then they will both have output every 6 hours which will be same as pressure fields.

If you are use time of 0 6 12 18 and step of 0 and 12 that would mean for
time 0 you get output at 0 and 12
time 6 you get output at 6 and 18
time 12 you get output at 12 and 00
time 18 you get output at 18 and 6.
The converter program isn't meant to handle that.

You want to provide the converter program with output (either analysis or forecast) for each variable at only one time and
with time that match eachother in all the files.
ryanxia
Posts: 19
Joined: July 1st, 2016, 1:05 pm
Registered HYSPLIT User: No

Re: trajectory runs get stuck for ERA-interim meteo files

Post by ryanxia »

Hi Alice,

I am successful to do this.
I download the 2D file at hour 0/12 and step 6/12 and selected 2m temp, 10m u/v wind and total precipitation. I should not select step 0. It looks like the "forecast" file also include the analysis field data.

Stream:Atmospheric model
Type:Forecast
Dataset:interim_daily
Step:6, 12
Version:1
Grid:1.0° x 1.0°
Time:00:00:00, 12:00:00
Date:20170101 to 20170107
Type of level:Surface
Parameter:10 metre U wind component, 10 metre V wind component, 2 metre temperature, Total precipitation
Class:ERA Interim

The final ERA meteo files converted by HYSPLIT could be run for back-trajectory with rainfall rate data along trajectory. It is supposed that this ERA files are just like GDAS and that I could run them on python version of HYSPLIT.

Thanks. And welcome to comment me if I was wrong somewhere (I am not atmospheric expert).

Ryan
ryanxia
Posts: 19
Joined: July 1st, 2016, 1:05 pm
Registered HYSPLIT User: No

Re: trajectory runs get stuck for ERA-interim meteo files

Post by ryanxia »

Hi Alice,

I am so sorry for that I found new problems about my ERA meteo files.

As I posted above, I use 2D data from 0/12 hour and 6/12 step (2m temp, 10m U/V wind, total precip). The converted ERA meteo files have missing parts.

For example, for the ERA meteo file covering Nov 2016 week 4 (11/22-11/28), the 2D field for the first day 0 hour are all zero:

Profile Time: 16 11 22 0 0
Used Nearest Grid Point ( 181, 91) to Lat: 0.00, Lon: 179.50
2D Fields
SHGT TPPT U10M V10M T02M WDIR WSPD
m mm m/s m/s oC deg m/s
1013 0 0 0 0 0
3D Fields
HGTS TEMP UWND VWND WWND RELH TPOT WDIR WSPD
m oC m/s m/s mb/h % oK deg m/s
1000 64.0 25.1 -6.0 -1.4 -0.51 72.9 298.3 76.6 6.1
975 286 22.8 -6.3 -1.5 0.34 79.9 298.2 77.0 6.5
950 513 20.8 -6.5 -1.4 0.77 88.8 298.3 77.7 6.7
925 744 19.7 -6.6 -1.2 1.3 88.1 299.5 79.9 6.7
900 981 18.4 -6.3 -1.2 1.4 90.1 300.4 79.3 6.5
875 1223 18.5 -5.4 -2.0 0.66 61.6 303.0 70.0 5.7
850 1472 19.6 -4.4 -3.2 0.76 34.6 306.7 53.8 5.4
825 1728 19.4 -3.3 -3.7 0.78 25.8 309.1 42.4 5.0
800 1992 18.1 -3.0 -3.4 0.94 22.7 310.5 41.6 4.5
775 2263 16.8 -2.6 -3.0 0.71 22.9 311.9 40.8 4.0
750 2541 15.3 -2.0 -2.9 0 22.3 313.2 35.3 3.5
700 3122 11.9 -0.85 -2.9 0 26.1 315.7 16.3 3.0
650 3737 7.9 0.66 -2.9 0 30.8 317.9 347.1 3.0
600 4393 3.9 1.5 -2.1 0 33.3 320.7 324.3 2.6
550 5095 0.37 1.6 -2.0 0 26.9 324.5 321.4 2.5
500 5853 -3.9 1.4 -1.2 1.8 27.0 328.3 309.9 1.9
450 6677 -8.8 0.98 3.9 1.5 32.5 332.2 194.1 4.0
400 7582 -13.2 2.0 8.2 2.0 15.9 337.8 193.9 8.5
350 8587 -19.2 2.4 8.5 4.2 14.3 342.9 195.7 8.9
300 9715 -27.8 1.7 9.3 2.7 11.5 346.3 190.6 9.4
250 10995 -38.8 2.6 8.9 0.18 21.0 348.4 196.2 9.3
225 11708 -45.4 3.8 7.9 0.34 43.6 349.0 205.4 8.8
200 12483 -52.5 4.5 7.6 0.67 68.9 349.7 210.8 8.9
175 13330 -60.1 5.4 9.3 0.17 93.0 350.8 210.2 10.7
150 14273 -68.2 6.2 12.3 -0.48 98.2 352.6 206.7 13.8

Also the last day (here it is 11/29 0h as I include 12h forecast field from time 12h for 11/28), 3D fields are all zero:

Profile Time: 16 11 29 0 0
Used Nearest Grid Point ( 181, 91) to Lat: 0.00, Lon: 179.50
2D Fields
SHGT TPPT U10M V10M T02M WDIR WSPD
m mm m/s m/s oC deg m/s
1013 0 0 -5.9 1.0 26.2 99.6 6.0
3D Fields
HGTS TEMP UWND VWND WWND RELH TPOT WDIR WSPD
m oC m/s m/s mb/h % oK deg m/s
1000 0 0 0 0 0 0 273.2 99.6 6.0
975 0 0 0 0 0 0 275.1 99.6 6.0
950 0 0 0 0 0 0 277.2 99.6 6.0
925 0 0 0 0 0 0 279.3 99.6 6.0
900 0 0 0 0 0 0 281.5 99.6 6.0
875 0 0 0 0 0 0 283.8 99.6 6.0
850 0 0 0 0 0 0 286.2 99.6 6.0
825 0 0 0 0 0 0 288.6 99.6 6.0
800 0 0 0 0 0 0 291.2 99.6 6.0
775 0 0 0 0 0 0 293.8 99.6 6.0
750 0 0 0 0 0 0 296.6 99.6 6.0
700 0 0 0 0 0 0 302.5 99.6 6.0
650 0 0 0 0 0 0 309.0 99.6 6.0
600 0 0 0 0 0 0 316.1 99.6 6.0
550 0 0 0 0 0 0 324.1 99.6 6.0
500 0 0 0 0 0 0 333.1 99.6 6.0
450 0 0 0 0 0 0 343.2 99.6 6.0
400 0 0 0 0 0 0 355.0 99.6 6.0
350 0 0 0 0 0 0 368.8 99.6 6.0
300 0 0 0 0 0 0 385.4 99.6 6.0
250 0 0 0 0 0 0 406.1 99.6 6.0
225 0 0 0 0 0 0 418.5 99.6 6.0
200 0 0 0 0 0 0 432.8 99.6 6.0
175 0 0 0 0 0 0 449.7 99.6 6.0
150 0 0 0 0 0 0 470.0 99.6 6.0

Because of this, the monthly daily run will get stuck too. It seems that using the mixture of analysis and forecast fields are very hard to be converted to meteo files that are able to be run for trajectory on HYSPLIT. Do you still have caveat or suggestions to help me? I just want to use ERA files just like GDAS.

Best,
Ryan
alicec
Posts: 418
Joined: February 8th, 2016, 12:56 pm
Registered HYSPLIT User: Yes

Re: trajectory runs get stuck for ERA-interim meteo files

Post by alicec »

Right - you can't get the forecast field time period to exactly match the analysis field time period.
The forecast is missing for the first time period because the first time is at the step of 6.
The forecast also has an "extra" time period at the end. so it will be shifted with respect to the analysis fields.

This is what I do when I am using ECMWF data.

1. download 2D forecast fields with time of 00/12 and step of 06/12
2. download 2d analysis fields with time of 00/06/12/18 and step of 0
3. download pressure fields with time of 00/06/12/18 and step of 0

Download all data for one day before and one day after you actually need it.

4. Use ecm2arl to convert the three grib files (and the invariant grib file if you are using it) to arl format.
5. use xtrct_time to remove the first and last time period from the arl file.
Cggc
Posts: 8
Joined: March 28th, 2018, 1:21 am
Registered HYSPLIT User: Yes

Re: trajectory runs get stuck for ERA-interim meteo files

Post by Cggc »

If you are still having problems, check the start and end dates and times of your ERA metdata files. I found that if my trajectory went beyond these bounds then I too would get stuck.

I have a few ERA metdata files. From 2013 -2009 ERA-Interim. If you have a very large online server and would like these, let me know. I can share them. They are already converted for HYSPLIT use. The largest file is about 750MB, so I feel your pain with the large file sizes.

CG
Post Reply

Return to “Trajectory Model”