Hi Tom,
WXSIM missed again totally a quite heavy thunder we had this morning…
Also for today, WXSIM says no thunder, GFS have for this coordinates xxCAPE at around 1500 J/kg and average Lifted Index at -4
plaintext.txt (2.5 KB)
Hi Tom,
I still have same problem in the Netherlands. Also in the Netherlands is a warning for thunder but WXSIM is showing nothing.
I have set convective sensitivity to 1.3 and also in WRET, but no thunder at all.
also I have set minimum sky cover…(tab 4) for allowing any precipitition at 1.
EDIT: Must say that it isn`t every time it misses thunder in my case. Last week there was thunder and WXSIM placed it in the forecast.
Just saw that again for coming wednesday it is in the forecast. Remains the agressive setting from convective sensitivity at 1.3
also it misses sometimes totally thunder in the forecast while there was strong lightning that forecast period.
Oebel
Did a followup of today’s afternoon’s/evenings forecasts and saw some variations depend on when forecast was runned, yes i know it can’t be blaimed on WXSIM all that, but, the GFS-meteogram i generate every 6 hours have had the CAPE/Lifted-values quite stable for today/tomorrow for days now so its not only GFS what fail here…I has not have same variations at all what have seen in WXSIM…
As note sunset was 22:57
Forecasts runned about 10 minutes before the hour seen (WXSIMATE 20 min before the hour) and value are the thunderamount for next 2 days in plaintext (1 thunder = thunder at one forecast):
12h - showed 1 thunder
15h - showed 1 thunder
18h - showed 3 thunders
21h - showed 0 thunder
24h - showed 2 thunders
What i have mostly seen this season are that WXSIM shows thunders in days 1-4 but when it comes to the day D it has deleted the thunder. Actually i think i have no hits where Wxsim have thunder = also thundering so far this year.
Instead i have hits where WXSIM showed thunder before day D but deleted it when it came to that day + then where thunder but WXSIM haven’t had it at all.
Henkka
Thanks for the continued feedback. I will take it into account. What I would like to do is make it so that the zero setting is, on average, best for the most people. Then, if a certain location’s climate is somehow different, it should be easy to accommodate with the -2 to +2 range I now have on the sensitivity.
Any more observations, from anybody?
Thanks! 
Tom
Just a thought …
Are you northwestern European users using the NW European option, as opposed to North American? I think you should, but I’d be interested to hear what happens if you experiment with the North American indices.
The differences are detailed in manual.doc, but in short, despite considerable overlap (lifted index, Showalter, and some others), European and North American researcheers somewhat independently developed their own indices, and I wanted to take that into account. In particular, the Boyden index was developed for the UK, especially for cooler, cloudier conditions, often involving frontal passages. The KO index was developed in Germany and is also “tuned” for conditions that might be encountered there. However, I’ve run across information that KO is not really any better (and maybe not as good as) LI.
KO would be inappropriate in my regions (SE U.S.), as it would often go totally “off the charts” with values like -15, which would never be reached in northern Europe. Also, we get K index values here that you would never see in Europe.
I think the “NW European” choice is mainly best for somewhat maritime climate areas of Europe, mainly the UK, Ireland, NW France, the Netherlands, and maybe Denmark. I’m less sure about Finland, or continental parts of northern Europe, and I doubt it’s appropriate for Italy. On the other hand, “NW European” might be best for New Zealand, even if it is pretty much at the antipode!
Hope that helps,
Tom
Hi Tom,
I think that for Europe it would be usefull to take a look to the European official storm forecast site (is a joint project by volonteer metereologist)
www.estofex.org
they talk about SWEAT index and more…
http://www.estofex.org/guide/
Sante
Tom,
I did a run and used North American option. There was no difference.
Turned back to NW European.
Now have convective sensitivity set to 2 ( max).
For wednesday there is thunder expected.
Yesterday evening (friday july 1, I think forecast of 7 p.m. and 10 p.m., there was thunder in the forecast for Wednesday.
This morning all thunder is gone again in WXSIM. Still the weatherservice expects thunder.
With set to 2, my lifted index is higher.
Oebel
Will test with North American.
Attached a compare GFS vs WXSIM Plaintext (NW European).
There are rain in GFS what WXSIM ignore totally (!) and then as result (as i see) also ignore the thunder, now with CAPE values above 2000 j/kg in GFS.
Today we had thunder well inside the 80km radius from here, my Boltek picked as max 206 strikes/min, not bad for beeing here.
WXSIM, had at forecast runned at 12h Several thunderstorms. That looked more than ok regarding the weather outside.
Next run, at 15h the thunder was totally gone. And the max in Boltek was 2+ hours later.
Run with North American done, no difference at all
Also a run with sensitivity 1.6 done, the Lifted etc values rised yes, but, again, because wxsim igores the rain it also ignores the thunder. I use btw 25% limit for clouds/rain in wxsim.
henkka
This is really strange.
You arnt getting enough thunder over there but I have too much in the forecast here.
The LI issue has reared its ugly head again and now I am getting -5 LIs.
Main issue from what i see is that WXSIM for some reason reason ignores the rain whats in GFS (for the record, using nomad-servers), for rest of today GFS says 10mm but WXSIM <2mm and GFS has rain for whole day while WXSIM got it to only one 3 hours section.
Currently here are thundering (again) + raining, WXSIM says clear/sunny, GFS says thunder (CAPE ~1000 and LI -4) + rain.
One what i have noted last 3 days with thunder is that no rain = no thunder NOT works in reality, here have been 5 hearble thunders in last 3 days of what it has rained in 3 of them, rest 2 has only had some boom-boom’s.
Do someone have 12.8.6 or 12.8.5 handy to send me?
Got a tought if test with a older version and see if there are any difference…
Henkka
Hi Henkka,
You have mail.
//Oebel
Found a 12.8.7 from middle of May and runned that. It shows muuuch more thunder than 12.8.8 do, now i got thunder for next 4 forecasts in plaintext including also rain in each while with 12.8.8 i got in only 1 forecast.
Today afternoons forecast jumped from “Clear” to “High risk of severe thunderstorms with possible tornados” 8O
If someone have the wxsimupd.exe for 12.8.7 (from May) would be great. I have just a renamed wxsim.exe and can’t change any settings in it (gives me an error)…
Henkka
you got pm !
WXSIM should not be ignoring GFS-based rain, UNLESS certain settings are in effect. There are four things I can think of which would cause the program to omit any GFS-based rain. I will list them here, along with what you should do if you DON’T want it to leave out rain.
(1) There is an option (under Preferences/Settings) to omit mention of trace amounts of precipitation. make sure that’s NOT checked.
(2) There is that relatively new “minimum cloud cover for precipitation” item (adjustable in pro mode). It was needed for some users, and I had originally set it to 40% as a default. The default in 12.8.8 is now just 20%, but in pro mode you can set that to zero, as if the feature were never there.
(3) If you turn the GFS precipitation bias way down, it might omit some of the lighter rain. If it’s at 100, there’s no effect at all.
(4) There’s an option on the main form, at lower right, to Stop AM Rain. That’s useful here in the summer, as virtually all rain this time of year is afternoon thundershowers, and the feature helps refine the timing. If you want to make sure all the rain gets through, DON’T activate that.
Do note that the shower option (generally recommended during the warmer months) breaks the rain up into shorter segments occupying about a fifth of each three hour period. however, that doesn’t stop the rain. Also, I just recalled I may have something in the orographic effects (pro mode only) that could stop rain for downslope situations. Just make sure those settings are zero.
If, after confirming all the settings I just mentioned, you still see rain omitted completely from the forecast (look at both plaintext and rain amounts in the numerical summary), please let me know!
Tom
Tom,
Will re-check those, actually i don’t have access at all to perferences, got an error 380 and also WRET fails so no lastret are created, that with 12.8.8Beta, and have no idea how to fix it (but it runs ok) but from memory:
(1) Checked
(2) Set to 25%
(3) ?
(4) Not checked
orographic: not touched
But, with the GFS-meteogram from ex. yesterday, and wxsim not ignores rain + above settings, how then wxsim gived me zero rain (and no thunder) yesterday?
And soon as runned 12.8.7 gived me lots of thunder (not heavy amounts of rains tought but it hasn’t came here either) and the forecast jumped from clear/sunny to heavy thunder + tornados, that with same data from wxsimate?
Henkka
Got it to work ok,
Settings are now:
(1) Checked
(2) Set to 20%
(3) Where are this?
(4) Not checked
orographic: not touched
Anyway, like i saw was it yesterday, forecast runned now just before 21h with fresh data from wxsimate, no thunder at all, nor with 12.8.7 or 12.8.8Beta
BUT, one thing i saw was temperature rised with 12.8.8 as much as 4C from 28C to 32C, both forecasts runned within 5 minutes.
Getting confusing now…
Runned a new forecast at 21:05 with fresh data and got, again, a slightly different forecast, now there popped up some rain for tomorrow. Also, for tonight in lastret are “Scattered showers or thundershowers possible” shown but no rain.
Time shown in wdata.txt was in all cases 2011-07-03-12
plaintext_12.8.7_2040.txt (1.9 KB)
plaintext_12.8.8_2045.txt (2.16 KB)
plaintext_12.8.8_2105.txt (2.28 KB)
Hi Henkka,
You could set that 20% all the way down to 0%, and it would be just like in earlier versions. The GFS precip bias is on tab #4 under Preferences/Settings. It’s probably set at 95%.
Hope that helps!
Tom