I noticed the last few nights that wind speeds being forecast too high are causing night time min temp errors. Its common on clear nights here for winds to drop calm and temps to drop lower than wxsim is forecasting. For example tonight a min temp of 3c is forecast with winds gusting to 30km/h but currently its 2.4c with winds calm, This is at 21.40 so i would expect temps to drop below freezing.
If you compare my website forecasts at www.waterfordcityweather.com to a station 40 miles away carlowweather.com it looks like the Carlow forecast is a lot more realistic especially with regards night time temps.
I’m wondering is there any settings i can adjust to make the forecasts more accurate. Actually thinking about it some more i think i have always had these issues with temps especially in Winter time.
Hmmm … this can be a tricky issue. You’re apparently getting a good inversion keeping the winds from making it down to the surface. Those winds are adjusted some off of the GFS data, but perhaps not enough in this case.
Two questions:
(1) How is your anemometer sited and mounted?
(2) What did your low temperature turn out to be, and did it stay calm all night?
Tom
Not sure if this thread is concerning valley effect temps, but, I suffer the valley effect temps not picked up by WXSIM. Certainly the GFS data does not capture micro climate valley effects, but, looking at the WXSIM forecasts it does calculate a valley effect temperature. It would be nice if there was an option in WXSIM to introduce or add in a percentage of the WXSIM calculated valley effect to the forecast temperature. The user could select what percentage to add in. Just a thought…
Thanks
James
Hi James,
If you are running manually and have professional mode, you can change the urban heat island setting to a lower number to closely simulate a valley effect. If you take it all the way to 0 (not usually recommended), the main forecast becomes the fully “valley” temperature. I’m nervous about people just randomly messing with it when it’s an integral part of the customization process, so that’s why I’ve might it slightly inconvenient to mess with, but I also wanted to provide flexibility for cases like you’re describing. Let me know if that helps!
Tom
Hi Tom
I’ve talked to you before about this and, from what you said then and now, I hesitate to make any changes. Certainly, one reason is that low temps in the Tucson environment are hard to forecast and the average low forecast by WXSIM is not bad. Regardless, it’s frustrating to see actual temps much lower than forecast temps. Especially, on those occasions when my station is colder than the County stations near to me but at an altitude of 9000+ ft. with my station at 2765 ft. I still need to do a science project to better understand the data. If I ever complete that project, I’ll email the data to you and see what you think.
Keep up the good work!
Sincerely,
James
Hi Tom
Min temp on the night in question was 0c so 3c below what was forecast.
With regards the Anemometer its at 30ft in clear air from all directions.
Mike