Thanks for the analysis Miguel; I will re-run the model as soon as possible as you suggested.
In the meantime I have some other questions and comments; 1. One hypothesis which we have formed is that there could be a high sensitivity dependence on the forecast cloud cover (cc parameter) within the model. For example in the case provided here, the cc forecast values (in octas) from 17 UTC to 23 UTC on 14th March are : 17Z : 8 18z : 8 19z : 7 20z : 7 21z : 6 22z : 6 23z : 7 The forecast road temperature shoots upwards after 18z as your chart shows, just when the cloud cover values become less than overcast. i.e. the solar flux values at the road surface may become very high due to broken cloud cover. Do you have an opinion on this ? 2. For some locations where we run the model, the roads are in deep canyons which even now are not receiving direct sunlight (even when skies are clear or partly cloudy). Does the model take into account any terrain blockages to direct sunlight ? 3. It seems odd that the model would predict "melting snow" with road temperatures as high as 17 C; wouldn't the snow have to melt first and perhaps the residual water mostly evaporate before the road temperature could get that high ? Perhaps there's some error in the model physics formulation...or perhaps we are interpreting the "melting snow" condition incorrectly ? Thanks in advance Iain -----Original Message----- From: Miguel Tremblay [mailto:[EMAIL PROTECTED] Sent: March 20, 2008 11:09 AM To: Iain Russell Cc: '[email protected]' Subject: Re: [Metro-developers] High pavement forecast temperature with melting snow road condition Hi Iain, You will find in attachment an image (road_obs.png) showing the road forecast (lines) and the observation (circles). Blue is for road temperature; Red is for air temperature; Green is for dew point; The places where you don't see a red circle is because dew point and air temperature are the same (TD is drawn after AT). As you can see, the observed air temperature at the station is around 5°C less than forecasted. I suggest you try to replace the air temperature and dew point in the atmospheric forecast file by those who were observed. Rerun METRo with this enhance atmospheric forecast and look at the result. Tell us the result and, if it not accurate enough and if you still have questions, we will continue our investigation. Regards, Miguel Iain Russell wrote: > > Hi > > > > We are finding that the METRo model (latest stable release, 3.1.0) is > generating a road condition of melting snow with pavement forecast > temperatures in the region of 16-17 C. Input data files can be > provided if required, but as a summary, here are the input atmospheric > forecast conditions , for a site in BC, Canada valid at 2200 GMT on > 14^th March 2008; > > > > at 4 > > td -3 > > ra 0.3 > > sn 4.4 > > ws 7 > > ap 833 > > cc 6 > > > > Pavement temperature of 16-17 C seems too high, for melting snow to be > the road condition (i.e.we think that the snow would have to > completely melt, then the road dry out, before the road temperature > could get that high). > > > > Regards > > Iain > > > > _ > _ > > **Iain Russell** > > Manager, Content Development > > Pelmorex Media Inc. > > The Weather Network http://weather.ca/ > > MeteoMedia http://meteomedia.ca/ > > Tel: 905 829 1159 (1405) > > Fax: 905 829 5502 > > [EMAIL PROTECTED] > > > > ------------------------------------------------------------------------ > > _______________________________________________ > METRo-developers mailing list > [email protected] > https://mail.gna.org/listinfo/metro-developers > _______________________________________________ METRo-developers mailing list [email protected] https://mail.gna.org/listinfo/metro-developers
