Hi Miguel!!

I've been dealing with that issue as well: in my case, I developed an automatic selection method to detect sea breezes and for that, I had to reject days with cold fronts, which sometimes can be confused with sea-breeze events.

Thus, I recommend you to have a look at the 850-hPa level (from reanalysis data for instance) to differentiate the different phenomena: the sea breezes cause a wind change at surface very similar to cold fronts, whereas at 850-hPa the signal is evident only in cold fronts (wind-direction change and decrease of the temperature). Except for some isolated cases in tropical latitudes, in which the sea-breeze layer can be thicker, in sea-breeze cases at 850 hPa only the return flow is observed, which cannot be confused with a cold front.

The same is applied for some valley circulations and katabatic flows, driven by gravity currents, which cause the wind shift at the surface.

Regards from Spain,

Jon

Here I am just guessing, Miguel.

(Meteorology is of some superficial interest to me through my work on renewable energy, but that's about it !!)

But as a first guess, I would suggest that you may need at least two other weather parameters apart from the wind pattern:

If I were you, I would not overlook the fact that wind profiles can include turbulence effects, and that may make your met measurements sometimes quite tacky to interpret depending on wind alone !!

Cheers !

-Sanjay

Hi Miguel!!

I've been dealing with that issue as well: in my case, I developed an automatic selection method to detect sea breezes and for that, I had to reject days with cold fronts, which sometimes can be confused with sea-breeze events.

Thus, I recommend you to have a look at the 850-hPa level (from reanalysis data for instance) to differentiate the different phenomena: the sea breezes cause a wind change at surface very similar to cold fronts, whereas at 850-hPa the signal is evident only in cold fronts (wind-direction change and decrease of the temperature). Except for some isolated cases in tropical latitudes, in which the sea-breeze layer can be thicker, in sea-breeze cases at 850 hPa only the return flow is observed, which cannot be confused with a cold front.

The same is applied for some valley circulations and katabatic flows, driven by gravity currents, which cause the wind shift at the surface.

Regards from Spain,

Jon

Hi Miguel and Jon,

just to add to the nice reply fom Jon, that for the temperature decrease the equivalent potential temperature in the 850 hPa level is a bit more robust parameter than the 'regular" temperature of the 850 hPa level.

sincerely yours,

Malte

hi, Miguel,

if you are trying to identify the position of an airmass front based on surface measurements (for example METAR data) the first look is always to the windfield. If there is a significant jump of the wind barbs to the left (several stations should show such a jump) you can be quite sure to locate the frontal zone on the surface between the two adjacent stations were this jump occurs. The second step would be to check dewpoints as a more or less tracer for airmasses but near the surface this might be not as significatn as the wind jump. If you do have data from pressure levels higher than about 850 hPa the horiicontal gradient of the equivalent potential temperature is the best choice as Malte noticed.       

Thanks to all of you!! It would be very interesting to check in a higher level and try to do  some kind of convolution between different fields.  As a starting point I want to use only the wind, but as Malte and a colleage of mine said, the potential temperature may help a lot.

If you are talking about a cold front, especially one that has undergone continuous frontogenesis, then it would be easier to identify than a weak cold front (i.e. a less sharp leading edge).  If the cold front behind the leading edge is sufficiently colder than ahead of the front, then your winds should show an increase in strength due to baroclinicity.  If we are talking about the northern hemisphere, then the front is propagating at some angle from the northwest, with the wind vector, in a well-developed baroclinic situation, nearly parallel to the frontal axis and cold air to the left.  If you have profile information, which it seems from your message that you do, you can look at evidence of cold air advection, which results in a wind backing with height.  It would be good to have other information as others pointed out to understand changes in stability, humidity, and pressure, for they are all important for diagnosing a cold front, especially since the frontogenetical equation evaluates how temperature and wind influence the shape and strength of the front.  I recommend to cross analyze any surface charts with your idea as well, or conveniently pick a day where a front is moving through to see what you can pick out in the wind field.  Good luck. 

E

I see that I'm a late entry in this discussion, but I thought I could "add to the confusion"...:-). I would say in general no; just winds are not sufficient, just as only temperature or humidity are also not enough. This seemingly "simple" problem (everyone recognizes a front on a weather map, right?) is surprisingly difficult, for two reasons: 1) Huge variability between individual fronts, and; 2) false positives.

It is easy to design an algorithm that captures the majority of fronts, but it will likely also "capture" a number of non-fronts, like mesoscale phenomena such as the sea breeze, and often also coast lines. And the surface temperature doesn't always drop with the passing of a cold front or increase with a warm front, because it is determined by a number of other factors, like radiation and clouds.

I would add the pressure field (things like divergence and vorticity) and the gradient in equivalent potential temperature, and stay away from the surface; use  850 hPa that is less influenced by the local surface.

I would like to add that I agree with Michael; a change in wind strength is not a strong enough parameter to recognize an atmospheric front with no other variable available. But there is another parameter which is wind direction. Have you considered that?

Thanks Alastair for pointing this out. You're absolutely right; I was less than specific and changes in wind direction across a front is probably more more important than changes in wind speed. To my defense, some of that should be manifested in the pressure field - and I did mention that...:-).

I would also download this review article on cold fronts and interaction of cold fronts with topography.  The paper provides much of the information that you might need with plenty of good diagrams and resources within.  

I have just realised that change of direction is a characteristic of a cold front passing. The winds veer ahead of cold fronts and back behind the cold front :http://www.theweatherprediction.com/habyhints/48/

HTH,

Alastair

Yes, it will be a useful Tool. There will of course be many subclasses dependent on the strength  and other characteristics of the front..

BR

Søren