Addition of Fronts
It wouold be incredibly useful to add the layer of "fronts" into the layers palet.
Fronts are NOT drawn by numerical weather forecast models, but by skill weather forecast meteorologists.
As Windy shows maps directly issued from weather forecast models, there is no way to show fronts automatically.
Moreover, what does the drawing of the fronts bring? The position of the air masses? The changes of direction, and the eventual strengthening, of the wind? The type of weather, cloud systems, precipitation? All of this you find it in detail on Windy. The fronts were imagined in the 1930s, at the time when this synthetic and graphic presentation allowed a simplified representation of current and forecasted weather. Today's weather models provide much more detailed forecasts, even if they are not able to offer plotting of fronts.
Personally, I use to look at Temperature layer at 850hPa which gives a rough idea of the location of the different fronts...
planetjim last edited by
To add to @idefix37 fronts are sometimes unmistakeable phenomena, but sometimes a bit vague and sometimes very vague. There may be a single front or there may be several lesser fronts succession, or just a weather change over a period of an hour or two to cooler temperature and a different wind vector.
There'a judgement and even a psychological component to what constitutes a front. A human forecaster will take the human impact into account in choosing whether to draw a front, how far it extends horizontally and when it has degraded sufficiently to drop it. It's a tough job for a NWP model - more of a neural problem - but expect it one day.
It's probably better to look at rain, temperature and wind change then decide if that's what you call a front.
I agree with idefix: temperature layer at 850hPa is useful. Looking at winds also helps: wind abruptly changes direction by maybe 45 degrees at the front. Over ocean, surface winds are fine, but over land I'd look at winds at 850-900hPA as well. It also helps to look at precipitation: it rains there. It also helps to see the pressure contours: they have a little kink at the frontal lines. My favorite view at windy.com for that is: color as precipitation, pressure contours, and particle wind animation.
There are some related discussion here https://www.morganscloud.com/2017/06/28/grib-weather-files-you-gotta-see-the-rain/
@idefix37 - awesome, thank you so much. I'm much smarter now reading your answer. I will watch 850hPa. 👍🏼
Jacquiel last edited by
In aviation, before flight, the pilots use fronts and tipically stationnery fronts to avoid storms.
In fact, fronts are very inportants for aviation security.
An other way to improve security is to produce storms animations using lightnings data: the pilot can forecast where storms move and which altitude is the best for the trip he wants to achieve.
Although I don’t disagree with you, what would help immensely for pilots is for pressure change to be displayed with altitude change, using the same altitude slider tool and increments as the temperature and wind altitudes use. The pressure data with altitude change clearly exists in the model Windy has access too, as other features in windy use it already. So add an altitude slider to the pressure overlay interface. That would allow pilots to see the upper lifting structures, and their relative movement with time, more clearly than a moving line over a model ever would. The data and tools needed to implement it already exist.
The implementation issue I foresee is the need to extend the pressure overlay's range to cover the much lower pressure (150 hPa at FL450 for instance). I extended the pressure overlay down to 870 hPa. It would be easy to extend it to 700 hPa. Or even down to 500 hPa to cover the entire VFR range. It’s just a matter of tweaking the increments and color shading (I increased the temperature overlay's lowest T range already to deal with the entire globe’s temperature distribution above FL340). And IFR pilots will generally have flight service and institutional tools or avionics integration available to address WX issues at IFR altitudes. So there’s no particular need to display below about 500hPA (+/-100 hPa).
Regarding your link that pilot's advice is quite correct. However, an increase in short-term pressure situational awareness for a GA pilot just prior to taking a flight is a very good thing to have easy access to and Windy does have a potential to provide easily accessible rapid-assimilation of pressure changes they're likely to encounter at VFR altitudes. If the overlay emphasizes fine-details throughout its range they would easily see where the 'front' or trough is located, and moving to. No one would actually need a line drawn to see where it is and where it fades to ambient, as a detailed display would implicitly mark where such features are forecast to be with time.
PIREPS can 'ground-truth' the forecasting, but a general situational awareness boost reduces practical mental workload and stress which provides the mental-space to perform all other flight tasks better. Pressure changes can surprise you, and they want a tool that leaves less likelihood of surprises. Yes, a pressure model can be wrong, but a line on an air services forecast plot can be too, hence PIREPS.
Windy can easily provide that awareness without manually drawing a line anywhere as the model’s ‘pseudo-data’ will depict which way the pressure change structure lays and is likely to move.
That would be extremely useful to many non pilots as well.
Windy will not draw fronts in the maps because this is a manual process which a meteorologist have to do.
Another site has this. It is potentially in digital form (after someone has drawn it) and can thus be obtained. Perhaps some data trading is in order?
Yes by having a visualisation of temperature gradients mixed with moisture at each altitude. I don't think such visualisation is impossible to achieve/automate
@idefix37 I too would like to see fronts - thus to be all to draw conclusions relate to the 'types' of clouds you see. And to draw parallels with traditional synoptic chart trading. eg. C.Nimbus == cold front. And to be able to access when it's arriving. If fronts are truly only generated by meteorologist's interpretations, then there is an algorithm or set of rules that they use. And so, these could also be translated into a graphical layer....
Just because people in the past have not done it this way, it doesn't mean that the future techniques could be even better...
So to correct your statement, it should have read, "Fronts have traditionally not been drawn by weather forecast models... but they certainly could be, once we were to create and implement the appropriate algorithms..."
@cyder Exactly. A very good idea! It would be a superb addition.
As a retired pilot, I recently discovered windy.com and the pilot weather app. I've always been a weather nut and run multiple weather widgets on my phone. The shortcoming of all of them is the lack of METAR and TAF info, which the Pilot app provides.
I discovered this conversation when I couldn't find any frontal depictions on the windy.com site. And now I know why. So no worries there. Thanks for a great site!
yavordjonev | Premium last edited by
I would like to second this request from a sailor's prespective.
It seems that ECMF produces weather front data already: https://apps.ecmwf.int/r/cdb/
It would very useful to see it in Windy.
jidanni last edited by
As much as I don't like talking to robots at customer service,
I must accept it anyway these days.
So I would be, yes, happy with artificial fronts drawn by robots (algorithms)
instead of real weatherpeople.
vsinceac | Premium last edited by vsinceac
Afaik. that ECMWF front forecast are for Ensemble model; I wonder how could sailors cope with such a spagetti plot:
Bruzote | Premium last edited by Bruzote
@idefix37 - (For those reading this old topic. :) Fronts on regular NWS surface analysis maps are MOST CERTAINLY done by computers, not meteorologists. Likewise, Accuweather stopped using manual analyses. The change in the NWS policy was done many years ago. I remember, as I was an active forecaster at the time. The quality of the surface analysis maps plummeted. Symbols for High and Low were computer-generated as well. The symbols were put all over the map and sometimes right on top of each other - really useless. The fronts were at times helpful, but sometimes poorly placed (I think nonsensical errors are quite rare now). The automated maps are not as helpful as hand-drawn fronts, especially when fronts had kinks or discontinuities. If you look at the NWS maps coming out of the WPC, it is apparent that someone at the WPC has an approval responsibility and their name goes in the legend. However, if you look at the product, it is CLEARLY automated. Perhaps the NWS allows for map tweaking, but certainly the primary products are automated. Eventually, AI will do this better than humans, but for now I don't know if that has been achieved yet. Work on this requires money and fronts are more for layman consumption or forecast conceptualization rather than forecast accuracy.