Surface winds have always seemed (to me) to be wrong in MSFS, and despite multiple attempts to fix some basic errors, it still (to me) appears to be wrong.
So here are a few things to consider, from the Trivial to the extreme.
1
Wind speed, set to 39.3 but processed as 40
Should be set to NEAREST Integer, in this case 39
2
If the Gust is ZERO, from a different direction to the main wind, that does NOT make the winds VARIABLE !!
Winds are set to 40 (40 knots), but in fact they are not 40, they are more like 23 knots.
Its as if the wind speed is really being displayed in ft/sec !!!
This thing about winds being set @ 1000ft and reducing near to the ground has really messed so much up.
The wind for the airport is measured at near Ground level, NOT @ 1000ft, and should reflect that.
Then, if it is felt that the winds at 1000ft should be higher, make them so, but do not mess with the measured / set surface winds.
They’ve taken the METAR winds and decided that’s what they are at 1000’. At sea level they’ve just simply halved it (I think). This is wrong as we all know.
We need the surface winds reflecting what the METAR says, and then blending with the area winds as height is gained.
In live weather i think that the surface winds are accurate to metar. The gusts are not injected in live-weather though. In custom weather they are injected as 50% of the value we set on ground but there the gusts work as they should now. Making it hard to make it accurate to how it realy is because the wind not dropping 50% on ground maybe it should be more gusts near ground though.
I am suggesting that FIRST they get the non-live weather winds sorted out, BEFORE attempting to add the LIVE weather onto that base model of winds.
Getting the non-live weather sorted first, should also improve/fix the relationship between the AWOS / Metars and the actual surface winds.
THEN, start adding in the Live weather, and winds at whatever altitudes that live weather has data for.
While Metar data can be old, the AWOS/ASOS/Tower reported weather, "should have winds reported from the airport at ground levels, as well as the selection of ATC’s preferable runway (modified by local operating criteria)
But they cannot get access to that data digitally everywhere in the world. Often, the METAR and TAF at airports are the only weather data available. Even in the real world where I fly, asking the tower for the accurate wind often just results in the METAR readback - they don’t have access to, or share, the actual wind (or cloud etc), and certainly not digitally.
I think he/she means that if ATIS reports the winds that actually happens in the sim in real time correct and set the runways properly for all of us at the same time we do not need to check METAR. We only need to check ATIS for an update of the winds. Correct me if i’m wrong here? I think it would be cool if we could check some ingame generated METARS based on the weather occuring in the sim instead of real METARS. That should make it feels like the weather is actually simulated in the sim even if the weather is predicted from real weather forecasts and the weather will always match the METARS that are generated.
I hope that is what is planned for the big weather overhaul around next year. It would be amazing. We do not need to leave the sim, makes it possible to plan flights based on the weather that actually going to be in the sim when we fly if those METARS can be exported to flightplanning softwares it would be more awesome. Then they have an own weather system in the sim based of realworld forecasts at meteoblue. That i think would be the real future of this flight sim weather system.
I noticed that the reduction of wind only happens for “big” Airports which in fact have Metar. If i land on a small gras Airstrip, the Winds stay the same until the ground. At least with live weather
Another possible explanation as to why surface winds would be lower than METAR would be if the sim were (for some reason), injecting the wind in meters-per-second instead of knots. As it happens, for a given wind speed, meters-per-second is almost exactly half of that speed expressed in knots.
Winds aloft (using Live Weather) are definitely injected in knots. Comparing the winds aloft at a particular location and altitude in MSFS to the NOAA forecast winds aloft from www.aviationweather.gov typically shows a close correspondence.
Computer forecast models such as the one produced by MeteoBlue will normally internally store wind values in rectangular coordinates “U” and “V”, with the “U” value expressing the east/west component and the “V” the north/south component. Speeds are stored in M/Sec. These are converted to directional coordinates and magnitude by vector algebra, and the speed can be converted from M/Sec to knots by a simple formula.
Because injected Live Weather winds at altitude appear quite accurate, I am confident that the conversion is being done correctly when the LiveWeather model is the only source.
I have not tried spawning at an airport that does not report METARS, but will give that a try, because in this case, the surface wind would come only from the MeteoBlue model.
Of course, without METAR, it would be difficult to know what the actual wind value should be for the non-reporting airport. I will give this a try by spawning at a METAR-reporting airport, then moving to a non-reporting airport that is relatively close by - keeping in mind that the LiveWeather model’s surface wind is only a prediction - not “absolute truth”.
METAR winds are always reported in knots, with a few exceptions (such as Russia where they are actually reported in M/Sec). I can’t imagine why METAR winds would be converted to M/Sec (with knots as the original value), but that would account for the injected surface winds from METAR being just about one-half of the reported value.
What I know for a fact at the moment is that if you set the Sim weather to have a 40 wind speed, then look at the winds that the plane is experiencing on the ground, the winds are very close to 40 ft/sec, (28.6 in this case) (or about 50% of the value entered with the preset slider)
If, in fact the sim windspeed on the Ground is about 50% of what one is told it is, then, if nothing else, this promotes a false confidence, that one can land in excessively high winds !!
I spawned at CYXU - London, ONT (METAR reporting). METAR gave winds 340 at 14-16 - in the sim it was 340 at 10. I moved to CLW4 (a non-reporting airport about 7 miles from CYXU), and wind was 345 at 16.
Then I went to KSUX (Sioux City, SD). METAR gave 143 at 16-23. In the sim, it was 145 at 16 to occasionally 17 - reasonably close to the METAR with less gusts
Moved to 7K8, a small strip about 5 miles away (non-reporting). Wind was 138 at 18
It appears that although METAR injected wind might be less than the actual METAR, it is definitely not “half” the correct value. At Sioux City, it was actually the same as the lower (non-gust) value from the METAR.
Finally, I went to KLBB (Lubbock, TX). METAR wind and in-sim wind were both 180 at 16.
One thing that is somewhat concerning: METARS report wind direction referenced to true north, while ATIS, AWOS and ATC tower controllers give the wind direction referenced to magnetic north. Based on what I saw at CYXU, KSUX and KLBB, It appears the sim is injecting the wind direction directly from the METAR, without applying local variation.
I am using MSFS live weather, which uses current METAR for airport surface wind, temperature and pressure - otherwise (at a non-METAR-reporting airport), it uses the values from the MeteoBlue model directly (I assume).
The source of current METARS I am comparing against is the Foreflight aviation app.
I am monitoring the weather being injected into the sim at the aircraft’s current location using the SDK “simvar.exe” utility, specifically looking at:
AMBIENT PRESSURE, SEA LEVEL PRESSURE, PRESSURE ALTITUDE, INDICATED ALTITUDE, AMBIENT TEMPERATURE, TOTAL AIR TEMPERATURE, AMBIENT WIND DIRECTION and AMBIENT WIND VELOCITY.
I do not trust the in-sim ATIS to give correct data and although the Cessna with the NXI will show current temperature and pressure (after syncing the altimeter with the “B” key), it (correctly) does not show wind direction or velocity while the aircraft is on the ground.
The above simvars are an absolutely accurate depiction of the weather being injected at the aircraft’s current location, no matter what the source of that weather might be.
At the three METAR-reporting airports I mentioned, the injected temperature and pressure were identical to the most current r/w METARS. I have done this often enough to be quite confident that the existing METAR import function in the sim does obtain the most current r/w observations quite quickly after they are transmitted.
I always have the simvar utility running on a separate monitor whenever I fly in MSFS. I depend on that to get the current airport wind for departure. I never use ATIS. That utility will not help for arrivals though, as it gives the weather at the aircraft’s position (when in flight) not the weather on the ground. I am confident enough in the currency of injected METARS in MSFS, to assume that whatever wind is shown in the r/w METAR will be the wind that Live Weather will give on landing, and that is almost always indeed the case.
I mostly also get the correct METAR winds injected. I think the change near ground in live-weather is when the winds goes from meteoblue wind to the METAR reported winds. That will be better in the next update when they will blend those better. I use shift+z addon to see the windrate all the time on the screen.
And if the gust is from another direction (way off the Wind direction) and even the Gust is ZERO Knots, the MSFS system says the winds are Variable …
It has a long way to go to get everything working correctly – but time is a great healer !!
I wouldn’t be surprised if the same would hold for the windspeed, you probably want to be on the safe side with that (i.e. don’t underestimate the wind).
