interpolation issue with ASE

[Russ Hodgson 0]

I am a new user to ASE, I thought I understood how the interpolation works (ish) but obviously I don't. My virtual aircraft is at the moment sitting at TKPK Robert L Bradshaw international in St Kitts in the Carribean the current METAR is, 190854Z 18000KT 10SM SCT092 26/22 Q1013 RMK ADVANCED INTERPOLATION. Now firstly this is a full metar report so why does it need interpolation? and secondly and the most odd thing when I click the interpolation details the reporting stations it is using is, NE Dist 1603 CWSA SE Dist 53 TAPA SW Dist 390 SVMG NW 51 TNCM. I can understand it using data from TAPA and TNCM but CWSA!!!!!! is 1603 miles away and SVMG is 390 miles away surely using data from these stations that are so far away can't be right can it? I might not be Michael Fish but I'm probably sure the weather on Sable Island is not quite the same as it is on St Kitts. If it is a case that ASE when using interpolation data looks for the nearest station in a straight line then surely there should be a limit to how far away it looks for the information. So is this a bug with ASe or have I misunderstood how the interpolation works?

[Chrix 0]

Hi, as far as I can see I'm quite sure you fetched up how interpolation in ASE generally works . I conclude with you that it would be more "useful" if ASE skipped some "non-valid" data used for interpolation, but only under certain conditions to avoid possible "weather gaps"... for example you could say a station should be skipped if it's outdated and/or out-of-range (distance), but there need to be at least two (or better three) valid reports for proper interpolation ... just my thoughts ... Regards, Christoph Reule

[aua668 5]

but there need to be at least two (or better three) valid reports for proper interpolation

But sometimes one single station in vincinity (e.g. 20 nm) is MUCH better than 2 or 3 stations for interpolation being many miles away RgdsReinhard

[Damian Clark 1,257]

Hello, The interpolation is range-weighted, so a station over 1000 miles away while another within 50 is not going to influence things much. Because of the design, a station in each "quadrant" must be used, and in some areas you're not going to have one in that quadrant except many miles away. Consider the case where every station is over 1000 miles away (i.e. in the middle of the ocean), we need to interpolate from somewhere. There thus cannot be a limit on range. Also note that any METAR string with "RMK ADVANCED INTERPOLATION" was created by AS, this is not a real METAR. In your case about 20nm station existing while others are hundreds or thousands of miles away, the major influence for conditions should be from that 20nm station.Note that in some areas, we've significantly improved surface temperature and wind interpolation via the interim update beta build (651) available in a pinned topic on this forum. Please take a look.