Geolocation Exercises

If it's in the UK—which I'm assuming it is—then green signs indicate turns off to A roads for example.
View attachment 77936
Sorry to be pedantic but that's not the case. On a motorway itself the signs are blue, even where they indicate exits onto A roads.

The example you show there is on an A road, the A414 which runs parallel to the M1 at that point. (The Street View car is on the motorway but the sign is on the A road to the left of the motorway, on the far side of the barriers, at the point where a slip road comes off to join the M1.)

https://maps.app.goo.gl/zve1sivDu77jpPU86?g_st=ic
 
I'll try posting an image of the same location from Streetview tomorrow, but I will say that while the comet has been identified, both the date and time are wrong, as is the guess that it's in the UK.
 
Last edited:
Sorry to be pedantic but that's not the case. On a motorway itself the signs are blue, even where they indicate exits onto A roads.

The example you show there is on an A road, the A414 which runs parallel to the M1 at that point. (The Street View car is on the motorway but the sign is on the A road to the left of the motorway, on the far side of the barriers, at the point where a slip road comes off to join the M1.)

https://maps.app.goo.gl/zve1sivDu77jpPU86?g_st=ic
You're being pedantic, but that's expected! What I meant was green signs indicate turns to primary routes and it looks like that image has a motorway primary route going under the bridge.

External Quote:

  • Rectangular road signs: Rectangular signs inform. Blue rectangular signs give information on motorways, green signs direct you on primary roads, while white signs give directions on minor roads. Rectangular signs can also indicate bus lanes and congestion charge zones.
https://www.rac.co.uk/drive/advice/...ndicate bus lanes and congestion charge zones.
 
Because I know what a massive time suck these things are here's one method to solve this one.
@Mendel guessed correctly the "UFO" is the top of anaerobic digester. However, it doesn't appear in both photos which means it's relatively new (the Google maps image is from 2023 and I took the sunset photo in Dec '24—but, you'd have to have guessed that). It also does not appear on either Google Maps or Apple Maps satellite imagery. Because of its newness it's not been tagged in OSM (yet). So can't be used to locate this farm using Overpass Turbo or other tag searching method. So it's a bit of a red herring....

View attachment 77937

However, there are enough old and tagged things in the photo to get to the right answer, if we guess some general things, like that it's somewhere in England and the photos are taken from a road.

Specifically a trunk road: highway=trunk
We can see some general buildings, that look like a farm yard: landuse=farmyard
On the left are some trees: natural=wood
and spanning the field is a wall. Specifically a dry stone wall: barrier=wall AND wall=dry_stone

If you search for these elements within an estimate radius of 150m from each other then you should get around 52 hits, which is still quite a lot, but few enough to be able to review the streetview imagery for in order to match the photos. (the correct hit for me was #44).

From Bing Streetside:
View attachment 77940

*** Bonus points on offer for pinpointing when the digester was installed.***

Whilst Overpass Turbo is great, its effectiveness to find locations is obviously very dependent upon the accuracy & consistency of the OSM Database. @Giddierone 's complex solution highlights this. I struggled in the two recent 'Alan Partidge' cases to find the locations.

In the first case, trying to find this location using parking=layby when the layby wasn't in the database. I also struggled to isolate the dual carriageway like is shown in the image because there is no tag such as highway=dual_carriageway. The OSM forums suggest using lanes=2 , but that returns a bunch of other roads with 2 lanes. I solved this by using maxspeed=GB_nsl_dual to find roads in the UK at have the National Speed Limit for Dual Carriageways.

1741251713861.png


In the recent search for the House all my overpass queries failed because the two power poles were not in the OSM db (power=pole) - i should have used power=minor_line. Neither was the house (building) although the area was indicated by landuse=residential. Also my estimated distances from the camera to the pylons and poles was way off.

(I have since updated the OSM db)
1741252906571.png


Basically, what I'm trying to say is that sometimes we have to try a different attack method when using Overpass turbo due to the potential for inaccuracies or inconsistencies in the OSM db. Although we may be trying to find clear and unique combinations of specific and detailed features , sometime it is better to use features that are bigger and more generic which can then be filtered visually.
 
You're being pedantic, but that's expected! What I meant was green signs indicate turns to primary routes and it looks like that image has a motorway primary route going under the bridge.

External Quote:

  • Rectangular road signs: Rectangular signs inform. Blue rectangular signs give information on motorways, green signs direct you on primary roads, while white signs give directions on minor roads. Rectangular signs can also indicate bus lanes and congestion charge zones.
https://www.rac.co.uk/drive/advice/learning-to-drive/the-highway-code-uk-road-signs-and-meanings/#:~:text=Rectangular road signs: Rectangular signs inform. Blue rectangular signs give information on motorways, green signs direct you on primary roads, while white signs give directions on minor roads. Rectangular signs can also indicate bus lanes and congestion charge zones.
And in this case we are not even in the right country!
 
I've been trying to get Overpass Turbo to work but haven't ever managed to get any results without it timing out, even when searching quite small areas. Is there a problem with it or am I just not using it properly?

(I tried getting ChatGPT to write queries from plain language, but it was hopeless and always produced syntax errors)
 
I've been trying to get Overpass Turbo to work but haven't ever managed to get any results without it timing out, even when searching quite small areas. Is there a problem with it or am I just not using it properly?

(I tried getting ChatGPT to write queries from plain language, but it was hopeless and always produced syntax errors)

Sometimes it times out for me, but it works most of the time. I either use the Wizard or just edit scripts that I know have worked in the past.

Try this script over any major city and see if you can get it to work.

https://overpass-turbo.eu/s/1ZSO
/*
This has been generated by the overpass-turbo wizard.
The original search was:
"amenity=drinking_water"
*/
[out:json][timeout:25];
// gather results
nwr["amenity"="drinking_water"]({{bbox}});
// print results
out geom;

results:
 
1741289380984.png

Here's the day view of the same location, I'll also say that the exact local time was 21:28 pm, which given the sky should help guessing a latitude and date.
 
The comet is in the middle of the photo, there's also two planets visible and it's a Spanish speaking country.
"Spanish speaking country" strongly hints at South America.

Using this as a starting point from the alignment of Venus-Saturn-Formalhaut we can determine a likely date: I would say the 20.01.2025 fits nicely.

I tried testing several locations with Stellarium and given the height above the horizon of the comet/planets and the local time of 21:28 a place somewhere near the western coast seems likely. Probably Chile? Haven't found the precise location yet.

original_overlay.jpeg
stellarium_overlay.jpg
 
Yeah I had similar. Somewhere around S 38 degrees, (Chile?) but on Jan 21 around 22:00.
I just had a second look at Stellarium and compared the alignment of Venus-Saturn-Formalhaut on different dates. The original image (brightness and contrast adjusted) as reference:

original_positions.jpg


The alignments on the different dates at 20:28 o'clock. I used Santiago in Chile as reference point but for the alignment that shouldn't make a noticeable difference.
date_positions.jpg


I think only the 20th and the 21th would be valid dates. I feel like the 20th fits a little bit better.
 
Another possible hint to narrow down the search location.

In the original image there seem to be three planes visible:
planes.jpeg

All three planes seem to go or come from roughly a SW direction. A quick check with ADSB-Exchange shows some flights that are vaguely aligned in groups of three. First for 2025.01.20 21:29:
planes_adsb.jpeg


One day later adds two further groups:
planes_adsb2.jpeg

Groups 2 and 5 seem to be most likely since they are flying on the same path pointing somewhat SW. This is also true for group 6 and 7 however those planes are quite far apart. I am not sure if they would show up on the original image with such small spacing in between.

At the very least it should rule out a large chunk of South America as possible search locations. ;-)
 
Chile, the Milton Keynes of the South Americas (this isn't the location but seems close...)

-36.65448914402652, -72.22791959537233
Screenshot 2025-03-11 at 11.53.13.png
 
flying on the same path pointing somewhat SW
There are several flights all flying above 20,000ft on the flightpath over Mendoza on the 20th. Although I haven't find the road junction [motorway_link] where the photo was taken...yet.
I added some guides for the planes (green) and "W" for West to the image with an overlay from Stellarium.
1741123050956 copy.jpg
Screenshot 2025-03-11 at 16.14.28.png
 
There are several flights all flying above 20,000ft on the flightpath over Mendoza on the 20th. Although I haven't find the road junction [motorway_link] where the photo was taken...yet.
I added some guides for the planes (green) and "W" for West to the image with an overlay from Stellarium.
The screenshot from ADSB-Exchange says "2025.01.20 20:16 UTC". Isn't this several hours to early? Chile and Argentina are luckily both UTC -3 (https://www.timeanddate.com/time/map/) so ADSB-Exchange should imho be set to "2025.01.21 00:28 UTC" to match the given local time of "2025.01.20(?) 21:28".
 
venus and saturn dont line up on any dates further north..like Columbia or even Mexico? i cant use that software.
Not that also line up with the comet on the horizon just after sunset. You have to be further south, otherwise Venus appears higher than Saturn.
 
venus and saturn dont line up on any dates further north..like Columbia or even Mexico? i cant use that software.
Not that also line up with the comet on the horizon just after sunset. You have to be further south, otherwise Venus appears higher than Saturn.
Yeah it is a combination of both. I think I should share my thoughts here just because I am not so sure myself that I haven't made some mistake somewhere down the line. Feel free to correct.

The lineup of Venus-Saturn-Formalhaut itself is pretty independent of the location on earth and mainly influenced by the date.

However the further north you go the lower the comet will be over the horizon and at some point it will no longer be visible. Additionally we have to consider the given local time of 21:28 that can translate to different UTC times depending on which country we use as a reference point which will also influence the height of the comet above the horizon. Peru for example has a UTC -5 compared to UTC -3 of Chile so if you stand on the border of Chile/Peru at 21:28 local time Peru the comet will already be far below the horizon while at 21:28 local time Chile the comet is still visible. Same for Bolivia with a time zone of UTC -4. That pretty much rules out Peru and Bolivia and all other countries further north.

Also the longitude cannot be ignored since the triangle Venus-Formaulhaut-Alnair will be tilted to the left or the right depending on your location. After trying out several locations in Stellarium (the shared timezone between Chile and Argentina helps a lot here) I think that somewhere around Chile or the western parts of Argentina would probably fit best by comparing the constellations with the original image.

On top of that we have some additional hints like the hills to the west. These rule out an incredible large and completely flat part of Argentina. Also the northern part of Chile is mostly to arid to match the vegetation in the image. Sadly the remaining area is still the size of a medium sized country. And it certainly feels like Chile has no shortage of road crossings with bridges involved. Argentina on the other hand is the total opposite: I hardly found any bridges there and there are not that many planes around in the western part.
 
Not sure how helpful this is but to my eye the distant mountains in #168 appear similar to those at this motorway junction #176 measuring them on a map they are around 20km. So perhaps we're looking for a similar motorway junction with mountains ~20km to the WNW?
Screenshot 2025-03-12 at 16.30.09.png
 
The country is indeed Chile or Argentina, and the date is the 20th or 21st.

For now, I'll say that given the sizes of the two potential countries,
-36.65448914402652, -72.22791959537233
The location is at least 3° further North than these coordinates, which should help rule out like half of both countries

Also, as a sidenote
like Columbia
It's Colombia, not Columbia
 
found it. i was gonna let the other guys search cordoba because i always get lost trying to do google maps. but i actually found it. yay
Found! Were you already suspecting somewhere in Argentina or did the latitude restriction give it away?
 
Last edited:
These three aircraft seem to fit. What was the actual time on the image @Calter ?
Screenshot 2025-03-12 at 21.44.29.png

Also, for those hoping to use Overpass to identify this location (and who know how to use it properly) I wonder what the most elegant solution is?
Looking at it on Open Street Map there don't appear to be many tagged items. If you wanted to triangulate this location by searching bridge, tertiary road, highway for example I think you'd get a lot of useless hits. There are no tags for things like the grassy area, the motorway sign or any description of the bridge, (concrete instead of iron suspension for example).
It seems Overpass is very dependent on the fastidiousness (nerdiness) of the those contributing to it. For example if you look at any historical European town you can find tags indicating what kind of brick work a building has, what kind of tiling pattern a path has or what kind of seat a particular park bench has.
Screenshot 2025-03-12 at 22.00.36.png


https://www.openstreetmap.org/query....282610#map=18/-31.432190/-64.282065&layers=N
 
Last edited:
What was the actual time on the image
The metadata of the image says it was January 20th at 21:28:48

I'm honestly not sure if those three lines were planes or just some sort of artifact, I had seen some planes earlier that night and I'm surprised I missed three of them going in a line (though I was mostly focused on the comet once I managed to spot it)
 
These three aircraft seem to fit. What was the actual time on the image @Calter ?
View attachment 78169
Also, for those hoping to use Overpass to identify this location (and who know how to use it properly) I wonder what the most elegant solution is?
Looking at it on Open Street Map there don't appear to be many tagged items. If you wanted to triangulate this location by searching bridge, tertiary road, highway for example I think you'd get a lot of useless hits. There are no tags for things like the grassy area, the motorway sign or any description of the bridge, (concrete instead of iron suspension for example).
It seems Overpass is very dependent on the fastidiousness (nerdiness) of the those contributing to it. For example if you look at any historical European town you can find tags indicating what kind of brick work a building has, what kind of tiling pattern a path has or what kind of seat a particular park bench has. View attachment 78170

https://www.openstreetmap.org/query....282610#map=18/-31.432190/-64.282065&layers=N
its wild how your [see below] link above is the exact same shape..isnt it? (granted i was originally on the wrong side of the bridge by following your shapes..but since the sign matched i drove under the bridge to check. ). serendipity.

anyway you need a 4+ [double] lane highway (width of bridge) overpassed by a 2-4 lane route, with an off ramp (the off ramps are the petals). i put the word "overpass" (because where i live we call that an overpass) in overpass but didnt get anything
1741821471307.png
 
Last edited:
I'll throw some flavor to this thread by adding a dashcam video.

I recently helped someone figure out some potential drone sightings and I figured their footage would make for a simple exercise. I'm curious how much can be narrowed down from this footage alone, I'll narrow down near which town/city it happened in three days, which should basically give it away.
 
I'll throw some flavor to this thread by adding a dashcam video.
View attachment 78341
I recently helped someone figure out some potential drone sightings and I figured their footage would make for a simple exercise. I'm curious how much can be narrowed down from this footage alone, I'll narrow down near which town/city it happened in three days, which should basically give it away.

I still haven't worked out who he's overtaking for those 40+ seconds.
 
I see these signs in this order
MUTCD_W11-3.svg.png

W11-3 Deer
MUTCD_R2-2.svg.png

R2-2P Speed Limit
MUTCD_W3-3.svg.png

W2-3 Traffic Light Ahead
MUTCD_OM3-L.svg.png

OM2-L Left object Marker

MUTCD_R4-7c.svg.png


R4-7C Keep right of obstacle

Street lamps look like Connecticut or even North Carolina.
 
Last edited:
Back
Top