I take photos with my phone so I use the location finder on the iNat website. I mostly take photos in my yard or at a well-traveled park. Is there any need to give a location beyond 6 decimal places? Since the map (for me) is just an estimate, more than 6 decimal places seems strange to me.
Six decimal places after the point are equivalent to 1/9 m (111.111ā¦ mm) in latitude, or in longitude on the equator (multiply by the cosine of the latitude). Since you canāt enter an accuracy circle smaller than 1 m, you donāt need to enter more than six decimal places. The site displays it rounded to five.
10 Mm = 90Ā°
1 Mm = 9Ā°
1Ā° = 111.111ā¦ km
3 Likes
Obligatory post for this paper related to spatial data/scales and organisms (opens a PDF):
11 Likes
Love this article!
āI suggest that if you only know the village where you have seen an animal, 2 decimals is probably the best you can hope for. If you are at a plot, you are usually safe with 3 decimals. If you run after the creature, you are ok with 4 decimal places. If you are at a cave entrance or water hole, 5 may just be ok (if you trust your GPS). Six decimal places tell you which side of the snake it is you recorded.ā
Itās clear that 6 decimal places may actually be too many for me! I was intimidated by reports of new species where other people go to the location to confirm the report, although, since I am a newbie and mostly make observations in my (modest sized) yard, I donāt really expect to be observing new or rare species. Thank you so much for laying my concerns to rest!
6 Likes
22 Likes
As a general rule, if Iām hand editing coordinates for plant observations, I round to 4 decimals, or 3 if Iām unsure of the precise location. Having tested my phone GPS repeatedly on known points (USGS benchmarks), it has a repeatability within ca. 3 to 5 m in open areas, and 5-10 m in forests, so this is about the precision of 0.0001 degree.
3 Likes
My GPS gets within a few millimeters if it has a clear view of the sky. If I cook a point for three minutes, which is normal for traverse points, and the standard deviation is 135 mm, the accuracy is 10 mm (135/ā180). If it says DGPS or float instead of fixed, the circle is at least 2 m in radius, but I still enter six decimal places after converting the state plane coordinates to lat/long.
1 Like
Millimeters suggests that you are using some pretty high-end survey-grade engineering equipment. Are you sure you didnāt mean meters?
I am able to get down to about 4-5 meters accuracy with my phone, and with my recreational-grade GPS unit, so I donāt record values to more than 5 decimal places (roughly 1 meter precision).
1 Like
I meant millimeters. The receiver is by Champion and screws on top of a two-meter pole, and the data collector uses a device called Mi-Fi or Jetpack to get corrections to the position. The pole has a level bubble on its middle and a point at the bottom, which I put in the dimple in the top of the nail.
1 Like
Presumably this device you speak of uses GPS and GNSS in combination (since either alone is limited to accuracy of around 3 metres under optimum conditions)? Even the two in combination hits a theoretical limit of around 1-metre accuracy, so I am intrigued how they are getting millimetre accuracy!
There are reference stations which run GPS receivers at known locations and publish on some Internet site the difference between the known location and what the GPS receiver says. The data collector adds this difference to what my GPS receiver says, thus getting a more accurate measurement of location.
1 Like
If Iām manually adjusting coordinates, I also round. It would be neat if coordinates were automatically rounded depending on the precision value. It makes no sense at all to have coordinates with ten or more decimal places and a precision of 100 m.
2 Likes
Another thing to remember is that when you put gps readings onto google maps it will rarely be in the correct location, but it will be close depending on where you are located. Google maps uses an Auxiliary Sphere. (basically assuming that the earth is a sphere). I usually add 1 meter to the accuracy that my GPS estimates to account for this.
This conversation helps with something that has bothered me. In school, when we learned to calculate the area of a circle, we were taught to use a the wavy-line equal sign to show that it was only an approximate area. But a physical object in the shape of a circle has an exact area, so it always bothered me that we were, in effect, saying that we canāt know the size of the object.
The area of a circle with an exactly known radius is known exactly. But if the radius is rational, the area is not, so itās written as an approximation.
This xkcd reminds me of a question Iāve always had. If you get your coordinates from a GPS device, the geodetic datum is WGS 84. But if you get your coordinates by picking a spot on Google Maps, what is your geodetic datum then? Is it also WGS 84?
1Ė= about 100km
0.1 = about 10km
0.01 = about 1km
0.001= about 100m
0.0001 =about 10m
0.00001 = about 1m
4 or 5 decimal places (10m or 1m precision) seems about as much as is useful or realistic for most observations
4 Likes
From speaking to surveyors here, itās entirely possible to get mm accuracy, but theyāre also using equipment that costs enormous amounts (no doubt in the tens of thousands of dollars).
3 Likes
Yes it is WGS84. The difference with GPS is that instead of a Geoid, Google uses an āAuxiliary Sphereā, which assumes the earth is a sphere. Google can be as much as 20 meters off, depending on where you are. Also consider the fact that when accuracy of a location is calculated it is not likely to be accurate 100% of the time.
3 Likes
Unless youāre in China of course where Google Maps is typically 100ā700 meters in error due to legal restrictions on geographic data in China.