I don’t do identifications if there hasn’t been sufficient care in the location. The location of the observation is as important as the species name. The words used to describe the location need to corroborate the position of the pin. So if they don’t match, or the place is so vague as to be meaningless (e.g. Lincolnshire or Wales), or the location is given as a postcode which increasingly seems to be the case, I don’t spend time trying to do the identification.
Sometimes, especially if it is someone new to iNaturalist, I explain why the location is important, but there is very seldom a positive response and some people are just plain rude, so mainly I just skip over observations where the location is unsatisfactory. The commonest excuse is “what I have done is within the iNaturalist rules”, but we shouldn’t be aiming to do things as badly as the rules allow. We should be encouraging people to raise their game so that the observations are as useful as possible.
I know the times I have trouble with phone camera gps accuracy is when service is bad or entirely unavailable. that’s just from personal experience, so you may not be having the same issue.
Most people seem to be getting that interpretation from my post, so I think I may have phrased it badly.
My question was not about whether the location is important or if it should be correct (as I know it is and I agree with what you’ve said, it is just as important as the photo or the ID itself).
It was more related to:
whether I should change the location from the place I was in when I observed a given organism to the place the organism really was (e.g. ducks in the middle of a lake)
whether it is important to associate an accuracy value to the observation (so that it has an accuracy circle) or if it is okay to leave it as a point with no accuracy circle
Adding to the faulty location (or even missing), here in Portugal it is rather common for new users to not add an ID and, sometimes, a date (this last one seems to be the most problematic to solve)
I would say that changing the location to where the birds really are is best, but not essential. If you could see them from the location posted, that’s good. It is good to make the accuracy circle (what we used to call the fuzz factor) big enough to include the birds.
Including an accuracy measurement is valuable, worth going to some trouble to do. However, some records lack that and for some purposes they’ll still be useful.
Soon I’ll be uploading photos from the '80’s for which I didn’t record locations very accurately and my accuracy circles will sometimes be over 30 miles in diameter. Best I can do and, considering the size of the earth vs the size of my circle, not too bad.
If someone posts an observation for the purposes of learning about the organism in the image, with only limited information on location, that is not only within iNat’s rules but all the “game” anybody needs to bring to participating on iNat. If you don’t want to help them, that’s your call, but it really has nothing to do with iNat’s mission, which is promoting understanding of and learning about biodiversity.
Yes, point always should be where organism was (as precise as you can guess, it can be hard).
It is both important to use as data, but also you always need to check how well it was recorded, as you adjust the pin it always gets a circle, so it’s inevitable if you take care of that.
I usually check if the GPS recorded my location well enough before uploading photos to iNaturalist. That is why I asked whether I should assign an accuracy value to each observation, given that that proccess can be quite time consuming when uploading hundreds of observations at a time.
But what you say makes perfect sense, I’ll adjust the pin from now on. Thank you very much!
Most GPS apps/devices allow you to export the track in a standard file format such as GPX (GPS Exchange Format). There are many mapping apps available that can display such files, and also allow you to edit individual track points. When viewed in this way, it is quite easy to spot any wildly inaccurate points and tidy up the track before linking it to your photos. You can also adjust short sections of the track to better reflect the true location of the subjects. When overlaid on a satellite view, it’s easy to align the track with roads, footpaths, natural features, etc.
Doing things this way can be a lot more efficient than editing the observations individually. With properly revised GPS tracks, it becomes reasonable to assume the same accuracy for all your observations (based on some conservative average accuracy for your GPS device). This can save a lot of time if you have thousands of uploads to do.
" no value of precision" does not equal " location is quite precise"
If there is no value, then there cannot be any precision inferred.
A GPS enabled smart phone my be accurate to about a 5 meter radius. From my experience cell phones are not always consistent.
GPS signal is broad cast with an accuracy of about a2.0 meter radius, however signal blockage such as trees & buildings, satellite geometry, atmospheric conditions and the receiver can all affect the user’s accuracy. When you transfer the GPS reading to Google earth or Google map, you introduce another small amount of error which varies with your location because Google basically assumes that the earth is a sphere instead of an ellipsoid.
Using a dual frequency receiver or an augmentation system such as WAAS you may boost your accuracy to 2 or 3 meters under a good sky (from my experience), beyond that you would need very expensive equipment. https://www.gps.gov/systems/gps/performance/accuracy/
“whether or not the issue of missing observation precision is something that a user should worry about”
I think it is better to have more information, if available. When I see a plant on iNaturalist in my area that I would like to see in person or one of my observations I wish to revisit I look for an accuracy of 10 meters or less. If there is no accuracy, I usually wouldn’t bother looking for it. It’s all a matter of judgement. An observation with no accuracy is better than no observation with a high degree of accuracy.
Have fun
It seems to happen when I leave the roads and go on trails, or basically any time I am outside a settled area. I look at my timeline and it has a record of me walking on water, covering five miles as the crow flies in fifteen minutes, and other impossible feats.
But I have also found that I can eliminate this problem if I keep Maps open the entire time. Not so great for the phone battery, but a great improvement for GPS accuracy.
Trees and rocks also are big factors in distortion of signal, it’s a good tip to keep it on always so it doesn’t loose signal and doesn’t start seeking for it again, resulting in wrong location.
The GPS app that I use seems to be disrupted by eBird…
When I go out to observe, I usually have the GPS app and eBird open (this last one so that I can make a list of the birds I observe). Sometimes the GPS app will stop working, and I believe it is due to the conflict with the eBird GPS recording. So I’m surprised that having Maps open actually increases GPS performance.
A question. If mobile phone reception is poor or non-existent, would the phone GPS still work? I’m going to assume that a handheld GPS uses direct satellite communication so this would be less of a problem in a remote location. Since I know little about either of them, the question just popped into my head.
Early on with iNat, I would post a record and simply use the name of the location to generate the coordinates. So some of my early records – for example, from a wildlife refuge I often visited – include a large circle around the whole refuge. My location was accurate but not very precise.