New Cache Coordinates

[+AllenFamilyCachers]

I have a Garmin 60CS and when marking a waypoint for a new cache, I have been using the AVERAGE button to supposedly get more accurate coordinates. However, now that I have placed 2 caches, I am getting people saying that the coordinates are a litle off. I personally hate when this happens, so I would like to correct it on my own caches.

 

What is the best way with my GPSr to get the most accurate coordinates for my caches? Is averaging the best way or just place the bugger on the ground, let it settle and hit MARK?

 

HELP!

[+Miragee]

How far off are they saying the coords are? When you take the reading, do you check to see what the accuracy is at the time? If it is 30 feet or so, which I encountered for one of my caches in a canyon, I posted that on the cache page, asked finders to update the coords if necessary, and included a detailed hint.

 

I often take one waypoint, walk away in one direction and come back to the cache, let the GPSr settle, and take another waypoint. Sometimes I do this four or five times, but one time, when the weather was turning, I only took one waypoint. My caching partner put that in her GPSr (a 60CS) and walked right to the cache.

[+IBcrashen]

Averaging is the best way. How long to did you average for? I usually average for at least 5 minutes or more and if the accuracy is up there I will do it longer. There are areas that are just bad reception areas and it sounds like this might be one of them after reading the logs. And remember on the best of days your looking at being off 15 feet and then the finders can be off 15 feet. I just ignore comments of coords being off unless they are over 35 feet.

[+AllenFamilyCachers]

Well, the people seem to be finding the caches fine, but are making comments about the coordinates. When averaging, I watch the accuracy bounce from 15' -30' and everywhere in between. I try to hit SAVE when it gets closer to the 15' mark since I assume it means the point is accurate within 15' of my waypoint. I'm still new to my GPS and the manuals only seem to offer up 'light bathroom reading' instead of any REAL instruction.

 

So what is the best method? How do you all do it? Should I just set the GPS down for a few minutes while it's averaging... then just arbitrarily hit SAVE? Or am I looking for it to hit a certain number then hit SAVE? Thanks for your help... keep the suggestions coming.

[+Sputnik 57]

I think the "standard" strategy is to set your unit near the cache (or hold it overhead) for about 10 minutes while it averages. Markwell has more in his helpful site here.

Edited March 13, 2006 by Sputnik 57

[Photobuff]

Averaging will help improve noisy data, that is, where the datapoints are jumping around, but centered on the right location. If all the data is 50' off, averaging will just produce a more stable number that's still 50' off. If you watch data using a program like SA Watch, you'll see there are times when the data is very good, and with a little averaging, will produce a very accurate position. You'll also see that there are long stretches where the number of satellites, position, and signal strength, are less than optimal, and all the averaging in the world won't give you a good position. My point is that averaging alone won't guarantee good results- GIGO. Watch your error indicator and number of satellites. When the error is low, and the satellites exceed seven or so, you should be getting very accurate positions regardless of averaging. If you wanted the ultimate accuracy (without buying survey grade equipment), you'd set up a laptop with SA Watch, and collect data for a few hours. With weighted averaging you'd probably find a window where the position was good to a foot or two. IMO, ten minutes of averaging provides little guarantee of anything.

[+jon & miki]

If the coordinates are really critical for a particular cache placement, average the coordinates on more than one day. The different satellite configuration and even where you stand relative to the GPSr can affect your accuracy (and the estimated error can still look good even when it's not - after all, if the GPSr actually knew how much error there was, it would attempt to correct it, wouldn't it).

 

One way is to use the average function as you've been doing. Another approach (the one we use most times) is to take multiple readings (maybe 5 to 10 readings, moving away and back to the cache between readings, letting the GPSr settle, taking a snapshot, repeating), put 'em all in a spreadsheet, toss any that are obviously out of line with the others, and average the remaining snapshot coordinates. (We hunt and hide as a team and we each have a eTrex, so we also average from more than one GPSr - they are often surprisingly different). You can also view the multiple coordinate sets on a map if you've got the software handy, then toss out any outliers that way and even average by mark 1 eyeball.

 

For caches where accurate coordinates are really a necessity, we average multiple samples from multiple days or at least ones taken several hours apart so that the satellite configuration has time to change significantly.

[+AllenFamilyCachers]

I have the Nav7 software on my computer... what if I just zoomed in to as great of detail as possible and clicked the place on the map where the cache is. If I made that spot a waypoint, would that work even better?

 

I'd rather not spend 10 minutes sitting in a spot where stealth is required anyways. Hopefully the software idea would produce nice results.

 

Thanks for your help... I'll have to try this option out when I get home.

[+jon & miki]

Maps can often be out compared to gps readings, sometimes by several hundred feet. I'd suggest you double check your results by physically going to the coordinates you read off the map and see if they are close enough so that cachers will be searching in the right area.

[planewood]

I just went out to the center of my cul-d-sac and took a reading of 100 averages with my 60cx. I had WAAS turned off and was getting 8 satellite bars maxed out and a reported accuracy of 11'.

 

Now, I don't know how accurate Google Earth is, or Terraserver, but when plotting the point in GE, it showed to be about 1' away from the center of the cul-d-sac. The best I could determine with Terraserver was 1 meter, but that was with the 1M/pixel photo's. The GE photo's are the enhanced ones in this area.

 

I did this same thing a few days ago with a local benchmark (AW1881) and it showed about 1' off, also. Both in GE and when compared to the stated lat/lon for the benchmark. That tells me that GE must be registered pretty well in this area.

 

I suggest you find some local benchmarks or triangulation stations in your area and make some tests.

 

With the 60cx, it seems that the more solid bars you have showing, and the taller they show to be, that you get pretty dang good accuracy!

 

BTW, my cul-d-sac was poured in 4 equal parts and there is a nice X right in the middle of it!

 

Also, I haven't seen WAAS 35 for several days???

[Photobuff]

I've found Terraserver photos and USAPhotomaps as an interface to be extremely accurate. If you have a good landmark, you could almost dispense with the GPS entirely!

[planewood]

Went back to the center of the cul-d-sac tonight and did the same test. This time the point plotted 1' on the other side of the bullseye. LonSeconds=9.3" as compared to 9.4" last night. Last night the sky was cloudy. Tonight is clear.

[+Sputnik 57]

I've found Terraserver photos and USAPhotomaps as an interface to be extremely accurate. If you have a good landmark, you could almost dispense with the GPS entirely!

The TerraServer Urban Area Lat/Lon coordinates are a little off in Houston, Dallas, and Colorado Springs. In Houston, the error is about 521 feet! USAPhotoMaps doesn't correct for their errors. I'm hoping that Microsoft will fix them someday. Meanwhile, you have to use the older black & white aerial photo coordinates, which are not as up-to-date or detailed.

[+CheshireFrog]

I've found Terraserver photos and USAPhotomaps as an interface to be extremely accurate. If you have a good landmark, you could almost dispense with the GPS entirely!

The TerraServer Urban Area Lat/Lon coordinates are a little off in Houston, Dallas, and Colorado Springs. In Houston, the error is about 521 feet! USAPhotoMaps doesn't correct for their errors. I'm hoping that Microsoft will fix them someday. Meanwhile, you have to use the older black & white aerial photo coordinates, which are not as up-to-date or detailed.

 

You've got that right. According to TerraServer I've been going home to the wrong house every evening. I don't know how widespread this problem is, but it's pronounced here in Colorado Springs.

[+nonnipoppy]

I have a Garmin 60CS and when marking a waypoint for a new cache, I have been using the AVERAGE button to supposedly get more accurate coordinates.

 

After doing the average for 100-200 averages we save the coords. Then we back off 100 feet or so and select GoTo those coords. If it takes us to the cache all is well if not we adjust the coords a bit and retry. Approaching the cache site from two directions and getting satisfactory readings make us think it is good.

 

A word about comments however. We had a new cacher visit some local caches and repeatedly commented that the coords were off 10-15 feet. Inexperience led her to believe that the coords should be dead on for all GPSrs. Three of us all use GPS60CS and we will ALL have different readings if we do a goto a cache and lay the GPSrs down. So I would say let some comments slide off depending on who makes them.

[Photobuff]

Wow, that's not so great about Terraserver accuracy in some places. Around here (upstate NY) the coordinates are about as good as you can read them. I'll have to be careful if I venture away from home!

[planewood]

Wow, that's not so great about Terraserver accuracy in some places. Around here (upstate NY) the coordinates are about as good as you can read them. I'll have to be careful if I venture away from home!

I download some terraserver data (using USAPhotoMaps) for NE OKLA and wrote down the N,S,E,W lat/lon's. I then used it as an overlay map in Google Earth registering it with the coords I wrote down. The underlaying GE image was terrible for the area. But, after registering the overlay, the GE Roads matched perfectly with those on the Terraserver image. They actually matched the TS image much better than their own GE image.

[+Malloc]

Yep, I hate it when I go to someone else's cache and it's way off. A few around here are known to be 60-80 feet off. I didn't want to be one of those people either. For both of my caches I've got out (one is a 4 box multi-cache) I used the method below. (oh, and I use a 60CS as well)

 

First I stood on site and did an average reading for 150-200 readings (depending on the numbers of mosquitos ). If my dilution of precision was too high, say >15 feet, I'd back out a known direction and distance (yes, I carry a long tape and a compass into the woods when placing caches) until my tree cover was better, lowering my DOP. I'd then do a reading here, as above, and project my actual waypoint the distance and direction to the cache. This helps overcome taking readings under big trees as we often find outselves doing down here.

 

I always come out the next day or two, and see how close my GPS takes me to my coords. If it brings me back to the right spot, then I use them. If not, I start over as above, and compare results. Do this until you're consistent. That's really about as good as it gets.

[+alanfreed]

Yep, I hate it when I go to someone else's cache and it's way off...

 

One of my biggest caching pet peeves is when a cache log is full of comments saying that the coordinates are off, yet the cache owner never bothers to check up on it. I'm glad you're at least concerned about the problem!

 

I'd try punching the coordinates into your own unit, and pretend you're hunting for the cache yourself. See if the coordinates actually lead you to the right spot, or if you find that they're as far off as people are suggesting. A little role playing like that can certainly shed light on how people are experiencing your cache.

 

Like others have said, if your receiver has an averaging feature, go for it! It seems like a great tool to help ensure that the coordinates are as close as you can make them.