GPS for GIS Applications

Here's what I do. I'm working on my master's in environmental science with a GIS minor.

Aside from using school equipment and software that costs tens of thousands, I have a personal GIS setup that works.

I use a Garmin GPSMap76CSx for collecting data.
I use MN DNR Garmin freeware for getting my garmin data into .shp file format. It will put .shp files onto the gps, too.
Then, for my actual GIS data management and visualization, I use Manifold GIS most of the time. I bought a lower level license and I think it cost me around $300. Not too shabby since a comparable product from ESRI would be over $1000 easy. Sometimes I use QGIS, also, which is an open source multi-platform program. It has some good features, but others I don't like. Unfortunately, saved maps are not compatible from one program to another, so once you pick one you're stuck with it.

Also, good GIS software requires processing power. ESRI, last I was aware, was still stuck in XP-world and could not handle 64-bit software (and the large amounts of memory that allows). They only recently added support for multi-core CPU's. Manifold is keeping up with hardware better, so it can actually USE 10GB RAM if I have it, and it can use all 4 of my processor cores. QGIS is a lighter program not built for that stuff, so intensive functions take awhile. BUT, QGIS will run on Linux or Mac if that's your game. ESRI and Manifold won't.

As for the data you need, you'll need to talk to the county surveyor. You may have to digitize it yourself. It won't be perfect unless you get out there and survey, but should be close enough. It may be obvious enough on a basic topo or aerial image what the property lines are. Take a look. Old fencerows show up on those things for decades after they're gone.

At any rate, you can get by fine with a Garmin receiver...even a fairly cheap etrex. The big expense will be decent software once you get your data. I've never used XMap, so I don't have a clue how it compares to ESRI, Manifold, or QGIS.

PaintPeelinPbody said:

I've got access to arcmap 8.3 and probably would be able to use that for at least another year maybe more. Its the data acquisition that I'm interested in. I'm pretty strong. With data analysis and map editing with ESRI products but haven't done any data gathering although I may at my internship.

Natehawk,What do you hope to do with your graduate degree in environmental sci? I've often thought of doing that, Landscape Arch, or Civil Engineering after undergrad.

I think all of these could use this GIS cartography experience.

Click to expand...

Ouch...8.3....they've got you held back pretty good in your department, huh? 9.3.1 is the latest, and trust me, anything over 9.1 really is worth it. That trial/student edition is also worth a look if you can get a copy. They hand out copies at all my GIS classes these days. Does your school provide you access even to the ESRI teaching site? I'd hope so, there's good stuff in there, and unless you have paid extra access, the lessons are PRICEY. I took a seminar...free...last summer where they based the lessons off one of those things, it was about programming Arc in various languages to create customized tools and stuff. Really powerful, I wish I had room in my schedule to learn more.

I actually want to teach biology and/or environmental science...either high school, or as a naturalist at a nature center, or something like that. I'm getting the EnvSci degree, but my thesis is all wildlife biology, basically. Weird thing with how my school is set up, but my thesis involves heavy GIS modeling, which is more EnvSci than wildlife biology.

What area are you working in? If the county or municipality that camp is located in does not have parcel data, you can usually get a copy of the tax maps from the municipality (or county if it's unincorporated). Once you have a scanned tax map you can easily georeference it and draw in the property boundary. Trying to survey the boundary w/ a consumer grade GPS unit will probably be more difficult than is is worth, as you'd have to go by the description in the deed to locate the actual property markers (if there are any set) but if there are landmarks at the boundary, such as fences etc you can GPS them to help. In my experience, large wooded parcels in undeveloped areas are hard to capture because there aren't many control points to go by when georeferencing the tax maps.

Check out ESRI's ArcGIS online services to get free aerials (http://www.esri.com/software/arcgis/arcgisonline/map_services.html). There is pretty good coverage for ~1ft resolution aerials, but if they area you're looking in is pretty rural there might not be good quality aerials available.

As for the GPS, you can manipulate any GPS output into something you can use in ArcMap, but it will take some tweaking. I have a Delorme PN-20 and there is no direct translation into an ESRI readable format. I typically export a GPX file from delorme, open it in Google Earth to save as KML, then convert the KML into a shapefile (search KML to SHP in the ArcScript section of support.esri.com). You can also use GPSbabel to convert GPX to KML then use KML to SHP. There is also a script that will let you export data from AcrMap into a KML.

I recommend using a personal geodatabase to store your "final" data rather than a shapefile. You can create many feature classes and store them inside one geodatabase, so files are much more portable, rather than dealing w/ several files for one shapefile.

If there are fencerows visible as landmarks on an aerial photo, no need to gereference paper maps of any kind. No need to GPS in the field for boundaries, either. I worked for a park once under multiple jurisdictions and getting accurate property lines was nearly impossible. I took a combination of Google Earth, old park maps (from park website), and digitized an estimated boundary. Some areas were udner dispute, so I let other crews work with county parks to get that sorted out.

ESRI is not the place to look for aerials of any kind. The OP, being a student already, probably knows a local source of aerials and other base map data. His school GIS database may have better than 1ft resolution if he's lucky. That stuff is expensive, but available for some areas.

Dude, you have too many steps of converting! I'm sure you can find something to go straight from Delorme .gpx to .shp at least. I would expect the .gpx format to be universal enough that it shouldn't matter if it was Delorme or Garmin or Magellan. Maybe so, but Delorme does GIS stuff, so there should be a direct converter out there somewhere.

As for geodatabase, I'm not so convinced. I say use the format you can most easily use in ALL systems you plan on utilizing. If you will only do work at one place with one program set, then yeah, geodatabase works great. I'm a fan of the File Geodatabase, though. If, however, you will work on an Arc machine in one place, then go to a QGIS machine at home ('cause it's free), and maybe a third system somewhere else, ANY geodatabase is a bad idea since the different formats are not handled by all systems. Each system uses its own special type of geodatabase format. The best thing geodatabases work for can't use a consumer GPS, but a Trimble or other with ArcPad allows you to "check out" your data for editing in the field, and "check in" the new edited data once you return to the lab. You have to spend some time setting up your geodatabase, but once you do, the process of updating your data is SOOO much faster, especially if you have a small crew all collecting different, new data in the field.

I took a class on that so I could use it for my thesis project. I occasionally refer back to some ESRI lessons to make sure I get things right, though, as it's a learning curve involved.

rlb81 said:

What area are you working in? If the county or municipality that camp is located in does not have parcel data, you can usually get a copy of the tax maps from the municipality (or county if it's unincorporated). Once you have a scanned tax map you can easily georeference it and draw in the property boundary. Trying to survey the boundary w/ a consumer grade GPS unit will probably be more difficult than is is worth, as you'd have to go by the description in the deed to locate the actual property markers (if there are any set) but if there are landmarks at the boundary, such as fences etc you can GPS them to help. In my experience, large wooded parcels in undeveloped areas are hard to capture because there aren't many control points to go by when georeferencing the tax maps.

Check out ESRI's ArcGIS online services to get free aerials (http://www.esri.com/software/arcgis/arcgisonline/map_services.html). There is pretty good coverage for ~1ft resolution aerials, but if they area you're looking in is pretty rural there might not be good quality aerials available.

As for the GPS, you can manipulate any GPS output into something you can use in ArcMap, but it will take some tweaking. I have a Delorme PN-20 and there is no direct translation into an ESRI readable format. I typically export a GPX file from delorme, open it in Google Earth to save as KML, then convert the KML into a shapefile (search KML to SHP in the ArcScript section of support.esri.com). You can also use GPSbabel to convert GPX to KML then use KML to SHP. There is also a script that will let you export data from AcrMap into a KML.

I recommend using a personal geodatabase to store your "final" data rather than a shapefile. You can create many feature classes and store them inside one geodatabase, so files are much more portable, rather than dealing w/ several files for one shapefile.

Click to expand...

No one of us can say what will be right or wrong for the OP but I guess having a wealth of suggestions wont hurt, so he/she can decide what will work best for them. As far as the property boundary goes, I agree that any boundary that is also a jurisdictional boundary would be nearly impossible to get right especially for a student doing a side project. Any way you cut it, it's going to be an estimated boundary unless a licensed surveyor surveys the land and signs off on a plan that they drafted.

As for ESRI being a bad source of aerials, I must disagree. They are free, and it doesn't get any better than that. The only drawback is that you can't save them, but as long as you have a web connection they will always be there. If there are other high res flyovers that can be accessed for free then go for it, but that's not something you want to pay for unless it has some other use than a map landbase or locating some general landmarks. Anything better than 1ft resolution wont do much good for a tree covered area anyway.

You can pay for a student copy of ArcGIS (was $200 when I went for it) that doesn't expire so storing your final data in a geodatabase works in that sense, if you've got the software at home and work (or school). I haven't tried to create any sort of map output in software other than ArcGIS but I would tend to guess it has the most robust features in that area. I agree with using shapefiles for your intermediate data to be used on the GPS since that would be most efficient between the different software/hardware options you have to use outside of the professional or academic world.

I do go around my ass to get to my elbow in terms of managing my own personal data, but it works for me and I don't have to install anything that I didn't already have (google earth, topo usa) aside from GPS babel. I just don't put much trust in freeware so I tend to avoid it when I can.

ArcView 3.x definitely had its place back in the day and some of the "old timers" that I work with still fall back on it when **** hits the fan. I think there was a little resistance back when 8.x came out, but that's because the interface was completely revamped, so people had to re-learn a lot of things that they could already do w/o thinking twice in 3.x. It's definitely worth the learning curve at this point if you can get your hands on it, as 9.x is leaps and bounds ahead. Some of the newer releases were plagued with bugs, but 9.3 is a solid piece of software. 9.3.1 had some issues according to rumors I heard, but we still haven't tried it out yet.

NateHawk said:

If there are fencerows visible as landmarks on an aerial photo, no need to gereference paper maps of any kind. No need to GPS in the field for boundaries, either. I worked for a park once under multiple jurisdictions and getting accurate property lines was nearly impossible. I took a combination of Google Earth, old park maps (from park website), and digitized an estimated boundary. Some areas were udner dispute, so I let other crews work with county parks to get that sorted out.

ESRI is not the place to look for aerials of any kind. The OP, being a student already, probably knows a local source of aerials and other base map data. His school GIS database may have better than 1ft resolution if he's lucky. That stuff is expensive, but available for some areas.

Dude, you have too many steps of converting! I'm sure you can find something to go straight from Delorme .gpx to .shp at least. I would expect the .gpx format to be universal enough that it shouldn't matter if it was Delorme or Garmin or Magellan. Maybe so, but Delorme does GIS stuff, so there should be a direct converter out there somewhere.

As for geodatabase, I'm not so convinced. I say use the format you can most easily use in ALL systems you plan on utilizing. If you will only do work at one place with one program set, then yeah, geodatabase works great. I'm a fan of the File Geodatabase, though. If, however, you will work on an Arc machine in one place, then go to a QGIS machine at home ('cause it's free), and maybe a third system somewhere else, ANY geodatabase is a bad idea since the different formats are not handled by all systems. Each system uses its own special type of geodatabase format. The best thing geodatabases work for can't use a consumer GPS, but a Trimble or other with ArcPad allows you to "check out" your data for editing in the field, and "check in" the new edited data once you return to the lab. You have to spend some time setting up your geodatabase, but once you do, the process of updating your data is SOOO much faster, especially if you have a small crew all collecting different, new data in the field.

I took a class on that so I could use it for my thesis project. I occasionally refer back to some ESRI lessons to make sure I get things right, though, as it's a learning curve involved.

Click to expand...

I didn't say ESRI's database was bad, just that there are probably better local sites available. I wouldn't waste my time on super high res stuff unless it was already available. Long load times stink. But, the super high res stuff is good if you wanna print a big copy at Kinkos or on the uni plotter for the landowner. I have personally had bad luck with downloading maps from an online service. I've gone through database connections and all that, and big files (which imagery tends to be), take extra long to load. Not worth the extra effort, IMO, but if they have some good files of something else you need, go for it.

I personally don't care for the ArcView license...it's a little too limited for some functions I use. I'm used to the higher level licenses, but Arc really pisses me off on a regular basis. ESRI have done some things with that program that just don't make good sense. Nobody's perfect, of course, but for what I've paid, I'm happy with my Manifold license. It has a different interface that takes some getting used to, but it has a different set of limitations for its lower licenses than Arc, and that different set of limitations works better for me. Also, Arc flat out doesn't work on Vista, so that ruled it out. I wasn't aware of the student copy price, as nobody talked that up, but I imagine it's still one of the lower level licenses, anyway.

Some open source software is killer. OpenOffice is every bit as good as MS's Office, but for free. Some stuff fails muster spectacularly. QGIS is somewhere in between. It integrates with GPSBabel, so it's good for conversion. But, it doesn't work well with popular proprietary formats (like geodatabases), which hobbles it. It's really limited when it comes to generating attractive final printable maps. That's why I bought Manifold in the end. It still has uses, and since it's free, it's no big deal to install and just use when you need it. DNR Garmin is another free program that just works. It's small, does the job it's asked to do, and not much else. It's nice to take .shp files of say a boundary, put it on the GPS, and see where your position falls when scouting, or when checking to make sure a trail stays inside the park or something.

I remember 3.x. I think I still have a trial copy floating around somewhere. I remember how earth shattering 8.x was, and how much that changed things. Yeah, some versions of 9.x have been buggy, but 9.3 has been good to me. I haven't used 9.3.1 yet.