Recently, three colleges converged in western Nevada for four days of field paleontology. Over thirty students participated in a series of field tasks that represented different kinds of field scientific work. The tasks ranged from classic description to modern collection techniques. For the latter, the students mapped a trilobite-rich area using a Nomad computer and SX Blue II sub-meter GPS from Walcott Scientific.

Students found the gear not too cumbersome to wear and use in the field.

Trilobites are extinct arthropods, related to modern insects, crabs, and lobsters. In the Cambrian period, about 500 million years ago, they ruled the seas. Deep water deposits of that age are rich in trilobite fossils. The field exercise was carried out in Esmeralda County, Nevada in glorious weather. The students divided into teams of four members and each team had a guide to the common trilobites. Then, they students scurried about the hills looking for fossil-rich zones. When they found a particularly rich area, the students logged their location with an SX Blue II sub-meter GPS that communicated to a TDS Nomad via Bluetooth.

Nevadia weeksi, common trilobite in the lower field area.

Data was collected as a point-based shape file in ArcPad, set against a topographic map downloaded from a state of Nevada GIS website. In addition to the location information, the students listed the dominant fossil genus and estimated the age. Both the Nomad and GPS worked without a hitch and there was plenty of battery life at the end of the project.

Back at the college, the data was transferred to MapWindow GIS and sent via Shape2Earth to GoogleEarth as a .kml file. This process took less then five minutes. Each student was then e-mailed the .kml file so they could see for themselves just how the fossil bugs were distributed in the field area.

This assignment, and similar ones I have done with my courses, really shows the students both how easy it is to collect data in the field, and how GIS adds value to a project. The Nomad proved an excellent tool in the Nevada desert with the bright screen and rugged design. While many researchers still use tablets and PCs in the field, the use of small PDA data collectors is more efficient on cost, weight, and battery life.

I’ve written several times about COM ports and GPS receivers. Interestingly, in the last two days, I have spoken with four different folks about how this works, particular in regards to setting up a GPS receiver through a USB port on a tablet computer. It is valuable to understand how this works to avoid frustration in setting up software later.

What is a COM port?
A communication (COM) port streams data into the computer. Older computers had serial ports for this purpose. USB ports are not really COM ports, so you need to download a driver for the conversion. There are two facts about this process: you need a different driver for each different brand of GPS or other devices using the COM port, each USB port needs to be set-up as a COM port and each will be given a different port number. This COM port number will be used in any software that needs to connect to the GPS. (As a side note, you will also need to know the speed of the data stream from the GPS called the Baud rate.)

Setting up a COM port for an SX Blue II Sub-Meter GPS
The first step is to download the driver. I can’t post folders in the blog, but I put the driver on the Walcott Scientific website. This will comprise a folder with various .dll, security catalog, etc., files for the FTBI driver. The folder is zipped, so you need to unzip it. Then, place the folder somewhere you can find it, like the desktop. Now, power up the SX Blue II and plug it into the USB port you wish to use. The computer will open the “Found New Hardware Wizard.” No need for the internet, so click “No, not this time” and click “Next.” Then, choose “Install from a list or specific location (Advanced)” and “Next.” On this page, choose “Search for the best driver in these locations” and “Include this location in the search:” Use the Browse button to find the folder and click “OK” and “Next.” A warning box will appear that the USB Serial Converter has not passed the test. Click “Continue Anyway.” The driver is now installed and you will close the Wizard. But wait! Another Wizard will open and you will repeat the process. You’ve done it!

Finding the COM port
So, what is the COM port? You can use a ‘port sniffer’ like GPS Viewer which works pretty well for finding GPS devices. Alternatively, you can go to: Control Panel>Administrative Tools>Computer Management>Device Manager and click on the “Ports (COM & LPT)” to see the port number assigned to the SX Blue. This is the COM port you will use in your software. The SX Blue broadcasts at 9600 Baud, 8 stop bits, no parity.

As I have written in a previous blog, I am not very pleased with the integrated GPS on the new Toughbooks. However, the Bluetooth is very stable and opens up a variety of GPS options. The following goes through the steps of setting up an external GPS on a Toughbook 19 running Windows XP Tablet. Other models should be similar; Vista has not been tested.

1) Turn on the GPS. As always, make sure the battery is charged!

2) On the Toughbook, open the SD card slot and flip on the Wireless toggle switch, Then, click on the Bluetooth logo on the bottom bar. A box will appear. Click “New Connection”. Here is an important step: use “Custom Mode” or else the computer will assign a COM port that may not work with some software! Click “Next.”

3) The computer will recognize the GPS. Highlight the name and click “Next.”

4) The computer will assign the GPS to a serial port. Click “Next.”

5) On the next screen, uncheck “Use default COM port.” Now, you can choose your own COM port. Many software packages only go up to COM 9 for GPS, so I recommend choosing COM 8 and “Auto Connect.” Click “Next.”

6) On the next screen you can select an icon if you wish. Click “Next” and you are done! The COM port you have chosen will be the same for all software you use. This also means that you cannot run two different software programs at the same time.

To save battery life in the field, you can turn off the WiFi while you are running Bluetooth. In the lower right corner of the bottom tool bar is an icon of a lightning bolt with a circle. Click it to bring up wireless options. Choose Wireless LAN OFF and Bluetooth ON. The 802.11a doesn’t matter.

I had an interesting case this past week that while unique, might shed light on some problems. A researcher has been using a CompactFlash GPS card in a Recon handheld computer without problems for nearly a year now. He e-mailed me that the GPS was working, but was coming in on a different COM port. Well….that can’t happen! COM ports, or Communication Ports are the electronic version of docks. They allow for data to be exchanged and, like a real dock, only one ‘vehicle’ can be there at a time. All computers use COM ports. Some have designated COM port numbers but allow you to create virtual COM ports. On a PC, it is typical for a serial port or an SD slot to have a unique COM port. If you have an integrated GPS that has a designated COM port. Because USB ports are wired differently, they require a driver to convert them to COM ports. Once a port is identified as a COM port, it stays a COM port. That is, the numbers don’t shift! On the Recon, the serial port is always COM1, the CompactFlash (either slot) is COM2, and Bluetooth devices come in on either COM8 or COM9. In the case of the “Mystery of the Shifting COM Port”, the CompactFlash jumped from COM2 to COM3.

Knowing the COM port for a GPS is necessary to run any software to read the data stream coming from the GPS. Unfortunately, the program I wrote for the researcher was set for COM2 so he could not use COM3. (As a quick fix, I re-wrote the program for COM3, but that didn’t address why the port should change.) Many programs allow you to figure out which COM port the GPS is connected to before you begin to collect data.

Back to the mystery, there are three scenarios: 1) The GPS tortured the computer into shifting COM ports, 2) there was a hardware malfunction in the computer, or 3) there was a gremlin in the operating system of the Recon. GPS receivers don’t have the capability to shift COM ports so that was ruled out. I was hoping the computer didn’t fail because the researcher was in the field in South Africa. So, I settled on the operating system. His Recon has Windows Mobile 5 which has a cool applet that allows you to re-route the GPS from its incoming hardware COM port (in this case, COM 2) to any outgoing COM port. The benefit is that many programs can read a virtual COM port. The analogy is that you phone can only take one call, but many receivers in your house can listen in. We tried changing the parameters on the GPS applet but that didn’t work. Dang. It seemed the most likely scenario (and the cheapest) was that the operating system flaked out. With Recons, it is really easy to restore the operating system. You do a hard reset by holding down the power button for a few seconds. When the yelp of a rapid double beep occurs, simultaneously press the flying windows and enter buttons. This will restore the factory settings. Of course the factory didn’t put on any extra programs or data so if you try this yourself, BACK UP ANY FILES OR PROGRAMS YOU INSTALLED!!

The result? One e-mail later, the researcher restore the factory settings and the CompactFlash GPS is back on COM2. The “Mystery of the Shifting COM Port” is closed. The lesson is, if it seems like the operating system is tweaked, return to the factory settings before you toss the computer. Oh, and back up the files!

Recently, ABC put out a news snippet about the “child killing” inaccuracies of GPS vehicle navigation units. Glossing over the semantically misleading title, the report did a decent job of quickly summarizing the obvious—navigational software is prone to errors. It is not that the receivers are inaccurate, but rather the GIS programming behind the directional software is not foolproof. Having driven sub-par vehicles all my life, I am well cautioned to not blindly trust ‘foolproof’ gauges like gas meters, speedometers, turn signal indicators, etc. I do trust billowing white steam as a statistically likely indicator that my car is overheating. So, where is the fault with the navigational systems and what can you do to lighten your chances of becoming a GPS child killer?

The various driving programs packaged by TomTom, Magellan, Garmin, DeLorme, and others, rely on a base map of roads. Already, there is going to be a spread of accuracy based on the detail of the map and the age of the data. Roads change; in some areas, a map can be outdated within a year. I live on a street that dead-ends against a creek and then shows up on the other side with the same name. It is not uncommon for folks to knock on our door wondering where the rest of the street lies. Some software allows for upgrades but it is up to you to stay on top of things. And, it is unlikely that daily, weekly, or monthly closures will appear. Secondly, in the better mapping packages, the roads are ‘weighted’ for traffic speed and other flow parameters. That way, you can mathematically deduce which is the best route even if it is not the shortest. We all know how this works whenever we try to explain to a friend or spouse that our way is “better.” Herein lies the problem. How many software packages are going to be more accurate than your knowledge, experience, and intuition for your neighborhood? Then why do you trust it for other neighborhoods you have never been to?

I blame the problem for the accidents and lost drivers on losing our ability to read a map and growing up in an increasingly virtual world. At some point in the Late Atarian, we evolved into a race that trusts virtual landscapes more than what is outside our window. Or, we just would rather inhabit that space. Why else would so many minivans and airplanes have TV sets instead of looking out the blessed window?

At this point, you may be wondering why I am writing about this on a fieldwork blog. Do you think the software for backcountry work doesn’t suffer the same shortcomings? In future blogs, I would like to explore how to improve the software we use for monitoring and recording our field work.

What is the latest trend in GPS? For my money, there will be an expansion of ‘geocoded’ cameras; cameras that record the current longitude and latitude and affix this data onto the image. While the technology has been around for some time, both on the commercial market and as open-source software, the market demand seems to be increasing.

Camera with GPS or GPS with Camera?

Looking at the current market offerings there are two out-of-the-box solutions. First, there are a few cameras with GPS capabilities. For now, I am not overly impressed with cell phones that have cameras and GPS receivers as the quality does not seem that great. (Readers: if you have a cell phone that you are happy with, please let me know!) The leader is the Ricoh 500SE camera. This is a well-built, 8.0 megapixel SLR that comes with Bluetooth already installed and a GPS receiver. Simply take a photo and the location information is stored in the image tag (known as the EXIF header). I was not able to get details of the integrated GPS, but the camera does accept NMEA data from other GPS receivers. Therefore, you can use the camera with a top-quality sub-meter GPS like the SX Blue.

Screen shot of Ricoh camera image (from Ricoh website).

The other option is to buy a GPS, or GPS-enabled computer, that has a built in camera. The photos will be of a lesser quality but could be ‘good enough’ for most jobs where you are recording field data for projects. For instance, you might be out checking on damaged trees for a county assessment and you are recording the tree data, including location. At some trees, you may want to take a picture and store this information as well. The advantage of this solution over the Ricoh is that you get the full computer benefits.

Finally, a third path is to use a software that allows you to integrate GPS data and photos. The ESRI folks already have a program, GPS-Photo Link (ArcPad Edition) that allows you to do this, and more will follow suit. I have also come across quite a few links online to folks that have created their own software, but I haven’t tested any. Again, if you know of other software, let me know.

Test of the new Nomad computer with GPS

Recently I tested the Trimble (nee Tripod Data Systems) Nomad Computer with integrated 2.0 megapixel camera. Firstly, I am very impressed with the computer and recommend it to folks who need a powerful and ultra-rugged handheld computer—particularly if you need USB-in capabilities. The camera option has a small camera in the top of the computer, under the weather-resistant see-through casing. The good news is that the camera is easy to use, and has the basic functions such as variable size and quality settings, white-balance, etc. Taking photos and later retrieving them was very straightforward. The bad news is that the photos are not geo-coded even though the computer comes with a decent, SiRF III embedded camera. In fact, I could not find software on the computer for creating geo-coded images. It would seem to me that a rugged field computer with GPS and camera from the factory would come with software for geo-coding standard. Trimble, when will this be available?

Photo from Nomad camera.  Setting was small-size and medium resolution.  Slightly overcast skies and using default settings.

Here’s one of the lessons you learn the hard way.   Sub-meter GPS receivers, like the popular SX Blue, keep an internal power source running so the computer can maintain clocks and other functions.  While personal computers do this as well, the drain is much more significant on the GPS receivers.  So, make sure you keep the receiver charged up!  I went to do some field collection and had a dead receiver.  Fortunately, the new SX Blue has a power level LED function so you can monitor the levels.

I haven’t seen this in press before, but my experience is that GPS receivers, whether commercial or high-grade, will not perform as well with low power.  This also includes GPS receivers driven via USB or CompactFlash slots on other computers.  You will have enough power to set the lights off, but the reception suffers.  This may have no scientific validity, and I haven’t conducted controlled tests, but it seems to be the case.  Regardless, keep the batteries charged!

In the last blog, I gave an introduction to dilution of precision or DOP and explained when you need to know it and when you don’t. This time, I want to show you a very cool, free program to calculate your DOP values throughout the day. All you need to know is your latitude and longitude and day. The program is called “Planning” and is from Trimble. You can download it here.

To use this program, you need to first enter your position through File > Station. You can use the map feature to get ‘close’ but you really should use the correct latitude and longitude. On this screen, you also set the date and time and duration of observation. Then, you need the satellite almanac information. This is downloaded from the Trimble site as an ‘.alm’ file or you can use an ‘.ssf’ file.

Here we see a plot that shows the total DOP at 10 minute intervals for San Diego (where it is very lovely right now). Looks like 7:30 a.m. is the time to work.

That’s it! Now, play around with the different map features, it is surprisingly easy to navigate. Notice that you can use GPS (United States system) or Glonass (Russian system). WAAS is available but since those satellites don’t affect the DOP, they are not included in the almanac.

This cool screenshot shows the visibility of each satellite throughout the time period. You can turn satellites off and on to clean up the graph.

But wait…if the DOP is based on satellite positions, what happens if you cannot see a particular satellite because of mountains or buildings? Obviously, that satellite is no longer used in calculating the DOP. The “Planning” software doesn’t directly allow you to blank out areas of the sky (but you can turn off certain satellites). However, there is another great program that allows you to show the blocked view based on topography. Stay tuned…

GPS receivers are so cool because you turn them on, wait a few minutes, click a button, and bingo—there is your location. For most applications, you don’t even need to care how accurate the location is because either accuracy isn’t important to you or you are using maps where the scale is so large, a sloppy error circle wouldn’t show up anyway. By example, the pencil here on my desk makes a mark about one millimeter in diameter. Let’s say you have a GPS receiver that is accurate to about 10 meters (not unrealistic for a Garmin or Magellan). That means that at a scale of 1:10,000 or larger, you wouldn’t see the error plotted on the map.

But what about when location is critical because you are monitoring changes over time, or you have found Flint’s treasure and you don’t want to be digging three feet away and suffering the same fate as Ben Gunn? In that case, you want to record the accuracy as well as the location and that is the DOP factor.

DOP, or ‘dilution of precision’ is a unitless value that relates the accuracy of the satellite measurements based on their positions to the computed solution. Therefore, the lower the DOP, the more similar the two values are and the more ‘accurate’ the value. Since the satellites are constantly in motion relative to your position, the DOP changes as well. The more widely spaced the satellites, the lower (and better) the DOP. The best situation is one satellite directly overhead and several others spread out on the horizon, giving you a wide spread. There are several DOP values commonly reported: PDOP is the position DOP and is based on three coordinates, HDOP is the horizontal DOP based on two coordinates, and VDOP is the vertical DOP which is only based on height.

The DOP value acts as a multiplier to the error of the measurement (the calculated position). That is, with a DOP of 2 and an error of 5 meters, the accuracy would be 10 meters. Because of this, many programs will allow you to set a minimum DOP for data collection. Usually, the default is a value of 6. To get sub-meter resolution, you would need a DOP of 4.

Let’s say you are planning a mapping project and want to increase the DOP of your measurements, there are several things you can do in preparation. The simplest method is to set a minimum DOP and only collect data during those times. Again, the shift in value occurs because the relative spread of satellites is changing. Second, you can occupy one location for a longer time, gathering more data points to average. This usually results in a lowered DOP, though you could increase your average if the satellites get worse. Third, you can chart the satellites for your area and plan on collecting data during the best times. Because the satellites are in geostationary orbits, they will be in roughly the same positions each day for several weeks.

In my next blog, I will discuss a fantastic free program that charts your DOP for you.

Recently, I ran into an old professor of mine and we chatted about the use of computers in field geology. Both of us were trained using little more than compasses, 1:24,000 topographic maps and aerial photo interpretation. And we both agreed that students need to be trained in these methods. “So,” he asked me, “when do you decide to spend several thousands of dollars for good field computers?”

The answer I gave is that you bring a computer in the field if you need data handy. Data to read in the field or data to collect in the field. There is a real advantage to processing data in the field so you can shift priorities as you are working. Of course we are always collecting data, but not all data needs to be digested with a computer processor. One advantage I have seen recently is recording repetitive data. If you use a database program with drop-down lists, you are less likely to incorporate spelling errors and you can have the program prompt you for data so you won’t miss key features.

It is also standard these days to use a GPS receiver to locate key locations, but it is not as common to use the devises to track lines or areas. However, with a good receiver, you can often track your moving position directly onto a digitized map, saving time later.

Sounds great, but what is the catch? The obvious answers are that field computers are costly, take up space and add weight, and are totally useless if the battery dies. A less obvious answer is that we run the risk of creating a virtual ideal, even when we are outdoors. I have seen students rely directly on GPS receivers even when the receiver shows them in a gulley when they are standing on a ridge! Or you turn the world into a binary playing field when we know reality is probably more fuzzy.

The bottom line is that field computers and GPS receivers are fantastic tools, but we still need to teach the underlying fundamentals of location and data collection and interpretation. In future blogs, I will outline specific research and teaching projects I have done to illustrate successes with field computers—successes that justify the cost of bringing these tools in the field.