GPS
Myths
Alaska Support Office GIS Team
This document is designed to dispel GPS collection myths and give
sound advice for digitizing with GPS receivers intended for GIS.
Myth: My recreational grade GPS always gets data when my expensive
receiver does not. Is it worth to keep this expensive rock around
any longer?
Reality: GPS units vary in the ability to see GPS signals
and allow a position to be collected. Usually, for GIS data collection
purposes
(at least in the NPS), the horizontal position needs to be within
a couple of meters from the true location. More expensive mapping/resource
grade receivers can be set to automatically filter out poorer
quality positions. Recreational grade receivers will allow a position
to
be collected at any level of horizontal positional accuracy and
do not store any estimates of error.
Myth: The GPS unit is the most important item to purchase. Software
is secondary.
Reality: Not anymore. With Windows CE-based dataloggers
the choices of mix/matching the GPS receiver, the datalogger,
CE-based GPS
software and PC software must be considered before deciding on
a “GPS unit”. A GPS unit may be cheap, but after you
add the software, training and cabling to get a “system” to
work you may equal the amount of money on a more expensive GPS
system that pre-plans, structures the attribute collection, collects
and filters the GPS data and post-processes for the GIS. Don’t
get caught up in going cheap, since the real work begins AFTER
you have collected the GPS data. Some GPS solutions leave you with
limited means to pass into the corporate GIS, high quality data
and accuracy statements for documentation.
Myth: Now that I purchased a receiver with WAAS, I don’t
need to augment GPS positions with realtime DGPS equipment or post-processing.
Reality:
Relying on WAAS for 100% of your DGPS needs is not realistic.
WAAS is designed by the FAA to guide aircraft down to 200’ above
the surface of the earth. WAAS differential will not work effectively
under tree canopy and when the southern sky is occluded. Relying
on WAAS for your DGPS needs will introduce data into your corporate
datasets that may be more accurate sometimes and not so accurate
other times. How will you know? A better solution is investing
in real-time differential systems or those systems that utilize
post-processing differential. Using equipment that post-processes
ensures 100% differential processing for your GIS needs.
Myth: All
I need is the GPS receiver and no external antenna.
Reality:
The key to receiving high quality GPS data is a clear view of
the sky and an un-interrupted strong signal from at least
4 satellites in a good geometric pattern overhead at any one
time. Since many mapping missions may be in hilly or mountainous
regions, under tree canopy or from within a moving vehicle,
holding a GPS unit in your hands can block crucial access to satellites.
An external GPS antenna placed above a mapper’s head will
maximize the greatest number of satellites to be used in a GPS
solution, and often times increase your efficiency in poor GPS
environments. An external antenna also allows you to free up
your hands to write things down or place the GPS receiver in
a storage location (like a pocket) during long stretches of monotonous
terrain. We encourage buyers to research the accuracy differences
between their GPS receiver’s internal and external antenna
and buy the best one available.
Myth: With GPS accuracy getting
better and better, there is no reason to average a point location.
Just collect an instantaneous
position and move on.
Reality: Averaging more than one GPS position
while standing still will generate a more accurate position due
to simple math. GPS
errors are still present and fluctuate; averaging these multiple
positions provides a more accurate point.
Purchase GPS units that
Average: The following Garmin GPS receivers allow averaging (Garmin
III+, 12XL, Map76, 60C). The eTrex series
cannot average. All Trimble receivers average positions by default.
Log
enough positions. Be sure to log at least 180 positions (For
Garmin receivers, that’s 3 minutes at 1 position per second.)
Stand
still, but wiggle! This may sound silly, but when averaging a
position, you must be not moving. However, in poor GPS environments,
like a dense forest, a small movement of the antenna may provide
you a position that a few seconds ago was not possible. These
subtle
antenna movements can provide the crucial signal that allows
a position when other GPS quality filters (like PDOP/HDOP, SNR,
Elevation
Mask) are set appropriately for the situation.
Myth: I simply don’t trust storing my GPS locations on the
GPS. I write all my locations down and transfer them into GIS when
I get back to the office. I think this is the best route:
Reality:
This is a bad idea. Not electronically storing a GPS position
places one in the situation where datum’s, coordinate formats
and accuracy can all be dropped or confused when entering into
a GIS. User error now vastly exceeds GPS error especially since
electronic storage of GPS data in receivers is tightly controlled
in certain coordinate formats, significant digits and datums allowing
for the best position possible. Once the data is stored internally
in the GPS, the appropriate software used back in the office will
ensure that the data is transferred in the highest quality possible.
Contact your park’s GIS coordinator to find out what software
should be used to download the GPS. If you must write down a coordinate,
then you must write down the datum chosen on the GPS screen display,
units and estimated error during collection.
Myth: I don’t need to worry about coordinate systems and
datums when using ArcPad.
Reality: Three questions must be ascertained
about coordinate systems and datums before you invest in an ArcPad
solution for your GIS
needs. Not asking these questions first will cause considerable
heartache, pain and errors when merging ArcPad data back into
your GIS.
Question 1: What is the datum of your GIS Layers?
You should
have PRJ (Projection) file assigned for all your GIS layers
BEFORE using
ArcPAD and they must be all in the same projection. This is
because ArcPad “listens” to the PRJ file to assign how to
transform GPS input on-the-fly into a shapefile. If you add a
layer with
no projection (.prj file) to a map with a geographic projection,
ArcPad checks if the data appears to be lat-long (geographic)
and will assign those data a datum from the apDatums.dbf file.
If multiple
datums exist with the same name (there are 21 variants of NAD27),
then you must use the Select Default Datum tool in the Layers
dialog box to select the datum you want. Incorrectly defining
the layer’s
datum in the field can cause transform errors in relation to
the GIS layers back in the office. Be very careful and ensure
you use
ArcGIS to change the projection of your data BEFORE adding it
to your ArcPad map.
Question 2: What is the datum of your incoming
GPS positions from the GPS receiver?
ArcPad has no control
over incoming GPS data,
but relies on GPS manufacturer protocols to send data in
certain formats and datums. Garmin receivers set to NMEA protocol
are
dependent on the screen display of the datum. Altering the
datum on the Garmin
can alter the GPS input so that ArcPad receives data in the
wrongly assumed datum! Always set any Garmin in NMEA protocol
to WGS84
Datum and never change!. You do not need to worry about TSIP
protocol receivers (Trimble), or PLGR’s since GPS input
is always WGS84.
Question 3: Are your GIS layers in NAD27?
If so, you should
not be using ArcPad for high resolution mapping. Since ArcPad
cannot use NADCON, a high quality Datum transform engine, considerable
error may result in your GPS positions (1–15 meters).
In order to fully utilize GPS technology for your high resolution
GIS mapping needs, you should transform all your Park data
to NAD83
as soon as possible. Not only will this remove datum transform
errors inherent in the older spheroid/datum based coordinate
systems, but ensure that your GIS is in the best position
possible to receive
data from the most incredible digitizing tool we have—GLOBAL
POSTIONING SYSTEMS.
Happy GPS’ing
Joel Cusick
Alaska Regional Office
National Park Service
Document Date: 9/14/04
PDF
version—best for printing (32 kb)
|