[Acoustics] GPS/ADCP questions

[Rainville,Francois [CIS-WSB]]

Hello Chris,
The fact that bottom track looks smooth does not mean that it is
superior to GPS under potentially biased bottom tracking conditions.
But as you pointed out, GPS as a reference has its own problems...
 
The vertical banding is usually caused by sat change errors or other
such thing.  Depending on the quality of data during acquisition, some
of these errors may cancel each other out since a sat change or
multipath is often followed by another one of same but opposite
magnitude.  These errors are common near the channel margins (due to
tree canopy) where the effect is then minimal on the total discharge
(less flow at the edges...).  Another source of such weird velocity is a
temporary loss of correction, a rarer event.  You should see velocity
spikes in the Ship Track (looking like ambiguity velocities).  There is
no way to filter these errors out right now, but even if you could, the
effect of filtering was found to be of no consequence on the final
result.  The GPS should overall be adequate, regardless of these faulty
points.
 
On the other hand, there has been cases where GPS data was not treated
in the computation on the same schedule and then the banding might
relate to a different kind of error.  The fact that coordinates match
near the end between your trimble (high precision) and a handheld unit
(lesser quality reference) may not indicate that the synchronisation of
data for its treatment was done right.  And yet another source of error
are the magnetic storms. This would screw up your GPS data by a lot.
But I do not believe such storm took place recently.
 
About rates, I do not know what the baud rate would do (but it has
nothing to do with your problem I would presume).
On the other hand, there is a debate about the refreshment rate for the
GPS reference.  It could be argued that you should have as many new
points for the GPS reference as you do for the ADCP, so that you get an
adequate match.  Some results seem to show that a smaller frequency for
the GPS output may have a negative impact on the quality of the value
(by how much and when remains to be determined).
 
Thanks,
Francois Rainville
Water Survey of Canada

Message from: 
francois.rainville at ec.gc.ca  613-992-1673 

	-----Original Message-----
	From: acoustics-bounces at simon.er.usgs.gov
[mailto:acoustics-bounces at simon.er.usgs.gov] On Behalf Of Christopher A
Curran
	Sent: Wednesday, May 10, 2006 12:42 PM
	To: acoustics at simon.er.usgs.gov
	Subject: [Acoustics] GPS/ADCP questions
	
	

	Hello, 
	
	1)  We did some ADCP meas. recently with an Ag132 GPS that shows
vertical banding in the WinRiver velocity magnitude contour plot when
GGA is used as reference, but when Bottom Track is toggled as reference,
the bands dissappear and the transects looks much cleaner. I was using a
usb-serial adapter for GPS, a 5Hz position rate and a 9600 baud on the
NEMA output.  The transects were done in saltwater bay during a tidal
change and were long,10-15 min. each. The GPS coordinates at start and
finish agreed with a separate handheld unit. The compass calibration on
the 1200KHz RioGrande was done correctly, but the unit is about 6 yrs
old now. Any ideas what's causing that apparent velocity banding w/ GGA
plot? 
	
	2) What's the best ADCP baud rate to use with GPS? 
	
	3) What's the best GPS baud rate into laptop (Ag132, output port
A for NEMA, I think 38.4K is max for the unit)? 
	
	4) Do I want to use the fast output option, 10Hz position rate? 
	
	Thanks for your answers. 
	
	Chris Curran
	U.S. Geological Survey
	WA Water Science Center
	Tacoma, WA 
	(253) 428-3600  x2614

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://simon.er.usgs.gov/pipermail/acoustics/attachments/20060510/493e587c/attachment.html