Advanced Review of TRDI SxS Pro Data

Video Transcript
Download Video
Right-click and save to download

Detailed Description

This video discusses the basics of performing an advanced review of midsection ADCP data collected in TRDI SxS Pro utilizing BBSlice and WinRiver II.

Details

Image Dimensions: 480 x 360

Date Taken:

Length: 06:14:00

Location Taken: Augusta, ME, US

Transcript

Hi, my name is Nick Stasulis and I am a hydrologic
technician with the Maine office of the New

England Water Science Center.

In this screencast I will discuss how you
can perform a more advanced review of data

collected in SxS Pro, when needed.

First, let’s start with one important point.

This advanced review is not something that
should be done or needed for most midsection

data collected in SxS Pro.

Proper review during the measurement and at
the site once the measurement is complete

should eliminate most issues that would need
this detailed advanced review.

Collecting good quality data will allow you
to eliminate most of these issues, if they

are detected and addressed during data collection.

That said, let’s take a look at a measurement
that has some irregularities.

Notice the very slow velocities at the surface
nearly in every vertical.

The velocity profile plot indicates this data
is large magnitude negative velocity, which

is likely not realistic.

We see in the contour plot that water error
velocities are fairly high in some of these

bins, but not all.

We have the ability to change the error threshold
for each vertical, or the up threshold for

the entire measurement, but there are no good
tabular views in SxS Pro to determine reasonable

values.

Rather than use trial and error, let’s load
this data into WinRiver II so we can utilize

the better data viewing options there.

First, open BBSlice, a program that comes
as a part of RDI Tools.

For the input file, select the pD0 for the
measurement, which is the raw data file.

For the output file, I like to use Sliced,
which will create a file for each vertical

named Sliced000, 001 and so on.

Click the Slice button and the utility will
create a new output file for each vertical,

and will include unused verticals.

In the measurement directory, we now see the
new sliced data files with the .000 extension.

In WinRiver II, select create measurement
from data files, navigate to the measurement

directory, type *.000 in the file name box,
and select all of the sliced files.

For each vertical, now a transect, you will
get a warning box stating there is no associated

configuration file.

Simply click ok for each.

You’ll notice in the playback configuration
that there are no processing settings, no

transducer depth, or other information, it
is simply the raw ADCP data.

You could enter this information, but we will
leave it as is for the purpose of this data

review.

You will also notice the reference is set
to bottom track, which could cause errors

if there was a significant bias in the bottom
track data, so consider changing the reference

to none.

Now that the raw data is loaded, you can reprocess
each transect, essentially allowing you to

review one vertical at a time, with access
to the tabular and graphical options in WinRiver

II that are not available in SxS Pro.

In this measurement, we were seeing negative
velocities near the surface that seemed to

be erroneous in SxS Pro.

Brining up a velocity tabular allows us to
see that most of the error velocities near

the surface are high.

Our water track error threshold in SxS Pro
was set to 1.969, which is high enough that

it really isn’t screening much of the invalid
data near the surface.

Scanning through the velocity tabular in WinRiver
II for several verticals indicates a value

of about 0.5 would remove most of the data
near the top with high velocity error values.

Back in SxS Pro, we note an as-is discharge
of about 18.3 cfs for this measurement.

We can first go into the Processing dialog
and set the good ensemble to cell setting

to 5, which will force the software to require
5 valid ensembles to compute a bin at each

depth in the vertical.

Then, we go into each station and manually
change the error threshold to 0.5.

Be sure to check the profile plot and ensure
all of the invalid data is being screened.

If you still see erroneous values, decrease
the threshold value, and if you feel too much

data is being screened, increase it so the
erroneous values are removed, but enough valid

data remains.

Once all the erroneous data is removed, the
final discharge is around 46 cfs and looks

much more reasonable in the contour plot and
profile plot, though the indicated angles

are inconsistent.

You would want to be sure to document in your
notes what changes were made and how the values

you applied were determined.