Sequoia LISST-SL, Руководство по эксплуатации

Страница: 46 / 56

LISST-SL V2.1
User’s Guide
42
APPENDIX D: Background Files
The single most important file, other than
the data file, is the Background file. This
file indicates the health of the instrument.
It shows:
If the optics are aligned;
If the windows are clean;
If the water you used was at
close enough temperature to the
instrument that scintillation did not bias
your Background measurement. We explain these further and how to decide
when you can proceed further.
Are the optics aligned: If the Background you capture is very similar to the factory value,
with the ratio Lp/Lref also within 98% or so of the factory value, the optics are in excellent
shape. Accept the Background file and proceed.
When the optics become mis-aligned, the following are seen on the Background: the inner
rings will become higher, but only ALTERNATE rings will be higher. This is usually
accompanied by a lower Lp/Lref ratio also. Small changes may be acceptable.
Are the windows clean: Dirty windows usually add light on the middle rings. In this case,
your Background will be higher in the middle than the factory value. Try cleaning the
windows with the window cleaning tool. Try 3 times. If it does not change, minor scratches
on the window may have developed due to sand grains blasting the windows. It may be OK
to proceed. [Read the Noise Test below].
Is it scintillation: Scintillation is the scattering of light by turbulent eddies with temperature
differences between water and instrument. The identifier for this is that ALL INNER RINGS
increase in magnitude, not just alternate ones (which indicate misalignment). The simple
cure: let the water come to instrument temperature. It is for this reason that we advise you
to make a Background measurement in the river with a 15 minute dip period.
“NOISE TEST” FOR ACCEPTANCE OF A BACKGROUND:
At any time, and particularly
after cleaning windows, and if the instrument optics are not misaligned, the background is
still higher than factory, the question becomes: can I go further? The answer is yes if the
new background is stable. If it is not stable, this will add noise to the measurement.
To test for noise, do the following. After capturing a background, switch to the MEASURE
mode, with the same water running through. In this case, the instrument should see no
sediment. The outputs on the touch-screen should be small and fluctuating numbers. This
is the instrument noise. Now you can judge if the noise level is acceptable. In turbid rivers,
a noise of 5 mg/l may be acceptable. In cleaner waters, you may tolerate only 1 mg/l or
even less.
In our experience, this number is often less than 1 mg/l. This noise is the most
vital criterion for accepting a background measurement.
In the end, it is noise on your measurement that determines your confidence in the data.