gauge configured in a full Wheatstone bridge
repeat. Unless there was radio interference or the
powered by 10 V DC. The output of the strain
batteries were weak, the first scan was within the
gage bridge is a function of the drive voltage. To
error checking. The status of the fish was read
reduce the precision error, the transducer's drive
following the transducer readings at 15-minute
voltage was read as part of each scan. To reduce
intervals.
random error, each scan was the average of five
A beacon channel was used to ensure that the
readings.
data acquisition system and the receiving equip-
Instrumentation to receive and process the sig-
ment were aligned and monitoring the same fre-
nals from the instrumented fish consisted of an-
quency. The beacon had a unique frequency and
tenna, receiver, scanner programmer, processor
timing pulse along with a corresponding unique
and recorder (Telonics 1991). Transmitted signals
DC voltage. If this unique voltage appeared any-
are received by a 9-dB directional antenna on
where in the scan list other than the last channel
shore about 200 ft from the pier. Each instru-
the entire scan was rejected.
mented fish transmits at one unique frequency
between 140.010 to 140.220 MHz spaced at 10-
kHz increments. The receiver, which is tuned to
INITIAL RESULTS
one of the active frequencies by the Telonics scanner
programmer, isolates and amplifies the timing
Hydrometeorological data
pulse signal. Using an adjustable threshold, the
Water discharge records for water years 1992
recorder would time between signal pulses. For
and 1993, from an existing USGS gaging station
the active state the period was 2 ms, compared to
in West Hartford, Vermont, are plotted in Figures
10 ms for the inactive state.
16a and b, as well as in Appendix A. The gage is
An analog signal corresponding to the period
approximately 7 miles upstream of the study area,
and amplitude of the time pulse was read by the
and these discharge records have not yet been
HP 3421 measurement and process controller. The
corrected for ice effects. The apparent sudden
output voltages for the active and inactive state
discharge increases are attributed to an ice jam
were 2.5 and 3.5 0.4 V, respectively.
affecting the gage. An ice cover or ice jam blocks
Interrogating the status of the instrumented
a portion of the flow area and creates a hydraulic
fish was a multistep process. First, a digital pulse
drag on the water surface, which reduces the
from the HP3421 would step the scanner/pro-
river 's conveyance. This loss in capacity results
grammer to the next preprogrammed frequency.
in an increase in stage. Estimating discharge us-
The programmer automatically tuned the receiver
ing this higher stage with the open water rating
to the unique transmit frequency. Software would
curve results in the calculated discharge being
allow the receiver to settle in on the new fre-
unrealistically high.
quency for at least three cycles (30 ms) before
The average daily temperatures for the White
taking data. Once the active switch was tripped,
River region during the 1991-92 and 1992-93 win-
the faster pulse rate was transmitted for 2 min-
ters are also plotted in Figures 16a and b, respec-
utes. Readings were taken in a matrix format of
tively. Similar plots for water years 19601993
five readings per column and six columns wide.
appear in Appendix A. Ice formed during a sea-
To minimize variation of the signal, each column
son is a function of the duration and severity of
was averaged. To ensure the fish remained in the
the temperature. Both factors can be expressed
same state, i.e., active or inactive during the sam-
as freezing degree-days (FDD) using the expres-
sion 32F, where F is the average daily tempera-
pling aperture, the averages of the first and last
column were compared. If the readings were within
ture in degrees Fahrenheit. The volume of ice
5% of each other, the scan was acceptable and the
grown during the winter is reflected in accumu-
lated freezing degree-days [Σ (32F)]. In sum-
matrix average was used as a value for the scan.
If the scan was outside the tolerance, another
ming the degree-days, if the sum was negative,
complete scan was taken and the error checking
indicating warm weather, the accumulator was
process repeated. If an acceptable scan was not
reset to zero. Accumulated freezing degree-days
achieved in six tries, the sixth reading was flagged
for water years 19601993 are included as part of
as questionable and hard-copy diagnostic data
the plot in Appendix A. The AFDDs for the win-
ters of 1991-92 and 1992-93 were 1087 and 1029F-
were printed. The programmer would sequence
to the next preprogrammed channel using a digi-
day, respectively, with both years being approxi-
tal pulse from the 3421 and the process would
mately 15% warmer than the average AFDD for
11
Previous Page
