WMOD00412

• Title: Final Report on Utah Cloud Seeding Experimentation Using Propane During the 2003/04 Winter
• Date: 3/1/2005
• State: UT
• Weather Modification - Doc Type: Report

I

I

The Goodrich stmsor and mounting hardware were again removed from the HAS tower on 10 February.
A new mounting arrangement was manufactured by the lead field technician. The new mounting placed
the instruments higher above the previous mounting plate, and reduced the amount of structure against
which rime ice could build. Also, the unit was rotated 180 deg which faced the cooling vanes into the
prevailing southwest winds with the hope this would improve performance. A number of fair weather
days followed during which an older style Rosemount tower type icing sensor was removed from the SIR
station and mounted near the HAS Goodrich instrument to provide a basis for comparison. The
Rosemount did not indicate icing rate by outputting frequency shift but only heater relay "trips" when the
instrument internally detected a sufficient decrease in frequency. These trips do not provide the high time
resolution of the Goodrich as typically only one or two trips would occur during a 40 min EU, or perhaps
between EUs. But the trips certainly indicated periods when SL W cloud existed for tens of minutes or
more. No further changes were made to the Goodrich mounting, orientation or 34 Hz threshold
requirement, during the remainder of the 2003/04 field season which ended on 30 March.

I

I

I

I

A rapid storm passage on 17 February resulted in a single pair of EUs, with northwest flow. This was
the first EU declaration with northwest flow since the initial storm of 14-15 December. Examination of
the raw 1 min data revealed that a limited amount of icing occurred with a wind shift from southwest to
northwest just before the first EU was declared. No significant icing took place during the pair of EUs.
The Rosemount icing sensor now mounted near the Goodrich did not trip and only trivial snowfall was
observed by the gauge network.

I

I

Storm passages on 26 and 27 February resulted in several EUs with the Goodrich sensor output
appearing normal, and frequent trips by the nearby Rosemount icing sensor. A major storm impacted the
area from 1700 MST on 28 February until 2000 MST on 29 February which produced well over 2 inches
ofSWE in most of the gauges. This was the most intense storm of the season and, according to the lead
field technician, the most intense seen locally in five years. The Goodrich sensor triggered the first EU at
2015 MST and its record generally appeared normal with a few minor fluctuations until about 0810 MST
on 29 February. At that time frequency data again became noisy and the final pair of EUs, starting at
0838 MST, had suspect data. The Goodrich did not trigger any further EUs for the remainder of the
storm.

I

I

I

Many HAS Rosemount trips occurred from 2000 MST on 28 February until the last trip at 0250 MST
on 29 February, over 5 hr before the Goodrich data became noisy. This suggests that rime ice built up on
the Rosemount housing until the sensing rod was shielded from further icing. The same likely occurred
several hours later on the Goodrich instrument. Supercooled liquid cloud may have existed for several
more hours during the storm passage but the two HAS icing sensors could no longer respond to its
presence.

I

I

A field technician was dispatched to the HAS the morning of 01 March to examine and photograph the
icing sensors and their housings and mounting hardware. One of those photographs is shown as Fig. 13.
Two to three inches of rime ice extended from vertical tower supports, various hardware and cables, and
from the icing sensor housings. The Goodrich cooling vanes are seen to be packed with rime ice. The
Goodrich and Rosemount vertical sensing rods had rime ice buildup to windward of the rods. A "finger"
of rime ice, tilted toward the left, is apparent as the highest visible object on the Goodrich sensor. (The
long vertical lightning rod itself well behind the sensor.) This rime finger does not appear to be attached
to the sensing rod itself but is located windward of it, in a position to block further icing on the rod. That
blocking effectively rendered the sensor useless. This same phenomenon was observed during field
testing of similar icing sensors conducted on the Grand Mesa of Colorado, and also in laboratory tests
(Super et al. 1986).

I

I

I

It was obvious that the long period of intense icing was more than could be kept ice free by the sensor
rod heaters and heated cone shaped structures from which the rods extend upward. An attempt was made
on 01 March to improve the deicing capability of the Goodrich sensor by reprogramming which allowed

I

27

I
I