The Story of a DFIR

The last 10 days have been far busier than I anticipated. The weather for this field season, since I arrived, has been spectacular with mostly sunny to sunny days on nearly every day and warm temperatures. The result of that has been a continuous run of field work. We did two flights on the Friday before the Thanksgiving Saturday celebration. That Sunday evening we went into the field to one of the local sites for two hours, and returned that next morning for a long day at the site. On Wednesday we had a flight to Lorne APS in the morning and in the afternoon we took a quick drive to the Phoenix APS. Then on Friday we had a long afternoon with a crew of five at the Willie Field APS. The good news is that we are very close to being done with the field work. The bad news is that it has been a week since I have written a post.

The Double Fence Intercomparison Reference (DFIR) shield is the World Meteorological Organization standard for providing a shield around a precipitation gauge measuring snow. As a part of this project, we included a DFIR shield so that we could compare the measurements made with our Antarctic-focused wind shield and precipitation gauge design. The thinking is that we'd be able to develop a function that indicates the multiplier, in relation to wind speed, to be applied to our shield and gauge design to that of the DFIR.

The construction of the DFIR at the Willie Field APS site in November 2017 with the assistance of two carps.

The DFIR is a large and cumbersome shield to assemble. This is why we had no intentions of installing a DFIR at all four sites. Instead, we constructed the DFIR setup at the local Willie Field APS site so that it would be near to town to support the construction and maintenance of the DFIR. Part of the DFIR was constructed in Colorado before being shipped to Antarctica. Other parts and the assembly of the DFIR was done a year ago in McMurdo at the Carpentry shop. We also had two "carps" in the field with us for an entire day installing the DFIR. The finished DFIR has an appearance similar to that of a snow fence. The top of a DFIR "picket" was 78" above the surface and the outer shield has a diameter of 40 feet. The finished product looked great.

The finished DFIR at the Willie Field APS site in November 2017. The top of the shield is 78 inches above the surface and the diameter of the outer ring is 40 feet. 

During this past February we were seeing in the observations exactly what we had hoped to see. The plot below shows the three different precipitation gauges (we also installed one with the Antarctic double-shield at the same height as the DFIR) at the Willie Field APS site. The plot below shows the three different rates of accumulation depending on the wind shield and height configuration for the precipitation gauge.

A plot of precipitation accumulation at the Willie Field APS for early February 2018. The three different lines show the different rates of accumulation dependent on the different precipitation shield design and height above the surface.

However, as the austral spring turned to winter, something odd happened in the observations. The accumulation in the DFIR precipitation gauge (solid blue line) increased much more dramatically than the others. By mid-March, the DFIR gauge leveled off at values greater than 110 mm (liquid water equivalent), meanwhile the other two precipitation gauges were in the 60s. From afar in Colorado, this seemed really odd. The observations did not match what we were expecting. In fact, there was even some question if the instrument was failing. Adding to the drama, in mid-August the precipitation gauge, with the Antarctic double-shield at the lower height, had a similar characteristic to that of the DFIR in March.

The odd behavior of the DFIR shielded precipitation gauge (solid blue line) in March 2018.

The mystery was solved with some photos provided to us a few days before we arrived in McMurdo. It was also confirmed in our first field visit to the Willie Field APS site. We found the DFIR wind shield, installed at a height of 78 inches, to be nearly completely buried. I expected the Willie Field site to have an annual accumulation of 1-2 feet and now it was over 3 feet at the instrument tower and over six feet at the DFIR. In the end, we created our own snow drift and it resulted in a dome of snow centered around the DFIR and extending to our other instrument locations.

The buried DFIR shield and precipitation gauge at the Willie Field APS site in November 2018.

This also presented a big question as to what do we do now. Initially, there were naive thoughts that we would dig it out. The more we thought about it, and showed the photos around town, that idea quickly disappeared. In fact, the photos for the DFIR getting buried have been a big hit around town, including many senior support personnel. Thoughts were put into an operation that could remove a move a much larger quantity of snow. Along the way, we also agreed that we did not want to install the DFIR again for another year. The chances of it burying again, as well as all of the effort to install it again, were not appealing. Instead we came up with a modified experimental design for the site.

After discussions among various support work centers around town, the implementer for our project was able to put a plan together. We were going to get a bigger shovel to assist us. The plan was for a bulldozer to work during the night shift, Sunday evening after Thanksgiving, to clear the snow away from the outer ring and to push the snow away from the DFIR. The work that we had to do Sunday evening was to finish digging out the instrument cable and to provide instructions to the bulldozer operator.

The bulldozer clearing away the snow from the outer ring of the DFIR shield.

The finished work by the bulldozer removing the snow from the outer ring of the DFIR.

We returned the next morning and were greeted by an excavator to finish the work with us. The bulldozer and done a great job in clearing the snow away from the outer fence. We had to dig down at the joints for each outer section and unscrew each section and then carry each section to the truck. After doing that for all 16 outer sections, the excavator started doing the major work. The excavator progressively removed the snow that had accumulated between the outer and the inner DFIR shields. Once that was done, we had to remove the sections from the inner DFIR shield. Then the excavator had to remove the snow leading to another shield and finally the instrument. Fortunately, the excavator operator was amazing in his operation of the excavator and he was able to gently carve away at the snow with the giant bucket. It was a master piece of work. Eventually we reached the precipitation gauge at the center and removed all of the remaining hardware and the installation tower. The entire day with the excavator took six and a half hours, and that included our lunch break. It was amazing to have the assistance and support of the bulldozer and the much, much bigger shovel of the excavator. I don't know how else it would have gotten done.