The day started early, up before 5 a.m. so that Jeff and I could collect status reports from the various sites before a 6:30 a.m. briefing. Jet lag worked in my favor -- I'd been awake since 3:30 a.m. The bike ride to the Operations Center on the south side of Kwaj was really beautiful, with a last quarter moon riding high in the night sky above the constellation Orion in the east.

The Operations Center is fairly spartan, consisting of two office trailers joined by a roof. Aircraft flights are coordinated here, as well as the collection of the various kinds of data. One trailer contains offices, while the other contains the work space, decorated with folding tables, various charts, and the all-important cluster of computer workstations. The computers are important because KWAJEX continues the trend of assembling, visualizing, and assessing the data in the field, rather than waiting until after the experiment finishes. This allows the researchers to have a relatively good idea of data quality while it's still possible to gather further data, but it requires considerably more computer power than the field researcher of ten years ago imagined.

When Jeff and I got in, we immediately launched into phone calls to the sites away from Kwaj. The status reports gave us an idea of what we would later see in the data, such as loss of data above the 3,000-ft level in one of their soundings at Meck Island. It was also a chance to update housekeeping issues. Woja Island reported a conversation with one of the electrical generator specialists on Kwaj regarding the failure of their main generator. Each of the "outer islands" had been equipped with three generators for just such a problem, but it's still important to maintain the full complement. After all, these islands lack electricity, and thus can't power the weather balloon support equipment, computers, flood lights, refrigeration, air conditioning, and so on. On the positive side, Meck Island reported that some of their data failures seemed to be related to a loose computer cable.

In the usual routine, the 6:30 briefing prepares for an 8 a.m. planning meeting that reviews the previous day's activities, the current day's equipment status, and the weather forecast. Today, the weather forecast turned out to be sufficiently bland that the day was declared a "no-fly" day, implying that the aircraft were not needed. This is fortunate because the Citation and the Convair were both experiencing problems. On the other hand, both the forecast and a TRMM satellite overpass made it likely that there would be considerable interest in flying tomorrow afternoon. A subsequent meeting assessed the overall status of the experiment to date, and considered various upcoming issues. In particular, a lively debate ensued on the right strategy for flying data collection tracks through small convective clouds. Even though the aircraft aren't supposed to be launched for clouds under 10 km in width, a recent flight ended up only having small clouds to penetrate. Even after two weeks the various researchers are still learning the attributes of each other's equipment and priorities, so the discussion took longer than an outside observer might expect.

Well before 11 a.m. the scheduled activities were over, although we were left with a massive pile of details that I needed to learn in order to take over Jeff's job . . . tomorrow!

George Huffman