Information on the accuracy and representativeness of new observing systems is critical for their optimal use in NWP data assimilation systems. In an attempt to provide these data for the WVSS-II moisture sensing instruments being currently deployed on commercial aircraft in the US, a series of objective study have been conducted both to assess the accuracy of the humidity data and to help determine how best to use these data as a supplement to or surrogate for traditional upper-air reports.

Two assessments were made during June 2005 and November 2006 using facilities provided by the UW mobile observing system located at Louisville, KY where about 80% of the WVSS-II equipped planes land and/or take-off daily. A total of between 25 and 30 UPS B757 aircraft provided en-route and profile reports data for the assessments.

Intercomparison data sets included 1) an infrared AERI system (providing very high time frequency boundary layer T/q profiles), 2) surface GPS (providing total atmospheric moisture content), 3) standard surface observations (Temp, Wind, Moisture and Ceiling), 4) a portable GPS rawinsonde system, and 5) geostationary and polar satellite data.

Approximately 2 weeks of collocated radiosonde and aircraft data were be collected during each test period with radiosonde launched 3 times nightly, immediately before, between and after periods of multiple aircraft arrivals/departures. On average, more than 5 co-locations (within 1 hour and 50 km) were obtained daily, including Temperature, Wind Direction / Speed, and Humidity reports.

The WVSS-II assessment focused primarily on assessments of the accuracy of water vapor measurement taken throughout the troposphere during November 2006. In addition to testing the engineering aspects of the WVSS-II systems, statistical evaluations were made of the performance of a variety of factors important for the optimal objective use of the aircraft data in combination with other data sources by assessing: 1) Similarity of reports from the different observing systems and different aircraft, 2) Biases between ascent and descent reports from individual aircraft, 3) Variability between different aircraft (to assess instrument calibration and effects of aging), 4) Capabilities of the systems to capture sharp moisture gradients accurately, including after an aircraft emerges from clouds, and 5) Comparison with previous test results.

Results showed generally excellent agreement, especially in the lower troposphere, where the RMS fits approached 0.5 g/kg for mixing ratio and at places bettered 5% for relative humidity. Additional details of the assessment results will be presented, along with a discussion of future program plans both for instrument improvement and for expanded WVSS-II system deployment.