A Comparison of WVSS-II and NWS Radiosonde Temperature and Moisture Data

Measurement of atmospheric water vapor by commercial aircraft began in the late 1990s, in a cooperative venture by United Parcel Service (UPS) Airlines and the National Weather Service (NWS). A sensor employing a thin film capacitor became known as the Water Vapor Sensing System (WVSS). It was installed on six UPS aircraft between 1997 and 1999. A two week study was conducted by the NWS in Louisville, Kentucky in the fall of 1999 that compared WVSS data from UPS aircraft with radiosondes launched from a nearby mobile sounding unit. A comparison was also made in the second half of 1999 of these six aircraft when they were near NWS radiosondes around the 00 and 12UTC sounding times. These studies showed that WVSS data to be comparable to NWS radiosondes. The availability of sixteen WVSS aircraft in the spring of 2001 prompted a comparison of this new data source with NWS radiosondes over a several month period. Data was accessed from the Forecast Systems Laboratory's (FSL) Aircraft Data web page and was compared to NWS radiosonde data that is also available at that site. Nearly 1100 data comparisons were made at various mandatory sounding levels from 925 to 250 mb in the three month period from May through July 2001. The two data sources compared reasonably well, especially below 500mb. Air temperatures at 925, 850, and 700mb differ by an average of 0.97 degrees C at the three levels, with dewpoints differing between 1.31C and 1.74C. While the WVSS supplied data of reasonable quality, they required maintenance at intervals unacceptable to the airline. In response to this problem, and in an attempt to get water vapor data of superior quality, a new sensor called the Water Vapor Sensing System II (WVSS-II) was developed. Instead of a thin film capacitor to measure relative humidity, it uses a more accurate technology employing a laser diode. It had been installed on 20 UPS aircraft by the spring of 2005. In order to determine how the WVSS-II compared to WVSS-I, a study was conducted that was nearly identical to the WVSS-I to radiosonde comparison. In the spring and summer of 2005, the authors collected WVSS-II soundings when they were available within 50km of an NWS radisonde site, and within one hour either side of the morning and evening launch times. Preliminary data suggest that the WVSS-II compares far more favorably to NWS radiosondes than it's predecessor.