Accuracy of TAMDAR Sensors in Pre-Storm Environments

Atmospheric variables (pressure, temperature, dewpoint, wind speed, and wind direction) from the Great Lakes Fleet of Tropospheric Airborne Meteorological Data Reporting (TAMDAR) sensors were compared with 18 UTC and 00 UTC radiosonde (RAOB) observations in pre-convective environments. The selected cases occurred from May through August during 2007-2010 across the upper Midwest. For the purpose of this study, pre-convective environments had at least one severe thunderstorm report (tornado, hail greater than 1”, and/or wind damage) that occurred within 3 hours of a 00 UTC RAOB and within 50 nm of a airport where TAMDAR ascent soundings were available. In addition, TAMDAR observations were only used if they: 1) were at an airport where at least five TAMDAR flights took place between 18-00 UTC, 2) within 20 nm of a RAOB site, 3) had a complete temperature and moisture profile to at least 700 mb. Data from seven dates with multiple TAMDAR-RAOB comparisons were analyzed. In addition, a case study will be presented to illustrate the utility of TAMDAR data over the Minneapolis, MN area where tornadoes developed around 00 UTC 18 July 2010. Two TAMDAR soundings obtained at 18 UTC were compared to a special RAOB at the same time. Mean dew point differences of 3.3 C and -1.7 C, and mean temperature differences of -0.296 and 0.6 C, respectively, were found in the profiles extending to 650 mb. These results are representative of mean differences across the sample and indicate that TAMDAR sensors do in fact match relatively well with RAOB data. These results suggest that the higher temporal/spatial resolution of TAMDAR profiles can supplement RAOBs by providing critical short term observational information for forecasters between standard RAOB times and locations, leading to a better understanding of capping evolution.