Midwinter Suppression of the Pacific Storm Track Activity as Seen in Aircraft Observations

Abstract

Unassimilated observational data, from the years 1979 to 1993, are analyzed and compared to the NCEP?NCAR reanalysis data to confirm the magnitude of the seasonal cycle in Pacific storm track activity. Such a comparison is necessary since recent publications have indicated that the reanalysis data may contain biases and spurious jumps that may affect climate signals, and eddy variance/covariance statistics are particularly sensitive to model biases, as well as changes in data coverage and quality. High-pass-filtered variance in the 250-hPa meridional wind component is taken to be the indicator of upper-tropospheric storm track intensity. Rawinsonde observations located mainly over the land masses, and aircraft observations over the oceans, both display seasonal cycles very similar to that seen in the reanalysis data. Both the midwinter suppression in Pacific storm track activity, as well as the more recent finding that the Pacific storm track is significantly stronger during the late 1980s and early 1990s than during the early 1980s, are confirmed. Complications involved in computing variance statistics from raw aircraft observations, as well as preliminary comparisons between aircraft and rawinsonde observations, are discussed in the appendixes.