Satellite-based Diagnostics of Tropical Extent

Abstract:

Long-term satellite measurements of variables relevant to tropical extent offer the advantage of high-frequency near-global coverage, but may not be accurate enough to accurately estimate the changes in the width of the tropics over time due to problems with inhomogeneity and calibration drift. The focus of this work is to develop novel diagnostics that can be used to estimate the width of the tropics, but are less sensitive to long-term changes in instrument calibration. For example, daily zonal means of tropospheric temperature from MSU and AMSU can be examined to find the north/south extent of measurements within a certain value of the maximum value for that day. This diagnostic would be completely insensitive to overall errors in satellite calibration. Another idea is to investigate ratios of maps of total column water and lower tropospheric temperature. Using the temperature maps, estimates of the total column water vapor assuming complete saturation of the atmospheric column are assembled. These can be compared to the measured total column water vapor and areas where the difference does not exceed a threshold value (columns are close to saturation) are judged to be regions of atmospheric ascent, and thereby part of the tropics. Regions where the differences exceed this value are judged to be descending, and thus part of the subtropics. We will present preliminary results of both these approaches, as well as validation of the techniques using model results where variables such as vertical velocity are available for comparison.