Examination of High AMSR-2 Water Vapor Values in Tropical Cyclones

Since 2012, Remote Sensing Systems (RSS) has used brightness temperatures from the Advanced Microwave Scanning Radiometer 2 (AMSR-2) to measure global over-ocean geophysical variables with high temporal resolution and large spatial coverage. The most recent RSS AMSR-2 product suite (V8.2) represents the latest iteration of the following geophysical retrievals: sea-surface temperature, near-surface wind speed, columnar cloud liquid water, sea-surface rain rate, and total columnar water vapor (TCWV). An important update in the new version is the elimination of an upper boundary for valid TCWV (previously set to 75 mm in earlier versions). The new V8.2 TCWV retrievals occasionally reach values as high as 100 mm in intense tropical cyclones (TCs). In this work, we examine the validity of these high TCWV satellite retrievals in TCs by comparing them to in-situ datasets and hurricane models. In particular, we aim to understand whether these extreme values of TCWV are realistic, or rather due to an algorithm limitation in extreme rain.

We will present a comparison study in which we validate AMSR-2 TCWV in sample TCs with in situ observations from dropsondes deployed in hurricane cores during NOAA Gulfstream-IV flights, and with surface-based GPS stations. In previous studies, RSS has validated water vapor in non-TC conditions with GPS stations (TCWV < 75 mm); this is the first study that validates the high TCWV ( > 75 mm) in TCs.

In addition, we will compare the AMSR-2 TCWV validation results to numerical models such as the Hurricane Research Weather and Forecasting (HWRF) and the National Center for Environmental Prediction (NCEP) Global Data Assimilation (GDAS) models. We will end the presentation with a robust statistical analysis performed for all datasets in order to determine the overall accuracy of the high AMSR-2 TCWV values in TCs as compared to in-situ and numerical model estimates.

The aim of this study is to provide the forecasting community with the confidence needed for satellite observations of TCWV to be used in assimilation/validation of numerical weather prediction (NWP) models and precipitation forecasting of intense TCs.