Observational Constraints on the Water Vapor Feedback Using GPS Radio Occultations

The air refractive index at L-band frequencies depends on the air's density and water vapor content. Exploiting these relationships, we derive a theoretical model to infer the specific humidity response to surface temperature variations, dq/dTs, given knowledge of how the air refractive index and temperature vary with surface temperature. We validate this model using 1.2-1.6 GHz Global Positioning System Radio Occultation (GPS RO) observations from 2007 to 2010 at 250 hPa, where the water vapor feedback on surface warming is strongest. Current research indicates that GPS RO data sets can capture the amount of water vapor in very dry and very moist air more efficiently than other observing platforms, possibly suggesting larger water vapor feedback than previously known. Inter-comparing the dq/dTs among different data sets will provide us with additional constraints on the water vapor feedback. The dq/dTs estimation from GPS RO observations shows excellent agreement with previously published results and the responses estimated using Atmospheric Infrared Sounder (AIRS) and NASA's Modern-Era Retrospective Analysis for Research and Applications (MERRA) data sets. In particular, the GPS RO-derived dq/dTs is larger by 6% than that estimated using the AIRS data set. This agrees with past evidence that AIRS may be dry-biased in the upper troposphere. Compared to the MERRA estimations, the GPS RO-derived dq/dTs is 10% smaller, also agreeing with previous results that show that MERRA may have a wet bias in the upper troposphere. Because of their high sensitivity to fractional changes in water vapor, and their inherent long-term accuracy, current and future GPS RO observations show great promise in monitoring climate feedbacks and their trends.