Observing Complementary Variables on Multidecadal Time Scales

For several decades, Remote Sensing Systems has been providing long-term, climate quality records of a number of geophysical retrievals estimated from microwave satellite sensors. These records now cover the last 36 to 45 years. Constructing these datasets requires ongoing effort to intercalibrate results from a long chain of overlapping satellite missions. Two of these variables, lower tropospheric temperature (TLT) from microwave sounders and total column water vapor (TCWV; over the ocean), are closely coupled by the Clausius-Clapeyron relationship. This relationship enables an important mutual validation analysis of TLT and TCWV on decadal time scales. For both variables, it is not feasible to validate decadal-scale stability by comparing with in situ measurements, i.e., radiosonde profiles, because of inhomogeneity problems in the radiosonde record and the sparse spatial sampling of long-term radiosonde sites. By evaluating the changes in both temperature and water vapor, we can determine if the two datasets are internally consistent.

The Clausius-Clapeyron relationship indicates that a given change in temperature will result in a fractional change in water vapor, assuming no change in average relative humidity. The exact value of the fractional change in total column water vapor for a given change in the temperature of a deep-layer satellite temperature retrieval is difficult to calculate analytically because it depends on many details, such as the scale height of the vapor column and the average degree of saturation. We arrive at an expected value over the tropical oceans by considering both climate model output and seasonal scale fluctuations in the satellite retrievals. For the latitude band from 20°S to 20°N, results from phases 5 and 6 of the Coupled Model Intercomparison Project (CMIP5 and CMIP6) are closely clustered around a value of 6.35%/K, which is consistent with the short time scale (monthly to interannual) behavior of satellite observations. On decadal time scales, a mystery emerges: satellite water vapor suggests more warming than is measured by microwave sounders over the last three decades.

In our presentation, we will briefly discuss the history and challenges of intercalibrating satellite retrievals, covering results from both our team and from researchers at other institutions. For the temperature retrievals, our understanding of the errors caused by choices made during the intercalibration and merging process has been greatly enhanced by the involvement of multiple teams using different approaches. The implications and possible causes of the inconsistency between the temperature and vapor datasets will be considered. We will conclude our presentation with a discussion of the challenges associated with continuing this critically important work – which informs our knowledge of global-scale changes in Earth’s climate – given limited support for maintenance of long-term datasets.