A Climate Benchmark of Upper Air Temperature Observations from GNSS Radio Occultation

GPS (Global Positioning System), or more generally Global Navigation Satellite System (GNSS), radio occultation (RO) is a remote sensing technique that produces highly accurate temperature in the upper troposphere and lower stratosphere across the globe with fine vertical resolution. Its fundamental measurement is the time delay of the microwave signal as it travels from a GNSS satellite to the receiver in low Earth orbit. With a relatively simple physical retrieval, the uncertainty in the derived temperature can be traced rigorously through the retrieval chain back to the raw measurements. The high absolute accuracy of RO allows these observations to be assimilated without bias correction in numerical weather prediction models and provides an anchor for assimilating other types of observations. The high accuracy, coupled with long-term stability, makes RO valuable in detecting decadal temperature trends. In this presentation, we will summarize the current state of RO observations and show temperature trends derived from 15 years of RO data in the upper troposphere and lower stratosphere. We will discuss our recent efforts in developing retrieval algorithms that are more tailored towards climate applications. Despite the relatively robust "self-calibrating" nature of RO observations, disparity in receiver hardware and software may introduce subtle differences that need to be carefully addressed. While the historic RO data record came from relatively homogeneous hardware based largely on NASA/JPL design (e.g., CHAMP and COSMIC), the future data will likely be comprised of a diverse set of observations from Europe, China, and various commercial data providers. In addition, the use of non-GPS navigation systems will become more prevalent. We will discuss the challenges involved in establishing a long-term RO climate data record from a suite of research and operational weather satellites with changes in instrumentation and coverage.