Update on JCSDA Impact of Observing Systems Project

The capability to assess observation impact is key to JCSDA’s partner agencies for evaluating the extent to which changes to the global observing system impact the skill of numerical forecast models and the Agencies’ downstream mission services that are supported by those forecasts. The JCSDA Impact of Observing Systems is articulated into four main tasks:

1. Conduct routine Forecast Sensitivity to Observations Impact (FSOI) experiments across JCSDA’s partner agencies to detect in near real-time changes to their forecasting systems or to global observing systems,

2. Ensure that the generated results are scientifically sound and reliable,

3. Devise automated and adaptive algorithms for processing satellite data,

4. Evaluate new platforms.

This talk provides an update on progress with an emphasis on 1). In March 2019, JCSDA has set-up a near real-time inter-comparison experiments, in which FSOI diagnostic files are pulled from NASA/GMAO and NRL daily runs and displayed through a web interface (ios.jcsda.org). Diagnostics (total impact, fractional impact, etc.) can be visualized individually for each center, as well as on comparative plots. We have now several months of data and interesting features can already be noted. If radiosondes remain the observing system with the largest impact for both centers, the impact of satellites can differ somewhat. For example, the fractional impact of IASI observations is four times larger at NASA/GMAO than NRL. On the other hand, the fractional impact of SSMI/S observations on NRL forecasts is five time larger than on NASA/GMAO forecasts. This can be expected given the marine nature of SSMI/S observations. The UK Met-Office, and possibly the US Air Force, will soon be joining the experiment and FSOI diagnostics inter-comparisons between the four centers will be presented.