Two new ways to improve reanalysis: Use future observations to improve the analyses and forecasts, and minimize reanalysis "jumps" that result when new observing systems are introduced

Eugenia Kalnay

  25 June, Noon, in 2155


Ensemble Forecst Sensitivity to Observations (EFSO) can use future observations to identify whether each observation in the current analysis is beneficial or detrimental. Chen and Kalnay MWR, 2019a, and 2019b (under review) used Proactive Quality Control (PQC), to find and delete the observations found to be most detrimental, which can significantly improve the forecasts.

Since Reanalysis, by definition, knows future observations, EFSO/PQC can be used to delete detrimental observations, and improve future reanalyses. A second new application is to minimize the "jumps" (for example, in precipitation) that take place when a new observing system is introduced, due to the fact that the model climatology is different from the true climatology. We suggest that these "jumps" can be minimized by correcting the reanalysis climatology before the introduction of the new observing system (Zhou, 2014, DRUM).