Global Observing System Experiments within the Météo-France 4D-Var Data Assimilation System

Abstract

Observing System Experiments were undertaken within the 4D-Var data assimilation of the Météo-France global Numerical Weather Prediction (NWP) model. A six-month period was chosen (October 2019 - March 2020) where 40 millions of observations per day were assimilated. The importance of in-situ observations provided by aircraft, radiosondes and surface weather stations, despite their small fractional amount (7 %), has been confirmed particularly in the Northern Hemisphere. Moreover, the largest impact over Europe in terms of Root Mean Square Error (RMSE) scores comes from surface observations. Satellite data play a dominant role over tropical regions and the Southern Hemisphere. Microwave radiances have a more pronounced impact on the long range and on the humidity field than infrared radiances, despite being less numerous (10 % versus 80 %). Bending angles impact significantly the quality of the upper troposphere / lower stratosphere temperature of the tropics and Southern Hemisphere. Atmospheric Motion Vectors (AMVs) are beneficial in wind forecasts at low and high levels in the tropics and the Southern Hemisphere, but also in the humidity field. Such impacts are only significant during the first 48 hours of the forecasts. Scatterometer winds have an impact restricted to low levels which is kept at longer ranges. A comparison with Forecast Sensitivity - Observation Impact studies over a 3 month period using the same measure of short-range (24 h) forecast errors reveals that the ranking between the major observing systems is kept between these two ways of measuring observation impact in NWP. From our conclusions, recommendations are provided on possible evolutions of the global observing system for NWP.

A global observing‐system simulation experiment for the EPS–Sterna microwave constellation

“…(1999) and has already been used with the ARPEGE model, as presented in Chambon et al . (2023). This norm is computed using the difference between the forecast of the experiments and the truth of the nature run, and by summing this forecast error for four different variables: surface pressure, temperature, wind speed, and relative humidity, with normalisation factors specific to each variable.…”

“…This norm is based on the same norm as the one used in the forecast sensitivity to observations index (FSOi), which is widely used among meteorological centres (Eyre, 2021), with adaptations to use the nature run as the reference instead of an analysis. It is computed using the energy norm proposed by Ehrendorfer et al (1999) and has already been used with the ARPEGE model, as presented in Chambon et al (2023). This norm is computed using the difference between the forecast of the experiments and the truth of the nature run, and by summing this forecast error for four different variables: surface pressure, temperature, wind speed, and relative humidity, with normalisation factors specific to each variable.…”

Preparing the assimilation of the future MTG‐IRS sounder into the mesoscale numerical weather prediction AROME model

“…NWP centres often perform this type of experiment because of the need to test upgrades to operational systems and to conduct more systematic studies of the value of individual observing system components. Several studies have been conducted using OSEs, such as at ECMWF in the context of model upgrades (Kelly and Thépaut, 2007) and general observing system impact studies (Chambon et al ., 2022).…”

Impact of Varying Number of Radio Occultation Observations on Regional Weather Prediction Over India During the Summer Monsoon Season

“…An OSSE includes a "nature" run, or truth, generated by an NWP model; simulated observations produced using the appropriate forward operator; and data assimilation and forecast experiments to evaluate the impacts of the simulated observations using a different model. Numerous OSSE studies have shown the impacts of various observing strategies from terrestrial-based and space-based observing systems on weather prediction (Bormann et al, 2019;Bouttier & Kelly, 2001;Chambon et al, 2023;Cucurull et al, 2018;Errico et al, 2013Errico et al, , 2020Gelaro & Zhu, 2009;Jones et al, 2014;McCarty et al, 2021;Privé et al, 2021Privé et al, , 2022Singh et al, 2014;Wang et al, 2013). Harnisch et al (2013) used ensemble data assimilation techniques to evaluate the potential impacts of RO observation saturation and showed that RO saturation did not occur even with 128,000 soundings per day (corresponding to a global horizontal resolution of ∼63 km per day).…”