The representation of winds in the lower troposphere in ECMWF forecasts and reanalyses during the EUREC4A field campaign

Abstract

Abstract. The characterization of systematic forecast errors in lower-tropospheric winds over the ocean is a primary need for reforming models. Winds are among the drivers of convection, thus an accurate representation of winds is essential for better convective parameterizations. We focus on the temporal variability and vertical distribution of lower-tropospheric wind biases in operational medium-range weather forecasts and ERA5 reanalyses produced with the Integrated Forecasting System (IFS) of the European Centre for Medium-Range Weather Forecasts (ECMWF). Thanks to several sensitivity experiments and an unprecedented wealth of measurements from the 2020 EUREC4A field campaign, we show that the wind bias varies greatly from day to day, resulting in RSME's up to 2.5 m s−1, with a mean wind speed bias up to −1 m s−1 near and above the trade-inversion in the forecasts and up to −0.5 m s−1 in reanalyses. The modeled zonal and meridional wind exhibit a too strong diurnal cycle, leading to a weak wind speed bias everywhere up to 5 km during daytime, turning into a too strong wind speed bias below 2 km at nighttime. The biases are fairly insensitive to the assimilation of sondes and likely related to remote convection and large scale pressure gradients. Convective momentum transport acts to distribute biases throughout the lowest 1.5 km, whereas at higher levels, other unresolved or dynamical tendencies play a role in setting the bias. Below 1 km, modelled friction due to unresolved physical processes appears too strong, but is (partially) compensated by dynamical tendencies, making this a challenging coupled problem.

Fast Atmospheric Response to a Cold Oceanic Mesoscale Patch in the North‐Western Tropical Atlantic

“…In particular, it seems that the numerical model of the reanalysis underestimates the change in stability over the cold water, producing a too weak reduction of vertical mixing. This can be ascribed to the difficulties that the ABL numerical scheme has in correctly reproducing the lower atmospheric mixing, which is likely to be responsible for the observed wind biases (Savazzi et al., 2022). Different ABL numerical schemes produce different mixing, resulting in different air‐sea coupling (Perlin et al., 2014).…”

“…Even if Savazzi et al. (2022) showed that the assimilation of RS data is negligible on the wind field, the reanalysis fields should not be considered as independent data but as a physically based spatio‐temporal interpolation of the observations. Reanalysis data can be used to estimate the large‐scale variability signal to isolate the local one better.…”

The frictional layer in the observed momentum budget of the trades

“…{7}^{\circ } $$ and at 20 height levels between 100 and 2,000 m, with one scan roughly every 30 s. The radial velocities were corrected for ship motions using an accompanying GPS system, described in Savazzi et al . (2021). After motion correction, the wind vector is retrieved and hourly averages are used to study the composite diurnal cycle.…”

“…Operational high‐resolution (9 km) forecasts from the IFS as well as ERA5 reanalysis are used to assess the synoptic situation during EUREC$$$ {}^4 $$$A/ATOMIC and evaluate the pressure gradient force and advection terms derived from the circular dropsonde arrays. For the forecast, model output was extracted at the nearest four neighbours of 61 points placed concentrically around the center of the EUREC$$$ {}^4 $$$A/ATOMIC circle, matching the flight hours on flight days (Savazzi et al ., 2021). ERA5 data are extracted for an area 65°–25°W and 0°–30°N encompassing Barbados.…”

“…Considering the small magnitude of the 𝑣− wind compared to the 𝑢− wind (Figure2a,b) 𝑣 is proportionally large and opposes Ekman wind turning near the surface. Efficient vertical transport might explain the relatively weak wind turning in the lower atmosphere (Figure2f).The most prominent differences between the observations and the IFS are observed in i) the 𝑢 tendencies in the lowest 1 km, with differences maximizing near cloud base (∼ 700 m), ii) in F 𝑢 above 1.5 km, iii) in the frictional forces in 𝑣 below cloud base, and iv) in the large-scale dynamical tendency of 𝑣 in the cloud layer between 700 m -3 km Savazzi et al (2021). investigates these differences in detail and shows that they are consistent with biases in the wind profile.…”