Thomas Rackow 🧊

@trackow.bsky.social

832 Followers 495 Following

Scientist @ecmwf.int : climate & ocean variability, kilometre-scale modelling, and its visualisation. #art and music enthusiast. #scicomm

@fraser
Propagation characteristics of MJO convection and circulations composited from (a) observations, (b) IFS 9 km simulation with NEMO, and (c) IFS 4.4 km simulation with FESOM. Time–longitude diagrams of lagged-composite intraseasonal OLR (shading) and 850 hPa westerly wind anomalies (contours) averaged over 10° N–10° S. Contour interval is 0.5 m s−1, with zero contours omitted. Stippling in (b) and (c) denotes statistical significance of OLR anomalies at the 90 % level (all shading in (a) satisfies this significance). The number of detected MJO cases is denoted at the top of the figures together with analysis periods. Green lines indicate the longitudinal range over the Maritime Continent, and black, blue, and red lines indicate the centre of MJO convective envelopes for the observations, for the 9 km simulation, and for the 4.4 km simulation, respectively.
Diurnal cycle of land surface temperature (LST) difference between the city of Warsaw and its rural surroundings, for (a) the summer months (JJA) during clear-sky conditions (5-year mean). The IFS 4.4 km simulations are given with red lines (Cycle 2 dashed, Cycle 3 solid), and observations from LSA SAF are given in black. The bottom panels show JJA-mean clear-sky LST anomaly maps at 13:00 local time, with respect to the surrounding rural LST average, for (b) observations from LSA SAF, (c) IFS 4.4 km Cycle 3 with the urban scheme, and (d) IFS 4.4 km in Cycle 2 without the urban scheme and using older land use–land cover maps.
Imprint of simulated Arctic sea ice leads on 2 m temperature in the Laptev Sea and East Siberian Sea. (a) Sea ice concentration field and (b) 2 m temperature field. The panels depict 13 February 2020, at 08:00, in the IFS-FESOM Cycle 3 simulation with TCo2559 (4.4 km), coupled to the NG5 ocean (∼ 4–5 km resolution in this area). The black and white contour lines represent the border between land and ocean areas.
Eddy variability expressed as the standard deviation of daily SSH data. The left column shows data from the Cycle 3 simulation of IFS-FESOM model, while the right column shows AVISO multi-satellite altimeter data. In AVISO, the global mean SSH is removed from each grid point. AVISO data consist of the time period 2017–2021, while 2020–2024 is used for IFS-FESOM.
Snapshot for 31 January 2020 at 21:00 UTC of infrared brightness temperature (left) and hourly precipitation rate (right) over the Indian Ocean, from (a) observations (Meteosat 8 SEVIRI channel 9 and GPM IMERG) and on forecast day 12 of (b) IFS-NEMO 4.4 km and (c) 9 km (third row) simulations. The simulations use the nextGEMS Cycle 3 setup except that they are run with a satellite image simulator and, for technical reasons, are coupled to NEMO V3.4 (ORCA025) here.
Global-mean TOA imbalance as a function of global-mean surface air temperature (GSAT). Grey lines show the climatological range of individual CERES years (2001–2020) over ERA5 GSAT data (Hersbach et al., 2020). Thin lines are tracing monthly mean values, with a small square marking the final month for each simulation. Big squares depict annual means (dashed for Cycle 2, solid for Cycle 3), and for multi-year simulations, thick solid lines are tracing annual means for each year, with the big square marking the last simulated annual mean.
Supplement to Goessling, Rackow, & Jung; Science, https://doi.org/10.1126/science.adq7280
Credit: Alfred-Wegener-Institut / Yves Nowak
Figure 1 from Goessling, Rackow, & Jung; Science, https://doi.org/10.1126/science.adq7280
Global-mean 2m temperature evolution from daily 10-day weather forecasts under conditions representing pre-industrial (1955 as proxy), present-day (2023), and future +2.9K climate states (2049), for AIFS, GraphCast, and Pangu-Weather. All models were trained with ERA5 reanalysis from 1979-2017 (with eventual fine-tuning), and the global-mean temperature for 1979–2020 is shown (light grey lines) for context. Black dashed lines are global-mean temperatures for 1955 (ERA5), 2023 (operational analysis), and for 2049 (physics-based scenario simulation), representing initial conditions for the data-driven forecast models. For every day in the three chosen years, 10-day forecasts were performed with every model starting from 12:00 UTC, resulting in 9 forecast datasets. The global-mean temperature of these 9 forecast datasets are shown as thin (hair-like) colored lines.
Skill of data-driven 10-day weather forecasts for 2m temperature in conditions representing pre- industrial, present-day, and future +2.9K climate states, for AIFS, GraphCast, and Pangu-Weather. Root-mean square error (RMSE, in K) for all models, weighted by the cosine of latitude in a, 1955, a proxy-year for pre-industrial times, b, in 2023, representing present-day climate, and c, in 2049, the +2.9K warmer world. All RMSE evolutions in 1955, 2023, and 2049 are averaged over the daily forecasts.