Technical Updates / Information
20230919 Revision of the webpage. ETH hres products are switched of by DYNAMOS at ETH. We run extended-range weather regime forcasts, uncalibrated for the new operationsl 100 member IFS extended-range ensemble since its preoperational release in March 2023. In addition we currently try to run the Lagrangian medium-range WCB ensemble forecasts. Other browsing tools can be used to look into the archives.20230302: Ongoing server outage. Migration to ECMWF. E-suite daily 100member extended-range implemented. Medium-range and operational extended-range still manually.
20221127: since Oct 2022 limited availability due to server outage at KIT. I try to process 00UTC medium-range forecasts once per day.
20210201: extended-range now with 10d low-pass filtered data not 5d.
20190126 Major Updates:
Additional plots and refinements in the regime definition:
- Persistence now required for active WR life cycle attribution in plot "Weather regimes". Effective from initial time 20190126_00. For earlier initial times "Weather regimes" equals what we now call "WR Simple LC". We experiment with 4 day persistence as indicated in the plot.
- "WR Simple LC": instantaneous regime attribution if maximum IWR above 1.0 . For initial times before 20190126_00 this is plot "Weather regimes".
- "WR Members": Matrix plot showing the temporal evolution of active WR life cycles in each member. WR attribution requiring persistence.
- Plots of surface variables as mean of all members attributed to a WR. Before 20190126_00 this was based on instantaneous WR attribution. From 20190126_00 this is based on attribution including persistence (attribution as in "Weather regimes").
- "Z500 mean/std" now also based on attribution including persistence.
- WR products for bi-weekly ECMWF extended-range (S2S) ensemble forecast out to day 45: "Weather Regimes", "WR index", "WR maxIWR", "WR Simple LC", "WR Members". Here Z500 anomalies are 5 day low-pass-filtered prior for computing the projection. The sophisticated WR life cycle definition as in Grams et al. (2017) is used for WR attribution and includes merging of subsequent life cycles and a 5 day persistence.
The anomaly of 500hPa geopotential height (Z500) (unfiltered data) relative to a 90d running mean is computed for each member,
high resolution (HR) and control forecast (CF), and the ensemble mean (EM). The anomaly is normalized and
projected on the cluster mean anomaly of the year-round seven regimes of Grams et al. (2017).
The projection is normalized following Michel and Rivière (2011) to compute a weather regime index IWR.
A member/CF/HR/EM is attributed to a regime if IWR is above 1.0
Short description of available plots (medium-range):
- (a) "Weather regimes": relative number of ensemble members (including CF) attributed to a regime at each forecast step. The regime has the maximum IWR at that time and the time is within a period of IWR above 1.0 for that regime. Length "pers" of the period indicated in plot (currently pers=96h).
- (b) "WR Index": distribution of IWR in the ensemble at each forecast step.
- (c) "WR Index ana.": IWR for the unfiltered analysis during the last 365 days.
- (d) "WR Simple LC": instantaneous regime attribution. The regime has the maximum IWR at that time and IWR is above 1.0, no persistence.
- (e) "WR maxIWR": instantaneous attribution to a regime as in (d) but IWR must only be the maximum, not necesessarily above 1.0.
- (f) "WR Members": Matrix plot showing the temporal evolution of active WR life cycles in each member. WR attribution requiring persistence and IWR above 1.0 (attribution as in "Weather regimes").
- (g) "Z500 clim" 90 day climatological running mean of Z500 centered at initial time.
- (h) "Z500 mean/std"mean and standard devition of Z500 for all members attributed to the respective regime. ("Weather regimes" attribution).
- (i) "T2M, T2M Ano, W100M, W100M Ano, TCC, PRECIP": mean of 2m temperature, 2m temperature anomaly, 100m wind, 100m wind anomaly, total cloud cover, 6h precipitation for all members attributed to the respective regime ("Weather regimes" attribution). 2m temperature available for a "North Atlantic" and "Europe" domain
Former IFS Weather Regime Forecasts (for browsing, operational from about 2018-2022)
- twice-daily
medium-range WR products (0-15 days) (Browser window width optimally >1200px)
- bi-weekly extended-range WR
products (0-45 days) (Browser window width optimally >1200px)
- daily 100 member E-suite (non-official) extended-range WR
products (0-45 days) (Browser window width optimally >1200px)
- detailed documentation (update 01/2019)
Michel, C., and G. Rivière, 2011: The Link between Rossby Wave Breakings and Weather Regime Transitions. J. Atmospheric Sci., 68, 1730–1748, doi:10.1175/2011JAS3635.1.
Grams, C. M., R. Beerli, S. Pfenninger, I. Staffell, and H. Wernli, 2017: Balancing Europe/’s wind-power output through spatial deployment informed by weather regimes. Nat. Clim. Change, 7, 557–562, doi:10.1038/nclimate3338.
Grams, C. M., L. Magnusson, and E. Madonna, 2018: An atmospheric dynamics' perspective on the amplification and propagation of forecast error in numerical weather prediction models: A case study. Quart. J. Roy. Meteor. Soc., 144, 2577–2591, doi:10.1002/qj.3353.