All charts correspond to weekly means.
All charts correspond to weekly means.
The weekly mean anomaly charts display the anomaly between the forecast weekly mean and the corresponding weekly mean in the SUBS-M-Climate.
Charts available are: 2m temperature, surface temperature, total precipitation, mean sea level pressure, winds. Click on the central small icon in the bottom right of the web frame to show the colour scale of values appropriate to each display.
Colouring on the charts is implies >90% significance for the anomaly from SUBS-M-Climate. White shading on the maps only implies <90% significance of the anomaly and show where the ensemble forecast is not significantly different from the sub-seasonal range climatology (SUBS-M-Climate), according to a Wilcoxon-Mann-Whitney (WMW) test. It does not necessarily point to a forecast of average nor to a 'don't know' forecast of climatology.
Further information can be gained by inspection of the sub-seasonal anomaly meteogram. Click the mouse over any location on the ensemble charts to produce probability information in meteogram form.
The example (Fig8.2.3-1) shows white shading on the maps. When presented with white shading at the location of interest, the user should inspect the anomaly Meteograms and CDFs to decide the likely anomaly and assess the confidence.
Fig8.2.3-1: Consider T2m temperature forecast anomalies from model climate at the location in southern Sweden (shown at X).
For week 1, the meteogram shows a high confidence forecast of low anomalies (box and whisker symbol quite short centred near 0ºC) implying near-normal mean 2m temperatures. This is confirmed by the relatively vertical red CDF profile near the 0ºC anomaly line.
For week 5, the meteogram shows much less certainty in the anomalies (box and whisker symbol is extended) implying a larger spread of mean 2m temperature anomalies and mean 2m temperatures. The CDF profile suggests the forecast is much like climatology as the red line of the forecast anomalies is almost coincident with the black line of the model climate distribution of anomalies. Forecast and climatology are equally uncertain.
In both cases in this example there is white shading on the maps at location X. The example shows white shading on the maps cannot, of itself, point to a forecast of average (as at week 1) nor to a 'don't know' forecast of climatology (as at week 5).
The weekly anomaly probability charts display the probability that the weekly mean anomalies are greater than zero. The probability is calculated from the number of ensemble members which show an anomaly within the higher half of the the SUBS-M-Climate distribution.
Charts available are: 2m temperature, surface temperature, total precipitation, mean sea level pressure.
Two equally probable domains can be defined: below normal and above normal.
White on the plot can currently signify a number of different things - e.g. median of the forecast is about the same as the median of the re-forecasts. This fact that there are several potential interpretations can be confusing for users and ECMWF plans to offer improved products to address this later in 2024.
Note: On precipitation charts the range of colours available for "below average" anomalies (brown colours) is constrained by the local climatology in the SUBS-M-Climate. For example if all ENS members showed a dry week, the mean (-ve) anomaly could be no larger in magnitude than the mean in the SUBS-M-Climate. So in some locations the strongest dry signal one can ever see will only be in the first or the second of the brown shades.
Click the mouse over any location on the ensemble charts to produce probability information in diagrammatic form. Diagrams available:
Click on the central small icon in the bottom right of the web frame to show the colour scale of values appropriate to each display.
The weekly anomaly probability charts display the probability that the weekly mean anomalies are in the lower or uppermost third (tercile), fifth (quintile) or tenth (decile) of the SUBS-M-Climate distribution.
Charts available are: 2m temperature, surface temperature, total precipitation, mean sea level pressure.
The probability is calculated from the number of ensemble members which show an anomaly within the highest or lowest tercile, quintile or decile of the the SUBS-M-Climate.
Terciles: Three equally probable domains can be defined: below normal, normal and above normal.
Quintiles: Five equally probable domains can be defined:
Deciles: Five equally probable domains can be defined.
Click the mouse over any location on the ensemble charts to produce probability information in diagrammatic form. Diagrams available:
Click on the central small icon in the bottom right of the web frame to show the colour scale of values appropriate to each display.
Any pixel that has high probability for lower (or upper) decile should have at least the same or higher probability for lower (upper) quintile and even higher for lower (upper) tercile. On occasion this has not happened and is being investigated.
These multi-parameter charts charts display the ensemble weekly mean:
Click the mouse over any location on the ensemble charts to produce probability information in diagrammatic form. Diagrams available:
Click on the central small icon in the bottom right of the web frame to show the colour scale of values appropriate to each display.
Hovmöller or Time-Longitudes diagrams show the time evolution of the ensemble mean anomaly of a parameter. The x-axis represents the longitude, the y-axis represents the time evolution (time increasing downwards). Past results lie above the horizontal line and forecast results lie below.
The northern mid-latitude Hovmöller diagrams show the time evolution of the ensemble mean anomaly of geopotential height at 500hPa or 1000hPa, averaged over the latitude band 35N-60N (Northern extra-tropics) or 25S-50S (Southern extra-tropics). The anomaly has been computed by averaging all the members of the real-time forecast and subtracting the mean of the SUBS-M-Climate. Contours are plotted every 1.5dam. Since it is an ensemble mean, the structures shown below the horizontal line are much more detailed in the first days of the forecast (top part) than in the last days (bottom part). Shaded areas represent the ensemble spread and are displayed only when the amplitude of the anomaly exceeds 2dam. On average spread will naturally increase with forecast lead time, though occasionally, when moving to longer lead times, there can be a reduction.
The Madden-Julian Oscillation (MJO) Hovmöller diagrams show the ensemble mean anomalies of outgoing long wave radiation, zonal wind at 850hPa, and velocity potential at 200hPa averaged over a tropical band (15N-15S).
The mean flow and anomaly charts show, at global or regional scales, the weekly:
Click the mouse over any location on the ensemble charts to produce probability information in diagrammatic form. Diagrams available:
Click on the central small icon in the bottom right of the web frame to show the colour scale of values appropriate to each display.
Any pixel that has high probability for lower (or upper) decile should have at least the same or higher probability for lower (upper) quintile and even higher for the tercile.
The sub-seasonal range products should always be used with historical skill metrics in mind.
Broad indications of ensemble performance and predictability in the sub-seasonal range should be taken into account when considering the sub-seasonal range output. Users should consult verification information:
The Relative Operating Characteristics give a measure of the effectiveness of a forecast system to predict an event that actually happens balanced against forecast of an event that fails to occur. ROC is derived from measuring the area beneath the results plotted on a ROC diagram. ROC area values indicate:
The Reliability diagrams give a measure of the tendency of the forecast system to over- or under-forecast and event. The diagram plots the frequency of a forecast probability of an event against the frequency that the event occurs. Ideally these should match and is shown by the diagonal line. Where the plot lies:
An explanation of reliability diagrams and ROC diagrams is given in the annex to this guide.
The ROCmap shows the ROC score computed over each grid point with a resolution of 2.5 degrees longitude by 2.5 degrees latitude. The charts are anomalies derived from previous sub-seasonal range ensemble forecasts compared with SUBS-M-Climate.
The map colours show values of the ROC score shaded according to the scale above the map:
The RPSS map shows the RPSS score computed over each grid point with a resolution of 2.5 degrees longitude by 2.5 degrees latitude.
The Ranked Probability Skill Score (RPSS) compares the Ranked Probability Score of a probabilistic forecast system with a reference of some sort, usually climatology.
The Ranked Probability Skill Score charts give a measure of the sub-seasonal range ensemble forecasts and is a measure of how good forecasts are in matching observed outcomes. Forecasts falling into the upper or lower terciles of SUBS-M-Climate, are verified against the subsequent reanalysis or operational analysis (for precipitation, 24hr forecasts). This is equivalent to the Brier skill score where upper and lower terciles are used.
The map colours show values of the RPSS score (equivalent to the Brier skill score) shaded according to the scale above the map:
(FUG Associated with Cy49r1)