The CEMS-flood sub-seasonal and seasonal products are essentially the same across the two systems and similarly the same across EFAS and GloFAS. Below, the available products and their main features are introduced.
Product colouring
The forecast signal is shown by colouring on the map product layers, either for individual river pixels or larger basins (see the further product details below in the subsequent sections). Each of these river pixels or basins are coloured by the expected anomaly category and the uncertainty sub-category defined for the actual forecast. There are altogether 7 anomaly categories (from 'Extreme low' to 'Extreme high') and 3 uncertainty sub-categories (from 'Low to 'High'), based on the extremity level of the 51 ensemble forecast members in the 100-value climatological distribution and the mean and standard deviation of these 51 rank values. The details of the computation methodology is described in: Placeholder CEMS-flood sub-seasonal and seasonal forecast anomaly and uncertainty computation methodology.
In total, there are 15 forecast signal categories coloured on the maps. Out of the possible combinations of the 7 anomalies and 3 uncertainties (7*3), 5*3=15 category combinations are represented by colours, after the middle three anomaly categories ('Bit below', 'Normal' and 'Bit high') are merged into one 'Near normal' category. The choice of 5 anomaly categories for the colouring allows the users to focus on the larger anomalies, supplemented by the 3-level uncertainty representation. The 7 base and 5 simplified anomaly categories, the 3 uncertainty categories and the related 15 colours are shown in Figure 1.
Each of the 5 anomaly categories indicated on the maps have a distinct colour ranging from red ('Extreme low') to blue ('Extreme high'), with the 'Near normal' category indicated by the neutral grey colour, while the 3 uncertainty sub-categories are indicated by different intensities of the same colour, going from darker to lighter versions as the uncertainty increases.
Figure 1. List of the anomaly and uncertainty categories defined with the colours used on the map products.
River network map
The river network summary map layer shows the combined expected forecast anomaly and uncertainty signal in a simplified way for each forecast lead time (Figure 1). The lead time is weekly (always Monday to Sunday with the weekly average river discharge) in the sub-seasonal and monthly in the seasonal (always calendar month with the monthly average river discharge).
The users can navigate between the different lead times by clicking on the chosen weeks (in sub-seasonal) and month (in seasonal) in the lead time controller (see Figure 1a bottom left corner). This way the users can check the individual signal for each lead time, which currently is 5 or 6 weeks for the sub-seasonal (depending on which day of the week the forecast run) and always 7 months for the seasonal.
The forecast signal is shown by colouring of all river pixels above a certain minimum catchment area (currently 50 km2 in EFAS and 250 km2 in GloFAS). The 15 pre-defined anomaly and uncertainty category combinations are used with the 15 colours, as described in the previous section (see e.g. Figure 1).
Figure 2 shows an example, which highlights some river sections with the explanation of the assigned colours and the corresponding anomaly and uncertainty levels. Both Figure 2a and 2b include the colour legend with the 15 categories and the corresponding colours (same as in Figure 1), for easier interpretation.
The river network summary map also contains the reporting points, which are labelled in Figure 2b. These are river locations, where detailed information is provided about the evolution of the forecast signal over the forecast horizon. These reporting points are either fixed points, which are also used in the medium-range flood products and the basin-representative points, which are selected locations on a one point per basin basis. Further details about the basins and the representative points are available here: Placeholder CEMS-flood sub-seasonal and seasonal basins and representative stations.
a)
| b)
|
Figure 2. Example snapshots of the sub-seasonal and seasonal river network summary maps with the reporting points, lead time navigation and river pixel colours explained.
Reporting point pop-up window
At the predefined reporting point locations (either fixed or basin-representative) further detailed information is provided about the evolution of the forecast signal.
Point information section
The first table in the popup window ('Point information') provides metadata information of the station (Figure 2). These are the station ID and also an internal ID (Point ID), the station name (if available), country, basin and river names, and also coordinates (lat/lon) and upstream area in two flavours, the provided ones and the LISFLOOD river network equivalent. The provided coordinates and upstream area are from the users as those represent the real river gauge locations. These are available only for the fixed points (sometimes provided upstream area is missing). For the basin-representative points, however, only the LISFLOOD coordinates and upstream area are available, as these points were defined solely on the simulated LISFLOOD river network. The fixed reporting points have a Point ID in the metadata table starting with 'SI', while the basin-representative points starting with 'SR'.
Hydrograph section
Next item in the popup window is the hydrograph, which graphically summarises the climatological, antecedent and forecast conditions (see Figure 2 and Figure 3).
Figure 2. Example snapshot of the reporting point pop-up window product (for a seasonal forecast).
The left half of the plot, left of the horizontal dotted line, which indicates the forecast start date, shows the past (see Figure 3a). The black dots (connected by black line) indicate the so-called water balance, the proxi observations, which are produced as a LISFLOOD simulation forced with either gridded meteorological observations in EFAS, or ERA5 meteorological reanalysis fields in GloFAS. These black dots show the simulated reality of the river discharge conditions, as close as the simulations can go at the actual conditions over the forecast periods (average river discharge over months in seasonal and calendar weeks in sub-seasonal).
These black dots are added to the hydrographs retrospectively, after each week (in sub-seasonal) or month (in seasonal) passes and the weekly or monthly mean proxi-observed river discharge values become available. The users are encouraged to go back and check previous forecasts to see how well the earlier forecasts predicted the anomalies.
The right half of the plot covers the forecast horizon, in the displayed example in Figure 2 and 3 this means 7 lead times of 7 calendar months from August to February (next year) (see Figure 2a). The forecast distribution is indicated by box-and-whiskers, displaying the minimum and maximum values in the ensemble forecasts of all the 51 members and the lower and upper quartiles (which are the 25th and 75th percentiles as well) and the median (which is the 50th percentile).
The coloured background is the model climatology (see Figure 3b). This climatology is generated using reforecasts over a 20-year period. Further information on the climatologies and their generation is given here: Placeholder CEMS-flood sub-seasonal and seasonal forecast signal generation methodology. In the past half of the hydrograph, the climatology is always lead time 1, so first week (always as days 1-7) or first month (whichever month of the year it is), as that is the closest equivalent to the proxi-observation-based climatology. While in the forecast half, the climatologies are lead time dependent and for each forecast lead time the equivalent climatology is plotted with that specific lead time. From the climatology, the 5 anomaly categories are coloured, below the 10th the 'Extreme low' with red, above the 90th percentile the 'Extreme high' with blue, the 10th to 25th percentiles zone as 'Low' with orange, the 75th to 90th percentiles as 'High' with cyan and finally the remaining 25th to 75th percentile as 'Near normal' with grey. This 'Near normal' category is the extended one (it was mentioned also with the river network summary map above), by merging the original 25-40, 40-60 and 60-75 percentile categories, including the narrower 'Near normal', the 'Bit low' and 'Bit high' categories.
As Figure 3c highlights it, the seasonal hydrograph (it would not work for the sub-seasonal due to the much shorter weekly lead times) indicates a property of the model climatologies. The seasonal hydrograph is designed to have exactly 13 monthly periods, which guarantees that the last month of the forecast (February in Figure 2 and 3) will feature as a month-7 forecast climatology and also as a month-1 forecast climatology in the past period as the oldest period plotted. This way, the hydrograph gives a visual impression of the drift in the river discharge forecasts. Drift in this context means, the month-7 reforecast-based climatology percentiles could occasionally be very different to the month-1 reforecast-based percentiles and by this show a noticeable shift or drift in the forecast behaviour. This means, values going lower or higher from shorter to longer ranges (see Figure 3c for visual indication of this). In fact, for this comparison, the left-most and right-most parts of the hydrograph need contrasting. In the example in Figure 2-3, the shaded climatological categories highlight that 'Extreme low' and 'Low' categories shift only very little, with the median being very stable. However, the higher percentiles (75th and 90th) are noticeably higher in the month-7 climatology, indicating a noticeable drift for larger values in the reforecasts. While in the month-1 average river discharge climatological distribution, the 90th percentile is about 20 m3/s, so about 10% of the time the monthly mean can exceed this value, in the longer range month-7 reforecasts the same 90th percentile, the 10% of the time to exceed this value, increases to 25 m3/s. So, in sort, the forecast is more likely to show larger values in the longer ranges than in the short range.
a)
| b)
| c)
|
Figure 3. Different interpretation schemes (a-b-c) for the sub-seasonal and seasonal hydrographs.
Probability evolution table section
Finally, the last part of the reporting point popup window is the probability evolution table. This table shows all the 7 anomaly categories (extreme dry to extreme wet as left to right) and the related probabilities for all the forecast lead time periods and from all the previous forecast runs that verify during the most recent forecast horizon. For the sub-seasonal, this means 5 or 6 calendar weekly forecast lead time periods, depending on which day of the week the run date is, and thus how many calendar weeks the 45-day lead time in the forecast can cover, and 7 calendar monthly periods for the seasonal. For the seasonal forecasts, there is always 7 rows with the most recent 7 seasonal forecast probabilities (as Figure 2 shows). While for the sub-seasonal, including all the daily (00 UTC) forecast runs verifying in the forecast horizon, there can be 41 to 46 rows. Always as many, as many forecasts can verify in the forecast horizon of the actual real time forecast, and it again depends on which day of the week the run date is. The bottom right corner of the probability evolution table is empty, as those lead times are not available from the earlier forecast runs.
The cells in the table are not coloured, with one exception, which is the forecast anomaly category (of the whole ensemble). That cell's number is bold and the cell is coloured by the same colour that the river pixel has on the river network summary map. As described in Placeholder CEMS-flood sub-seasonal and seasonal forecast anomaly and uncertainty computation methodology, the dominant category is determined by the rank-mean (the mean of the ensemble members' ranks). The coloured cell's number is not always the highest, like in many forecasts in the example in Figure 2' probability table. For example, the forecast for August shows a gradual progression from near normal (grey colour, the original 'Bit low' of the 7 categories) and high uncertainty (lightest grey colour of the three), to 'Extreme low' category with low uncertainty (darkest red colour). Moreover, the number of the coloured cells is also on the increase generally, as we go towards the shorter lead times. Until the June forecast, the colours are the lightest of the three versions, highlighting high uncertainty (light orange in the June forecast), while in the July forecast the uncertainty drops to medium level (medium dark orange) and finally in the August forecast we arrive to the low uncertainty 'Extreme low' situation. However, at the same time, for all of these forecasts the more likely of the 7 categories are constantly the 'Extreme low' category, with the probability values gradually increasing from 2X to 100% in the August forecast (all of the 51 ensemble members being in the 'Extreme low' category). The reason why the earlier forecasts shift towards the normal conditions in the mean sense, is the larger uncertainty with most or all of the categories having some ensemble members.
Basin summary map
The basin summary map is the equivalent of the river network summary map averaged onto the larger basin scale. The basins are predefined, as described in Placeholder CEMS-flood sub-seasonal and seasonal basins and representative stations, with 204 basins in the EFAS domain and 944 basins globally in GloFAS.
Figure 4 shows an example snapshot in the same area as Figure 1a above. One can compare, how the variable signal on the river network averages into the basin signal. The averaging is done from the river pixel information. A Rank-mean and a Rank-std value will be calculated for each basin and with those the exact same method is used (as for the river pixels on the river network summary map) to define the forecast anomaly category and the uncertainty category and thus the colour of the basin (see the inset figure with the colour legend in Figure 4). The colours used are the same as for the river network summary map, except there is a slight shift in transparency, in order to allow visibility for all river pixels, even for those with the same colour as the basin colour.
The basin Rank-mean and Rank-std values are determined using all the large enough river pixels in the basin. Currently, only pixels above 50km2 are used in EFAS and pixels above 250 km2 are used in GloFAS in the averaging. The basin Rank-mean/Rank-std values are calculated as an arithmetic average of the Rank-mean and Rank-std values of the individual river pixels, weighted by the square of the upstream area, as described below:
Figure 4. Example snapshot of the sub-seasonal and seasonal river network summary maps with the animation panel and river pixel colours explained.
The forecasts can be advanced (or even animated if needed) with the lead time controller, both on the river network and the basin summary layers (see Figure 1a and Figure 4 bottom left corner) and the users can check the individual signal for each lead time, which currently is 5 or 6 weeks for the sub-seasonal (depending on which day of the week the run date is) and always 7 months for the seasonal.