Overview
The following Explore the AI/ML sub-seasonal forecast portal displays to view all sub-seasonal forecasts submitted to the AI Weather Quest. As a reminder, participants Participants are challenged to submit with providing quintile-based probabilistic forecasts of for near-surface air temperature, mean sea level pressure, and accumulated rainfall at a precipitation, targeting lead times of days 19 to 25 , (week 3) and 26 to 32 lead time (week 4). Use the guide below to navigate the portal and better understand its functionalities.
Latency of forecast display
Participants have four days to submit their forecasts (days 1 to 4). On day 5, all submitted forecasts are extracted and pushed to the forecast portal. In other words, forecasts initialised on a Thursday will not be viewable on the forecast portal until the succeeding Monday.
Navigating the home page
On the home page of the AI Weather Quest forecast portal, you will be able to see each product uploaded to the forecast portal. A product, characterised The homepage displays all forecast products submitted by participants.Each product corresponds to a specific variable from an individual model and is represented by a large thumbnail on the right, is defined for each variable submitted by an individual model.For instance, say I have a model called bestS2S and I submit forecasts for each of the if the model ‘bestS2S’ submits forecasts for three variables, then three separate products would will appear.
On Use the left you can -hand panel to filter products by the chosen parameter, team name and individual models.:
- Parameters– The forecast variables
- Teams– The teams’ names
- Models – The models’ names
Navigating individual forecast products
When clicking Clicking on a forecast product , you can view opens all forecasts submitted to the AI Weather Quest for that individual the corresponding model and variable combination. The title of the product is formatted as 'Each product is titled using the format:
teamname: modelname forecast variable quintile probabilities', for
For example: 'bestS2S: Model 1Mean sea level pressure quintile probabilities'..
Below each forecast visualisation, a brief description provides additional context about the forecast shown.
In the left-hand On the left selection panel, you can filter refine the shown displayed map by selecting:
- Base time
...
- – The Thursday forecast initialisation date
...
- Valid time
...
- – The forecast lead time (either days 19 to 25 or 26 to
...
- 32)
...
- Area
...
- – The geographic region shown on the map
- Quintile interval – The forecasted quintile probability bands (< 20%, 20 <= x < 40%, 40 <= x < 60%, 60 <= x < 80%, x > 80%).
Additionally, Next to the left of each selection filter, you will see an overview button the overview icon (four cyan blocks) . If you click on the overview button, it will open all allows you to display all corresponding forecasts within that selection filter simultaneously. For example, if you click on clicking the overview button icon next to the quintile interval selection, then forecasts “Quintile interval” will show maps for each of the quintile intervals will be displayedquintile range.
Once choosing your chosen selection, the right map will display the forecast.
There are several tools that you may chose to leverage on the map itself:
...
In addition, the forecast map includes several interactive features:
- Arrows–Switch between available lead times (weeks 3 and 4).
- Colorbar – Click the icon with horizontal lines (bottom right) to see the colour legend.
- Fullscreen – Expand the map with the enlarge icon (bottom right)
- Share – Use the top-right share icon to generate a link to the current view.
- Download – Click the box-arrow icon to download the forecast as .png or .pdf, or generate a URL to download the charts programmatically.
...
Finally, below the forecast visualisation you will find a brief description of the forecast shown.
Acknowledgements
...