As an example, when adding future climate scenario files for Canadian stations from NRCan’s v1 2022 dataset, I realized that our current metadata in the tooltip doesn’t distinguish between individual files because the current metadata fields do not reflect the changes/iterations in future weather files. That means you see many, many dots but don't know how they differ.
Taking the Canada example, each station with future files has up to 32 variants covering every combination of:
- File type: TRY-AvgTemp, TRY-MinTemp, TRY-MaxTemp, TMY
- Global warming level: +0.5°C, +1.0°C, +1.5°C, +2.0°C, +2.5°C, +3.0°C, +3.5°C, plus a 1991–2021 baseline
Since the existing metadata columns (period, elevation, 99% Heating DB, etc.) are identical across all variants for a given station and reflect historical measurements, not the future scenario, it is impossible for a user to pick the right one.
Hence, before adding future weather files, we should think about how to extract the additional metadata and what additional fields we need to distinguish between variants. For Canadian files, it’s encoded in the filename, but for other countries/regions, it might be different.
We also need to decide how to adjust the UI. We could add a point for additional files (like now) or consider other options, such as a sub-menu or pop-up window for locations with multiple future weather files. If it's not too many files, the Dash Leaflet cluster do a pretty good job at breaking it down into subgroups and the tooltips (if showing additional metadata) could help picking the right file.
There are probably better solutions, but I wanted to raise this issue so we can start thinking about it, as it will be a challenge when adding future weather files soon.
As an example, when adding future climate scenario files for Canadian stations from NRCan’s v1 2022 dataset, I realized that our current metadata in the tooltip doesn’t distinguish between individual files because the current metadata fields do not reflect the changes/iterations in future weather files. That means you see many, many dots but don't know how they differ.
Taking the Canada example, each station with future files has up to 32 variants covering every combination of:
Since the existing metadata columns (period, elevation, 99% Heating DB, etc.) are identical across all variants for a given station and reflect historical measurements, not the future scenario, it is impossible for a user to pick the right one.
Hence, before adding future weather files, we should think about how to extract the additional metadata and what additional fields we need to distinguish between variants. For Canadian files, it’s encoded in the filename, but for other countries/regions, it might be different.
We also need to decide how to adjust the UI. We could add a point for additional files (like now) or consider other options, such as a sub-menu or pop-up window for locations with multiple future weather files. If it's not too many files, the Dash Leaflet cluster do a pretty good job at breaking it down into subgroups and the tooltips (if showing additional metadata) could help picking the right file.
There are probably better solutions, but I wanted to raise this issue so we can start thinking about it, as it will be a challenge when adding future weather files soon.