Blog Finalization

[sfriedli1]

• Integrate data from Bremgartenwald
• Setup workflow to publish the blog
• Automate Meteo Swiss data download (see Add option to update data for continuous deployment #5)
• Add analysis of ERA-5 Land data

[stineb]

Great, let's go ahead with these points.

• Regarding analysis of ERA-5 Land data, it would be nice to present maps of PCWD across Switzerland (or western-central Europe) for (i) the same date in previous drought years, and (ii) the maximum value per year

More suggestions to make it even richer:

• Additionally calculate PET based on Thorntwaite (temperature only, no net radiation). That way you can use a longer reference (more than back to just 2010) for stations providing only temperature data.
• The VHI map could be made a bit larger if computational and memory limits allow (don't have to store the raster file in the repo). Ideally, we could show the region @VanEend uses for his analysis -- includes Seeland and Biel ;-)

[fabern]

• clean TWD values (outliers?) and add to processing in PR Excluded sensor outliers with thresholds #10
• (Fabian) wait for ERA5Land download to finish, run through CWD_global analysis and update Add era5 land cwd ii #9

[stineb]

Nice work @geco-bern/geco_members

• Title change to 'Bern drought monitor'
• Move stuff that is not ours (and not Bern-related) to a separate section ('Appendix' or 'Further links') at the bottom. This concerns SwissSMEX and TreeNet. We don't want to step onto other's terrain and duplicate efforts by BAFU in collating these info on their site.
• Use descriptive names instead of P1, P2, P3. For example Low, Mid, High (height above nearest drainage channel)
• Avoid dashed lines in line plots. (Fabian: I suggest color for the years. Use consistent colors between TWD and SWP. Then, row/columns for other separation.)
• Show lines for all individual trees or shaded range, separated by plot (row) and species (columns).
• Use log scale for SWP (Fabian)
• Include known drought years to be highlighted in the Thorntwaite-based PCWD plot. May chose these five: 1947, 1976, 2003, 2018, 2026
• Remove manual section numbers.
• Find a solution to identify and remove outliers of TWD and SWP measurements.
• Optional: Include these ERA5-Land-based maps as suggested, perhaps under a section that makes clear that this goes beyond Bern.

With these changes and points above completed, feel free to publish and communicate about it (e.g. LinkedIn post) as soon as you are ready. I'll be back on this earliest 27.07. (after holidays) and could also communicate then.

[fabern]

TODO: Communicate about the monitor at https://geco-bern.github.io/bern_drought_monitor/:

• LinkedIn
• add a link (e.g. a new tab) to https://geco-group.org/
• github.com/geco-bern in the README

And importantly:

• ensure regular updates (manually, or automatic??)

[fabern]

NOTE: it appears the VHI fluctuates quite a lot, maybe clouds?
E.g. the situation on 2026-07-18 [1] appears to be worse than on 2026-07-20 [2] for the region of Bern, and inverse for Kt ZH.
This is why I chose to include the three last dates in the blog.

• But maybe we need more than three. If I update now the three dates of July 20th, 21st and 22nd all appear much less stressed than July 18th.

• Furthermore: add link to "Appendix D - National drought platform
  A bulletin on national level can also be found on https://www.trockenheit.admin.ch/
  Open current bulletin of National drought platform""

[1] See: https://map.geo.admin.ch/index.html?layers=WMS||vhi_v100_2026-07-18t235959||https://wms.geo.admin.ch/?item=2026-07-18t235959||ch.swisstopo.swisseo_vhi_v100||1.3.0

[2] See: https://map.geo.admin.ch/index.html?layers=WMS||vhi_v100_2026-07-20t235959||https://wms.geo.admin.ch/?item=2026-07-20t235959||ch.swisstopo.swisseo_vhi_v100||1.3.0