The json in question:
https://data.geo.admin.ch/ch.bafu.hydroweb-messstationen_temperatur/ch.bafu.hydroweb-messstationen_temperatur_de.json

-> to get flussmessstationen dynamically, as they are projected in an maplibre-incompatible epsg out-of-the-box
-> We need to project them into EPSG:4326 (WGS84)

Promising leads:

• https://ogre.adc4gis.com/

Fix the paths when importing files, as apparently the gh-pages page can't find its resources.

see handballStats as well, as this issue came up there as well.

should be fixable though, once one path is adjusted to work correctly (handballStats do work locally with brackets, as well as a gh-pages page.

• Check papers for useful interpolation algorithms
  -->probably temperature-based (as is most significant parameter)
  -->see MeteoSwiss Paper (but very likely to be computationally too demanding)
  -->method must take into account height differences
  -->best choice (for now): Ordinary Kriging (using PyKrige from GeoStat Framework)

• Alternative approach:
  -->subtract a temperature difference according to ICAO standard atmospheric profile (-0.6°C / 100m height difference)
  -->likely to require a separate interpolation for temperature (less efficient!?)

Discussion (long-term considerations):

• less dependence from web-based API (by using GDAL for reprojection) ?
• local storage + fetching of files through python (up to discussion)
  --> Badewetter-Index files: fetched, stored + reprojected in python (for now - gdal snippet available).

Check licences of dependencies for compatibility

Once #5 is complete, adjust the VT-Style according to these design specifications.

Use https://maputnik.github.io/editor to alter the style file inside this repository.

*.overpassql is now the recommended file extension for Overpass QL: OSM Wiki: Overpass QL § File extension
(Disclaimer: I added that section after some discussion.)

I propose to rename all *.query files in this repo to *.overpassql for consistency.

Context: GitHub's syntax highlighting engine requires that 200 repositories use the same file extension before they consider supporting that syntax and file extension: github-linguist/linguist#5890

Licences to check before "going live":

• Licence of this repository
• MapLibre licence
• OGD swisstopo licence
• Other Dependencies

Once all these check out, create github pages page and "go live" with it

Find out how to leverage a Raspberry Pi to collect data, calculate the index and push changes to the GitHub Repo

Use geojson format to add a custom polygon and a custom raster to the map.

• Include windchill + humidex
• Thinking of useful additional data (DEM, Geopotential, ...)
  --> keep in mind: should not get to computationally intense

can be done with qgis built-in tool:

--> choose directory, not mbtiles. then use a local path with /{z}/{x}/{y}.png

--> establish which resolution should be used (is needed) for BWI

Originally posted by @taetscher in #10 (reply in thread)

This Layer needs some improvements.

Specifically it would be nice if we had a QGIS workflow automated that:

• Gets the latest OSM data via Overpass Query (should be doable from within QGIS)
• Creates a class Attribute and populates the attribute (needed to distinguish icons when drawing the map)
• Dumps all non-essential Attributes (keeps only the Class and Name attributes)
• Saves the layer as a geojson with correct EPSG

The JS part also could do with a few improvements (only after a class attribute was established)

• no popups on POIs without a name attribute

To give the user suggestions where to go swimming, download and convert OSM data to geojson (point features):

• everything swimming related
• everything beach related

use overpass turbo to get the data and QGIS to convert it all (e.g. polygons) to point features.
http://overpass-turbo.eu/#

asdf

Establish the graphical design of the webpage (colors, accented stuff etc.)