The end of waterway map

General talk
, , , ,
26

I’ve added some new features! :gift:

Firstly, the method of allocating flow downstream has been improved. Most of the flow follows the way with the same name. This reduces the number of cases of “half the danube is being allocated to this random field”. :rofl: (for more see osm-lump-ways documentation).

Secondly, I now calculate the name (& wikidata & wikipedia) tag value of the last way that flow into the end point. This is a rough way to see the total upstream of every river in the world, by name! This is the “Ends stats CSV“ in osm-lump-ways, see the csv file format. It includes the rank of the end point by total upstream. I’m storing the top 1000 largest rivers into an ever growing CSV file. I ran it for every year since 2006.

The CSV is data.waterwaymap.org/waterwaymap.org_ends_stats.csv.zst.

You can access the CSV on the terminal like this:

$ curl -s data.waterwaymap.org/waterwaymap.org_ends_stats.csv.zst | zstdcat | ...

BTW qsv is a cool CLI CSV tool.

You can do things like:

  • № 1 largest river (by accomuilated upstream legnth) is the Обь (Ob) in Siberia. curl -s data.waterwaymap.org/waterwaymap.org_ends_stats.csv.zst | zstdcat | qsv search -s upstream_m_rank "^1$" | qsv table
  • The longest river in Ireland is the Shannon. With 5,476 km of total upstream, it’s currently the 324th longest river (:exploding_head:). By fixing wrongly connected river segments, I’ve been bringing it up, the leaderboard. I can check my status
    curl -s data.waterwaymap.org/waterwaymap.org_ends_stats.csv.zst | zstdcat | qsv search -s name ".*Shannon.*" | qsv table, and see the upstream_m_rank increase. I can check the next in the leaderboard with curl -s data.waterwaymap.org/waterwaymap.org_ends_stats.csv.zst | zstdcat | grep -C 1 ".*Shannon.*" | tail

Download it and make something cool. :slightly_smiling_face:

🪻ⓐⓜⓐⓝⓓⓐ🪻

7 Likes