Earliest Dam Year

First load the network, and attributes

network load_file("data/ohio-river/ohio.network")
network gis.load_attrs("data/ohio-river/nid-uniq.gpkg", "nidId")

nodes.is_usgs = NAME match "^[0-9]+";
nodes.is_dam = !is_usgs;
network count(nodes.is_usgs)
network count(nodes.is_dam)

Results:

1806

4181

Counting the upstream dams again,

nodes<inp>.ndam = int(is_dam) + sum(inputs.ndam);
nodes.no_dam_us = ndam == 0;
network count(nodes {no_dam_us & is_usgs})
network count(nodes {no_dam_us & is_usgs}) / count(nodes.is_usgs)

Results:

605

0.33499446290143964

We can see 33% of the USGS gages do not have dams upstream.

And for those that do, let’s look at the construction year,

env.max_year = 9999; # todo test it with nan
nodes.dam_year = int(get_attr("yearCompleted", env.max_year));

nodes<inp>.dam_aff_yr = min_num(inputs.dam_aff_yr, dam_year);
nodes<inp>.dam_affected = dam_aff_yr < env.max_year;

network count(nodes.dam_affected & nodes.is_usgs)
network count(!nodes.no_dam_us & nodes.is_usgs)
network count(nodes.dam_affected & nodes.is_usgs) / count(!nodes.no_dam_us & nodes.is_usgs)

Results:

1152

1201

0.9592006661115737

This shows we have the construction year for 95% of the USGS gages that were constructed after at least one upstream dams.