Stage and discharge measurements from Water Survey of Canada station 08LF051, Thompson River near Spences Bridge, British Columbia, used in the examples: every discharge measurement made at the station from 1994 to 2024, in date order.
Format
A tibble with 93 rows and 5 columns:
- date
Date of the measurement.
- stage
Stage, in meters (the mean gauge height for the visit).
- discharge
Discharge, in cubic meters per second.
- uncertainty_pct
Reported uncertainty of the discharge, as a percentage of it at two standard deviations (the "IVE method, 2-sigma value"), or
NAwhere none was given.- rating_table
The Water Survey's rating table in force at the gauging, such as
"11", orNAwhere none was recorded (all gaugings before 2020).
Details
Reported uncertainties are available for only 19 of the gaugings, so
uncertainty_pct is mostly NA. It is a percentage of the discharge at
two standard deviations, so the standard uncertainty of a discharge, in
cubic meters per second, is uncertainty_pct / 100 * discharge / 2.
Fits using var_spec() need a variance for every gauging, and so need
either the subset with an uncertainty or variances from elsewhere.
The rating at the station has shifted over the decades: gaugings from
different periods depart from a single fitted curve by several percent in
opposite directions, and the Water Survey's rating table changed three
times between 2020 and 2024. A curve fitted to all 93 gaugings is fine for
illustration, but a rating for a given period should use gaugings from a
period of stable channel conditions; see vignette("fitting").
One uncertainty may be incorrect: the gauging of 2018-11-01 has 0.0226,
where the others lie between about 2.5 and 11. It may have been entered as
a fraction (2.26%) rather than a percentage. It is kept as it appears in
the records, but leave it out of fits using var_spec(): weighted by it,
it would carry over 99.9% of the total weight, and the curve would be
forced through it.
Getting the data yourself
Individual discharge measurements (field visits) are not published on the Water Survey's Water Level and Flow website, https://wateroffice.ec.gc.ca, which offers only the time series derived from them; its Frequently Asked Questions list field-visit observations among the data "typically not publicly available on this website". These records came from the Water Survey of Canada, and further or more recent measurements would have to be obtained from it directly. The station's page on the website does report its most recent discharge measurement.