As a user, I want to ensure values from the USGS Daily Values service are using the correct times of day

[james-d-brown]

Given an evaluation that uses the USGS DV service
When timestamps are added to the reported dates
Then I want to be sure they are the correct timestamps

[james-d-brown]

Conversation today leads me to believe that I misunderstood how USGS records daily values.

I thought they were recorded at midnight in local time, which led to all sorts of shenanigans in accounting for daylight saving based on the metadata field uses_daylight_savings, as well as the option for a user to override this through declaration, daylight_savings, because it meant the values were recorded at a fixed local time, i.e., midnight local time.

The documentation is unfortunately silent and/or impossible to find, so I thought uses_daylight_savings was declarative meaning that "uses_daylight_savings": "Y", should be interpreted as the values having timestamps that are always in midnight in local time. However, I now believe they are always recorded at a fixed time in local standard time. In other words, if the metadata field "time_zone_abbreviation": "EST",, then the values are always recorded at midnight in EST, so always midnight UTC-5hrs!

Will need to try and corroborate this because, like I said, the docs are very vague indeed and in different places, but if so this will be a bug fix (that probably has no practical consequence because, to ensure the data forms part of an evaluation during local summer time, users would've had to use the uses_daylight_savings to ignore the daylight savings, but it would make this completely redundant and simplify the code a lot too, as well as the user experience).