Dataset Submission: HakaiQU5MooringProvisional about hakai-datasets

[JessyBarrette]

I created a metadata record for this dataset:
https://cioos-siooc.github.io/metadata-entry-form/#/en/pacific/tV5qE0aUgaOjSVmgPgiZ6MyHuSy1/-MW0Az3HzggNmnqgHDw3

[raytula]

QU5M seems like primarily a 'internal' name that may not mean much to people not associated with Hakai. I wonder if it could be generalized, or at least include 'Quadra' somewhere in the title.

Good to see the accuracy info in the description. Maybe we should mention how the data has only been automatically QCd and flagged, and that the data is not appropriate for research purposes --- or is it, within the stated accuracy limit? I was imagining we would produce a research ready version of this database, but not 100% certain if that makes sense. If we imagine adding conductivity and O2 measurements in the future, then I guess 'provisional' and 'not for research purposes' maybe more important to highlight.

[JessyBarrette]

yeah I agree. I'll add more info regarding location and auto QCing method. Regardless, it still good to have. It may not be necessary for now to have research dataset but yes with the CTD-Divers we will likely get a research dataset which would have the corrected conductivity temperature data.

Technically we could potentially also add a correction for those tidbit sensors to increase their accuracy to 0.05C

[JessyBarrette]

@raytula I just updated the metadata form

[raytula]

Looks good.

[JessyBarrette]

We need to integrate the air temperature to this dataset and we should be good to!

[JessyBarrette]

The development dataset is now available at https://goose.hakai.org/erddap/tabledap/HakaiQU5MMooringProvisional.html

Since those tidbit platforms are sampling at a 10min interval, I wonder if we should just drop the statistic variables since they are generally extracting stats from only one variable.

[raytula]

The development dataset is now available at https://goose.hakai.org/erddap/tabledap/HakaiQU5MMooringProvisional.html

Since those tidbit platforms are sampling at a 10min interval, I wonder if we should just drop the statistic variables since they are generally extracting stats from only one variable.

Yes, it makes sense to me to drop the statistic variables.

Note: in the case of stations like the Pruth Dock, KC Buoy and Limpet, the data logger typically takes a measurement more frequently (e.g. every 10 seconds???) and stores the five minute average, min, max and std values. However, there are also situation where the data logger only takes a measurement every 5, 10 or 60 minutes and measurements like this should likely be handled the same as these tidbit measurements.

[JessyBarrette]

Right it makes sense, looks like there's some interpolation happening on the 5min table at least for the tidbits. Just clarify things for me is it just a linear interpolation?

…

On Mon, Jun 21, 2021 at 6:40 AM Ray Brunsting ***@***.***> wrote: The development dataset is now available at https://goose.hakai.org/erddap/tabledap/HakaiQU5MMooringProvisional.html Since those tidbit platforms are sampling at a 10min interval, I wonder if we should just drop the statistic variables since they are generally extracting stats from only one variable. Yes, it makes sense to me to drop the statistic variables. *Note:* in the case of stations like the Pruth Dock, KC Buoy and Limpet, the data logger typically takes a measurement more frequently (e.g. every 10 seconds???) and stores the five minute average, min, max and std values. However, there are also situation where the data logger only takes a measurement every 5, 10 or 60 minutes and measurements like this should likely be handled the same as these tidbit measurements. — You are receiving this because you were assigned. Reply to this email directly, view it on GitHub <#18 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHICYOMMVH6G3IP4BFD7GKTTT46MNANCNFSM4YPXBCXQ> .

-- *Jessy Barrette M.Sc.* *Marine Instrumentation Specialist* Hakai Institute <https://www.hakai.org/> | ***@***.*** | (C) [phone]

[raytula]

Right it makes sense, looks like there's some interpolation happening on the 5min table at least for the tidbits. Just clarify things for me is it just a linear interpolation?

Yes, the tool use to import the raw tidbit data just does linear interpolation for 'short' data gaps.

[JessyBarrette]

This dataset was merged to production erddap