Spatial time periods in geological time

[CathJex]

Many of our articles cover time periods in deeper geological time, and a calendar input wouldn't work for us for the majority of our articles. E.g. 25 Ma or 10000 BP. Are there any plans to offer an alternative way of setting temporal coverage? And is there any plan for including temporal resolution of data reported in the article? E.g. A time period of 125000 to 110000 years ago, at 10–20 yr sample resolution.

[nuest]

We do currently not have named temporal intervals on the road map, but I see your point! Would these periods all be connected to a specific numeric period so that we can visualise/relate them?

Can you please clarify what you mean with "sample resolution"? Do you mean some kind of uncertainty? Or do you mean that data is available at a specific interval?

[nuest]

Just for my understanding, here are some links to lists of historical time periods:

• https://en.wikipedia.org/wiki/History_by_period
• https://www.totallytimelines.com/historical-time-periods/
• https://historyplex.com/historical-time-periods

[CathJex]

We do currently not have named temporal intervals on the road map, but I see your point! Would these periods all be connected to a specific numeric period so that we can visualise/relate them?

Just to clarify, I'm referring to geological time not historical periods.

I guess that there are two things here:

1. The time when fieldwork was carried out or the data were collected/analysed, which would always be a calendar input.
2. The time period covered by the data presented in the paper, which I presume is what the calendar input is intended for. But maybe I misunderstood?

Concerning point 2, for our journal, there are some cases where a calendar input would be work, e.g. fields like hydrology and modern climatology where they collect data on things happening in the present day or recent past. E.g. monitoring river water chemistry over a 2 week period.

But other fields, glaciology, paleoclimatology and geology, would present data representative of much deeper, geological timescales, i.e. microfossils in lake sediments from thousands of years ago, chemistry of ice cores from hundreds of thousands of years ago or samples of rocks from millions of years ago, etc. For these fields of study, the calendar input doesn't work. Instead, you could use defined geological time periods, known as eons, eras, periods, epochs, ages (see link to chart below). But these definitions do change over time, and so it might be safer to just ask for the age in years, and give the option for selecting the time units in ka, Ma, Ga etc. As that won't ever change.

Standard geological time is updated regularly in the International Chronostratigraphic Chart:
https://stratigraphy.org/chart

PDF version of the latest one:
ChronostratChart2022-02.pdf

The British Geological Survey have a 'friendly' break down: https://www.bgs.ac.uk/discovering-geology/fossils-and-geological-time/geological-timechart/

If there is no plan to implement something like this now, then for our journal, it would help if the calendar input was not a required field so that we can skip it when needed or even deactivate it in the plugin settings perhaps.

[CathJex]

Can you please clarify what you mean with "sample resolution"? Do you mean some kind of uncertainty? Or do you mean that data is available at a specific interval?

To clarify what I meant by sample resolution: I'm referring to the number of samples taken over a given period of time. E.g. a hydrologist monitors river chemistry between 2 Jan 2022 to 22 Jan 2022 (that would be the temporal coverage input into the calendar field). But the other interesting piece of information is how often they took a sample, i.e. the sample resolution. E.g., one sample a day, 1 sample an hour or 1 sample every 10 minutes. It's also useful information for people searching for publications/data.

In geology, an example could be an ice core, which dates to 100 000 years ago at the bottom of the core, and 10 000 years ago at the top of the core. The scientists took a sample of ice every 1 cm, which represents 1000 years of ice formation. So in the published paper, their data covers 100 000 to 10 000 years ago (often written as 100–10 ka) at a sample resolution of 1000 years.

Does that make sense? Perhaps it is irrelevant for this plugin, at least at this stage. But maybe something to think about in the future?

[nuest]

Thank you very much for your time to explain the issue to such a great level of detail and with all needed context. Some replies from my end.

• The calendar input at the moment should not be a required field, so that's a ✔️
• So the geological time is essentially a controlled vocabulary of names for time periods, right?
• Are you aware of any search interfaces where one can select geological time as input, or any metadata standards that model them?
  I'm wondering if these would be best saved in a database by their name, or by actually putting in the numbers.

---

• Notes to self:
  • in the database, one could store the name of the period plus the version/date of the standard geological time
  • is the current standard geological time available in some machine readable format? Yes, see JSON files in https://github.com/CSIRO-enviro-informatics/interactive-geological-timescale/tree/master/src/assets
  • https://stratigraphy.org/timescale/ based on https://github.com/CSIRO-enviro-informatics/interactive-geological-timescale
  • Interactive visualisation links to vocabulary in different representations, e.g., https://vocabs.ardc.edu.au/repository/api/lda/csiro/international-chronostratigraphic-chart/geologic-time-scale-2020/resource?uri=http://resource.geosciml.org/classifier/ics/ischart/Holocene
  • https://github.com/CGI-IUGS/timescale-data/

---

• As a user, would you be satisfied if you could just pick one / n-many period names from a drop-down list?
• I think the sample resolution (idea understood now) goes beyond what we had in mind for the OPTIMETA Geo Plugin... for now? I've created a new issue to track this idea, see Add support for sample resolution as data metadata #123

I'd really like to support your requirements, but don't see a short term opportunity to do so. Next up I will work on #62 to at least allow authors to enter whatever time period they want, and that might already help a lot. It's actually realistic that the calendar popup might be removed because of that.

[CathJex]

So the geological time is essentially a controlled vocabulary of names for time periods, right?

Yes, exactly.

Are you aware of any search interfaces where one can select geological time as input, or any metadata standards that model them?
I'm wondering if these would be best saved in a database by their name, or by actually putting in the numbers.

I'll talk with my scientific editors to see what they say on this.

As a user, would you be satisfied if you could just pick one / n-many period names from a drop-down list?

Yes, I think so. But I'll talk with my scientific editors to see what they think.

[nuest]

Interesting work on "stratigraphic age" incl. automatic extraction from text.

https://link.springer.com/article/10.1007/s12145-014-0171-5#citeas