Changing Times is an interactive tool for understanding, quantifying, and visualising the differences between stratigraphic age models. I initially developed this tool to provide an intuitive way of seeing how age models for the Ediacaran-Cambrian transition (~580 to 510 million years ago) have changed in recent years. This beta release of Changing Times is still in development and may change without notice until a first major version is released.
Changing Times has two built-in datasets: one for Ediacaran-Cambrian carbon isotope chronostratigraphy developed by Fred Bowyer, and one for the Miocene geomagnetic polarity timescale developed by Anna Joy Drury. The app is designed to be flexible and adaptable to other types of data and age models. If you find that your data do not work in the way you expect, please contact me .
When working with temporal data it is important to make sure you are measuring datasets in the same temporal reference frame. However, age models change over time as new data and methods become available. Both acknowledged and unknown uncertainties associated with age models can be large, especially when working with deep time data, and can significantly affect dataset interpretations (e.g. Westerhold et al. 2024 ). Changing Times helps to visualise and quantify the changes between different age model iterations, showing which time intervals have been most affected from one version to another. Overall, this can help show how stable age models are, and which time intervals may be subject to reinterpretation as chronostratigraphic methods and frameworks develop.
Thomas Wong Hearing developed the Shiny app and R code. Claude Code was used to refactor and write tests for the app. Fred Bowyer developed the Ediacaran-Cambrian stratigraphic age models (see Bowyer et al. 2023 ; 2024 ). Anna Joy Drury provided the Miocene geomagnetic polarity timescale data and age models.
Uploaded files must be .xlsx or .csv with one row per datum (e.g. an isotope sample, a fossil first/last appearance, or a magnetochron boundary). Required columns:
Model_
(e.g.
Model_CK1995
,
Model_GTS2020
).
Each column gives the age in Ma for that datum under that age model. Values must be numeric.
Leave the cell blank (or use
NA
) if a datum is not represented in a given age model.
For magnetostratigraphy-style data, also include two text columns named
Magnetochron_base
and
Magnetochron_top
.
Magnetochron_base
is the younger chron, the one with its base at this age;
Magnetochron_top
is the
older chron, the one with its top at this age. Polarity is inferred from chron names ending in
n
(normal) or
r
(reversed).
Rows with missing model ages are dropped from that model only; i.e. a datum can appear in some age models and not others.
Changing Times is still in development and this should be regarded as a beta release. Please contact me if there are specific features you'd like to see. I am planning to include an option for saving out tables of datapoint/event age volatility, as well as options to download the code used for specific calculations and plot rendering.