Skip to content

4.7.14 TimeOpt

Mingsong Li edited this page Mar 7, 2019 · 1 revision

The method is to determine the optimal sedimentation rates of the proxy series, in a procedure termed “TimeOpt” analysis (Meyers, 2015).

For a “test” sedimentation rate, the TimeOpt method extracts the precession-band amplitude envelope from the proxy data and evaluates the first correlation coefficient (r2envolope) between this envelope and reconstructed eccentricity model. It also evaluates a second correlation coefficient (r2power) between the reconstructed astronomical (eccentricity and precession) model series and the time-calibrated proxy series. Finally, a measure of fit (r2opt) combine both correlation coefficients using an equation: r2opt = r2envolope * r2power.

Monte Carlo simulation with a first-order autoregressive model is used to determine the statistical significance of the observed r2opt value.

This function is largely based on the TimeOpt R script in Astrochron by Steve Meyers.

  • Step 0: Select a time series in depth domain (interpolation may be needed if the sampling rate is uneven).

Warning: the unit of depth-series should be in “meter”.

  • Step 1: In the pop-up window, set the test sedimentation rate:

linear or log model?

Minimum, maximum, and the step of sedimentation rates. (Default values are usually okay)

  • Step 2: Set the middle age of data OR type frequencies of eccentricity and precession.

You’ll only need to give the middle age of the data; the frequencies will be calculated automatically from an astronomical solution of La2004.

The Taner bandpass cut-off frequencies are also adjusted automatically.

If the middle age is > 249 Ma, you may type the frequencies.

  • Step 3: Fit to precession modulations (default), and short-eccentricity modulation may not be reliable.

  • Step 4: If you have typed the frequencies in Step 2, you will also need to adjust frequencies here.

  • Step 5: Simulations are to evaluate the null hypothesis of the optimal sedimentation rate. This can be very time-consuming.

Clone this wiki locally