CRAN release ‘Luminescence’ v1.3.0

by RLum.Network (July 22, 2026)

We are proud to announce the release of 'Luminescence' version 1.3.0. This is a major release coming four months after our previous release and bringing a number of (breaking) improvements and new functionality.

Read up the post for the 1.3.0 release on the REPLAY website for more details on what has changed. The ones reported below are just a selection of the updates that have occurred in the package.

Highlights

  • fit_DoseResponseCurve() has seen a number of breaking changes which affect both the numerical results produced and the user interface. In particular, the default weights applied during fitting are now based on inverse variance, as this is more correct compared to the previous approach, which assumed that uncertainties scaled with the dose. The values accepted by the fit methods has also changed (for example, EXP is now called SSE, for single saturating exponential): this aligns the nomenclature used in the package to the one most commonly used in the literature. Also the object returned has seen some changes in the set of columns returned.

    The function has also been made more robust by slightly tweaking how the starting points for most models are generated. This should allow it to find a valid fit for more data sets than before; however, minor numerical differences are to be expected for data sets for which a fit could be obtained by earlier versions.

    Given that this function is called by analyse_SAR.CWOSL(), analyse_SAR.TL(), analyse_baSAR(), analyse_pIRIRSequence(), calc_Lamothe2003(), calc_Huntley2006(), plot_DRCSummary() and plot_DetPlot(), they are also similarly affected.

  • analyse_SAR.NCF() is a new function that implements the natural sensitivity correction for the assessment of SAR-based palaeodoses, proposed originally by Singhvi et al. (2011). This is currently in beta, pending more testing on real data.

  • calc_OSLLxTxRatio() supports the new od_rates argument, which can be set to activate an alternative approach for error estimation of Lx/Tx ratios based on Bluszcz, Adamiec and Herr (2015).

  • analyse_SAR.CWOSL() was enhanced to produce objects that can replot themselves via plot_RLum(), so it’s no longer necessary to re-run the whole analysis just to regenerate the plots. This is a change in behaviour from previous versions, where an abanico plot was generated: the old behaviour can be obtained by calling plot_AbanicoPlot() directly.

To see all the other gory details on what has changed, please see the 1.3.0 release page on GitHub.

Let’s enjoy spring together with Luminescence!

Your R Luminescence Team