REE input data

Upload a comma separated file (csv) containing 14 columns for each of the REE, in ppm. Each pattern should be in its own row. See sample file: the O'Neill 2016 database .

Anomalous elements to exclude from fit

Choose whether Ce, Eu, or Gd should be treated as anomalous and excluded from the fitted rare earth element pattern.

Fitting model options

Choose normalisation, lambda 4 fitting, and tetrad fitting options used in the REE pattern model.

Plot options

Choose which measured data, fitted curves, lambda components, and tetrad components are displayed in the REE pattern plot.

Analysis settings

Choose the F matrix parameterisation, uncertainty estimate, and variables used for bivariate plots.

Imputation input data

This function reconstructs incomplete REE rows using measured elements and, optionally, uploaded total rare earth contents. If a total csv/text file is uploaded, one total value must be provided per row in the same order as the rows in the table. If no total file is uploaded, imputation proceeds from the best fitted pattern using the other selected settings.

Run imputation

Choose whether to impute only the selected table row or all rows with missing rare earth element values.

Download latest imputation report

Available after running imputation. This ZIP records the most recent actual imputation run, not the live preview.

Imputation method and parameters

Choose either penalty-based or Bayesian imputation, then set the parameters for the selected method.

Tick a checkbox to impose a maximum absolute value for that lambda parameter.

Anomaly controls used for imputation

If selected, Ce and/or Eu are treated as anomalous during imputation. Measured anomalous elements are excluded from the lambda fit but still included in the uploaded total. Missing anomalous elements are reconstructed from the fitted pattern multiplied by the anomaly factor below (1 = no anomaly).

2026-08-12: Fixed horizontal scrolling issue.
2026-08-08: Added option to download imputation statistics, and other minor bug fixes.
2026-08-04: Visual upgrades and ARIA implementation.
2026-08-01: Added app metadata
2026-07-25: Bayesian imputation added.
2026-07-15: R code blocks moved to external files.
2026-07-09: CSS and JS moved to external files.
2026-06-30: Added geometry score to imputation stats.
2026-06-26: Added imputation stats.
2026-06-23: Added missing data imputation tab.
2026-06-21: Fixed error during chi-square plot download.
2026-06-18: Displayed row is now always highlighted, even when not in focus.
2026-06-14: Visual updates to fit ANU style.
2026-05-13: Substantially improved handling of poorly formatted csv files during upload.
2025-08-07: Fixed error in bivariate plotting.
2025-06-03: F matrix parameterisation option now updates correctly.
2022-01-05: Updated citable reference.
2021-12-08: Fixed bug that caused app to crash when data with many missing elements were uploaded.
2021-07-19: Updated citable reference to Anenburg and Williams (2021).
2021-05-17: Chi-sq distribution plot now correctly updates when changing uncertainty estimate.
2021-03-03: Added link to online workshop.
2021-02-19: Added option to plot the measured pattern, on by default.
2021-02-18: Plotted fit now takes into account selected anomalies.
2021-01-25: Fixed bug that everything was fitted twice, and added a progress indicator.
2021-01-24: Updated anomaly notation. Added source code link.
2021-01-11: Clarified adjusted r-squared notation.
2021-01-09: Added instructions. Updated sample file to use O'Neills database. Fixed bug with missing chi-sq values.
2021-01-08: Fixed bug with negative values, treated as missing data.
2021-01-03: Fixed bug with zero values.
2021-01-02: Tetrads now use corrected parabolas according to polynomial fit to ionic radii
2020-12-29: Added option for anomalous Gd.
2020-12-23: More decimal digits, improves precision for ppb-level data.
2020-12-15: Added bivariate plots.
2020-12-14: Added reduced chi-sq tab with histogram, observed, and theoretical distribution.
2020-12-11: Now using tau symbols for tetrad coefficients.
2020-11-30: Cosmetic upgrades, more detailed file export.
2020-11-29: Fixed a small mistake in the input.
2020-11-27: Implemented tetrads, added stats: r-squared, se, p-values.
2020-11-18: Added option to prevent chondrite normalisation (e.g. when using partition coefficients).
2020-11-17: Added MSWD.
2020-11-16: Improved data table appearance, added option not to fit λ4.
2020-11-15: Enabled downloads.
2020-11-14: Added file upload.
2020-11-13: Basic functionality working.
2020-11-05: Initial release.

License

The MIT License

Copyright © 2026 Michael Anenburg

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the ‘Software’), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED ‘AS IS’, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Source code

Download in R format:

https://lambdar.rses.anu.edu.au/blambdar/source/

Current prior mean

Current Bayesian prior mean values for lambda 1 to lambda 4. Values are editable only when the Custom prior is selected.

Current prior covariance

Current Bayesian prior covariance matrix for lambda 1 to lambda 4. Values are editable only when the Custom prior is selected.