eirikmn/bremla: Bayesian Regression Modeling of Layer-Counted Archives

Performs efficient Bayesian regression modeling of layer-counted climate proxy records. This model defines a probabilistic age-depth model using a generalized additive model (GAM), where physical processes and systematic counting errors can be described using a regression model and the remaining error can be explained using a noise process. The model is fitted to the accumulated number of layers (number of counted layers per unit of depth) using integrated nested Laplace approximation via the INLA package (<https://www.r-inla.org>). An ensemble of plausible chronologies are then sampled from the estimated posterior distributions, on which further analyses can be performed such that dating uncertainty propagation is treated rigorously. Additionally, the package also allows the chronologies to be constrained using supplied tie-points. The simulated chronologies can also be used to estimate the dating uncertainty of the onsets of climate transitions. Details on the age-depth model in Myrvoll-Nilsen, Riechers, Rypdal and Boers (2022) <doi:10.5194/cp-18-1275-2022>, and on how to incorporate tie-points in Myrvoll-Nilsen, Riechers and Boers (preprint).

Getting started

Package details

Maintainer
LicenseGPL (>= 2)
Version0.1.0.9001
URL https://github.com/eirikmn/bremla
Package repositoryView on GitHub
Installation Install the latest version of this package by entering the following in R:
install.packages("remotes")
remotes::install_github("eirikmn/bremla")
eirikmn/bremla documentation built on Jan. 25, 2025, 4:41 a.m.