- fixed a bug in
ar1Surrogates()
that would pop up with the isosopersistent option in rare occasions. See issue 89, and thanks @laurastreib for the bug report.
- Dealt with changes to R's handling of package versions that was throwing errors.
- Added some better error messages in
storey()
to try to help with issue 89
- Reference to
spatstat.explore::CDF.density
replaced withspatstat.univar::CDF.density
following update to spatstat v3.0.0
- added
oxcal.code.export.path
option torunOxcal()
to allow users to optionally save the oxcal model code before it runs
- Attempted to fix the bug that resulted from the rEDM v1.15.0 name change of
make_surrogate_data
toSurrogateData
- Fixed the many
if(is.na(parameter))...
statements that broke with R version 4.2 - Modified the use of the spatstat package (from spatstat.core to spatstat.summary) to accommodate the changes in that package.
- Fixed a bug in
writeBacon()
whereremove.rejected
didn't work properly if there was exactly one rejected date. Thanks @JanPetrik1 for reporting - see issue 71.
- Added an
ask.reservoir
option torunBacon()
that connects to the same parameter inwriteBacon()
. Settingask.reservoir = FALSE
will allow for non-interactive code that usesrunBacon()
and includes reservoir corrections. Thanks @JanPetrik1 for reporting - see issue 70 for details.
- Fixed shape mapping bug with unknown archiveType (thanks Sarah Ivory for reporting)
- Moved nuspectral to suggested package, as it's rarely used and requires compilation See issue 62
- Removed devtools from required packages - it shouldn't be necessary.
- Add an option to erase the bacon tempdir (if using) before each run, and set to default.
- Fixed bug in
writeBacon()
that was making the model overwriting option not work properly. Thanks Sofia Kjellman for reporting the bug.
- Better error messages in
binTs()
- Better error handling in
pcaEns()
- Fixed bug in
createOxCalModel()
that was causing errors with IntCal20 - Suppressed non important warning ("Density was normalised within the computed range of x values [0, 1]"") in
pvalMonteCarlo()
that occurred during ecdf estimation
- add time.var option to
runBam()
- better error handling in
ar1Surrogates()
- better logic and consistency checking in
mapAgeEnsembleToPaleoData()
- Create median variable option now included in
mapAgeEnsembleToPaleoData()
- Fix (at least temporarily) rare bugs in
pcaEns()
andfdr()
- A bunch of quality of life improvements
- Better integration with new lipdR classes
- A new function to create multi-model ensembles!
createMultiModelEnsemble()
- removed unused arima() in
createSyntheticTimeseries()
- Added explicit references to astrochron function
linterp()
incomputeSpectraEns()
. See issue 55
- Improved documentation
runBchron()
:which.table
parameter changed tomeas.table.num
for consistency.
- Removed jitterPositions option following
Bchron
update. See issue 53
- Added handling for shapes in
mapTs()
for tibbles.
- Fixed bug writing Bchron output into LiPD structure. Thanks @alexkjames for recognizing the bug.
- Modified the use of the spatstat package (to spatstat.core) to accommodate the changes in that package.
- Moved astrochron back to a "required" package, now that it's back.
- Moved astrochron to a "suggested" package, after it was removed from CRAN.
- Fixed sampleBaconAges() for handling different versions of R and rbacon.
- Improve and document the
export.quantiles
option inplotTimeseriesEnsRibbons
- Fix bug in
plotTimeAvailabilityTs()
when only one instance of thegroup.var
see issue 49 - Added support for flipping timeseries vertically in
plotTimeseriesStack
to match interpertations, based on TS metadata, specified ininvert.var
. Details in?plotTimeseriesStack
. See issue 30
- Better automatic detection for mapping ranges in
baseMap()
- Added interoperability with "tidyTs" objects, that is, TS objects expressed as tidy (or nested) tibbles.
- Fixed binding warning in
plotTimeAvailabilityTs()
- Fixed bug in
mapTs()
when the regions were really small
- Fixed a bug in
plotTimeAvailabililyTs()
where ifgroup.var
had a variable with a single instance it would fail.
- Added a
NEWS.md
file to track changes to the package. - Added pkgdown documentation, website available at nickmckay.github.io/GeoChronR
- Updated vignettes for consistency with Geochronology manuscript and version 1.0.0
- Fixed typo in
plotCorEns()
legend when usinguse.fdr = FALSE
- This release marks the submission of the manuscript describing geoChronR to "Geochronology", the manuscript, which is under discussion, is available here.
- Major changes to parameter names (terms within functions) were made for version 1.0.0, standardizing them all to lower case, period separated terminology. If you've been using geoChronR for awhile, this will almost certainly break some of your code, sorry.