- further steps in separating 'rice' from 'rintcal' within 'clam'
- now uses the 'rintcal' package for loading calibration curves, and 'rice' for additional radiocarbon/calibration functions
- now points to the renamed rintcal package
- calibrate() and calBP.14C() are now provided by the IntCal package
- replaced .hpd() as now provided as hpd() by the IntCal package
- replaced copyCalibrationCurve() with ccurve() in plots.R
- added vignettes
- some general cleanup of the code
- if depthseq is provided, the depth scale is now plotted correctly
- removed closeAllConnections, as directed by CRAN's maintainers
- repaired a bug where calibrate() couldn't find postbomb curves if required. Reported by David Hein-Griggs, University of Exeter.
- removed links to chrono website, as it has become unstable and could disappear any moment now.
- added an option 'rule' to allow more choice when extrapolating values (see ?approx). This is an internal option and shouldn't normally be important for users.
- now installs and uses the IntCal package to read the radiocarbon calibration curves (this to have less replication of files between packages that use the IntCal curves)
- repaired bug with Est not being the same size as smp (e.g., whenever dates were truncated or removed, either automatically or manually)
- repaired mixed.effect (however, does not use student.t any more)
- updated to the latest IntCal20 calibration curves
- Updated the code to deal with changes in how base-R deals with c() in loops, as suggested by Martin Maechler's e-mail 29 February 2020
- deptime.age and deptime.depth now invisibly return the deposition time values, for future manipulation such as summary(deptime.age(900))
- Updated help functions of clam and calibrate
- new function calBP.14C to find the mu (IntCal 14C age) belonging to a single calendar year (suggested by Andres Christen)
- removed suppression of warnings (suggested by Maris Nartiss)
- added sep argument to mix.calibrationcurves (suggested by Thomas Dye)
- calibrate now invisibly returns dat (suggested by Andres Christen) so now can be used as, e.g., cal <- calibrate(130, 30); cal
- related functions were put into separate .R files
- unnecessary references to Bacon were removed
- dummy -5 cal BP lines were removed from the IntCal13 and SHCal13 calibration curves
- Now a CRAN R package
- Repaired many sundry bugs
- updated the help functions
- updated calibration curves (IntCal13 for both hemispheres and the ocean, plus postbomb curves)
- corrected behaviour of est
- corrected use of alternative separator in .csv files (e.g. ";" instead of ","). Also added option to specify alternative decimal points (e.g. "," instead of ".")
- repaired bug that caused incorrect depths for plot.proxies
- repaired bug with slump, causing some dates to be assigned wrong depths
- corrected bug with drawing sample thickness when extradates applied
- new option for est, midpoint of entire calibrated distributions (above threshold, est=7). This differs slightly from the midpoint of the calibrated hpd ranges (est=3)
- minor changes to the manual, including more explicit citation suggestions
- added function student.t to show effect alternative calibration
- corrected wrong behaviour when using hiatus and slump, e.g., clam(hiatus=470, slump=c(120,140))
- polynomial age-modelling now correctly deals with weights (wght)
- uncertainties for age offsets can be included in calibrate()
- marine offset can now be specified when producing a mixed terrestrial/marine curve (mix.curves())
- new function glue.curves that can be used to extend SHCal04 to 50 kcal BP using IntCal09 and a specified SH offset (obsolete since release of SHCal13)
- corrected references in the manual
#clam 2.0
-
Depth segments of abrupt sedimentation can now be excised with the option slump
-
different colours for C14 and calendar dates
-
accumulation rates are calculated and added as a fifth column to the _ages.txt file
-
future ages can be avoided in the dates and the age-depth model
-
option to rotate the axes, or reverse the order for the age or depth axes
-
optional calibt calibration (Christen and Perez 2009)
-
option to provide .txt files with ages and probs for depths
-
single instead of mirrored calibrated histograms can be drawn in age-depth graphs
-
New function mix.curves
-
New functions pMC.age, age.pMC to convert between 14C ages and percent modern carbon
-
New function proxy plot
-
histograms and confidence ranges for accumulation rates can be calculated for specific depths or ages
-
probability distributions are now normalised, so that precise ages have higher heights than imprecise ages. This can be turned off
-
more consistent naming of dmax, dmin, etc.
-
Warnings are provided in case of clearly wrong settings (e.g. type > 5)
-
Internal calculations for C14 dates are now done in F14C (as in OxCal)
-
core name added to age-plots
-
can now leave thickness column out of .csv file (so, need 6 columns only)
-
instead of standard deviation, the statistically more correct probability is used (e.g., what was formerly sdev=2 is now prob=0.95)
-
use of file _depths.txt now needs to be activated explicitly
-
corrected a bug with calibrating single dates on old or young extremes of the calibration curve
-
corrected wrong behaviour when outliers and mixed.effect are used together
-
corrected a reservoir effect bug in mixed.effect
-
corrected a bug in hpds around 0 BC/AD when BCAD=TRUE
-
corrected a wrong reaction to clam(BCAD=TRUE)
-
corrected wrong yr.axis titles in png and pdf
-
corrected a bug in title option calibrate()
-
better error message when type=5 & hiatus
-
enhanced treatment of plotting parameters (par)
-
better rounding for ages.txt
-
squashed many more small to invisible bugs