paleocar is an R package implementing functions to perform spatio-temporal paleoclimate reconstruction from tree-rings using the CAR (Correlation Adjusted corRelation) approach of Zuber and Strimmer as implemented in the care package for R. It is optimized for speed and memory use.
This is based on the approach used in Bocinsky and Kohler (2014):
Bocinsky, R. K. and Kohler, T. A. (2014). A 2,000-year reconstruction of the rain-fed maize agricultural niche in the US Southwest. Nature Communications, 5:5618. doi: 10.1038/ncomms6618.
The primary difference between the latest version of paleocar and that presented in Bocinsky and Kohler (2014) is, here, model selection is performed by minimizing the corrected Akaike's Information Criterion.
A more recent reference would be Bocinsky et al. (2016):
Bocinsky, R. K., Rush, J., Kintigh, K. W., and Kohler, T. A. (2016). Exploration and exploitation in the macrohistory of the pre-Hispanic Pueblo Southwest. Science Advances, 2:e1501532.
This package has been built and tested on a source (Homebrew) install of R on macOS 10.12 (Sierra), and has been successfully run on Ubuntu 14.04.5 LTS (Trusty), Ubuntu 16.04.1 LTS (Xenial) and binary installs of R on Mac OS 10.12 and Windows 10.
- Kyle Bocinsky - Crow Canyon Archaeological Center, Cortez, CO
-
Development version from GitHub:
install.packages("devtools") devtools::install_github("bocinsky/paleocar") library(paleocar)
-
Linux (Ubuntu 14.04.5 or 16.04.1):
First, in terminal:
sudo add-apt-repository ppa:ubuntugis/ppa -y
sudo apt-get update -q
sudo apt-get install libssl-dev libcurl4-openssl-dev netcdf-bin libnetcdf-dev gdal-bin libgdal-devThen, in R:
update.packages("survival")
install.packages("devtools")
devtools::install_github("bocinsky/paleocar")
library(paleocar)This demo script is available in the /inst folder at the location of the installed package.
library(paleocar)
library(magrittr) # The magrittr package enables piping in R.
# Set a directory for testing
testDir <- "./paleocar_test/"
# and create it if necessary
dir.create(testDir, showWarnings=F, recursive=T)paleocar ships with test files defining a study area (Mesa Verde National Park), and pre-extracted data from the International Tree Ring Databank using the FedData package. See the data-raw/data.R script (or the documentation for FedData) to learn how to download these data.
# Load spatial polygon for the boundary of Mesa Verde National Park (MVNP) in southwestern Colorado:
data(mvnp)
# Get Tree-ring data from the ITRDB for 10-degree buffer around MVNP
data(itrdb)
# Get 1/3 arc-second PRISM gridded data for the MVNP north study area (water-year [October--September] precipitation, in millimeters)
data(mvnp_prism)paleocar can be run for either single location given by a vector of annualized climate data, a matrix of locations, or over gridded climate data such as PRISM in raster format. There are three primary functions:
paleocar_models()calculates the CAR-ranked linear models for all reconstructionspredict_paleocar_models()generates climate predictions over a specified prediction period, anduncertainty_paleocar_models()generates an estimate of model uncertainty over a specified prediction period.
Finally, the paleocar() method is a convenience wrapper that runs all three of these functions and returns a list with their output. See the documentation for each function for details.
paleocar may be run for a single location by providing a vector of annualized values to be reconstructed. Simply provide a numeric vector the same length as your calibration years as the predictands parameter.
# Extract a vector of annualized climate data (the first cell in the raster)
mvnp_prism.vector <- mvnp_prism[1][1,]
test.vector <- paleocar_models(predictands = mvnp_prism.vector,
chronologies = itrdb,
calibration.years = 1924:1983,
prediction.years = 1:2000,
verbose = T)## Calculating PaleoCAR models
##
## Prepare data and calculate CAR scores: 0.01 minutes
##
## Calculating models of with 1 input vectors.
## Define models: 0.02 minutes
## Calculate 5 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.02 minutes
## 123 cell-years remaining
##
## Calculating models of with 2 input vectors.
## Define models: 0.02 minutes
## Calculate 7 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.02 minutes
## 115 cell-years remaining
##
## Calculating models of with 3 input vectors.
## Define models: 0.02 minutes
## Calculate 7 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.02 minutes
## 41 cell-years remaining
##
## Calculating models of with 4 input vectors.
## Define models: 0.01 minutes
## Calculate 6 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.02 minutes
## 13 cell-years remaining
##
## Calculating models of with 5 input vectors.
## Define models: 0.01 minutes
## Calculate 2 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.01 minutes
## 3 cell-years remaining
##
## Calculating models of with 6 input vectors.
## Define models: 0.01 minutes
## Calculate 1 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.01 minutes
##
## Total Modeling Time: 0.1107242 minutes
##
## Optimizing models: 0 minutes
# Generate predictions and uncertainty (and plot timeseries of each)
predict_paleocar_models(models = test.vector,
meanVar = "chained",
prediction.years = 600:1300) %>%
plot(x = as.numeric(names(.)),
y = .,
type = "l")## Warning in predict_paleocar_models(models = test.vector, meanVar
## = "chained", : Chained mean-variance matching requires that the
## prediction.years include the calibration period. Changing prediction years
## to include calibration period.
uncertainty_paleocar_models(test.vector,
prediction.years = 600:1300) %>%
plot(x = as.numeric(names(.)),
y = .,
type = "l")
paleocar reconstruction for multiple locations using the same set of predictors (in this case, tree-ring chronologies)
Running paleocar on a matrix of locations (predictands) will generate reconstructions that select from the same set of predictors (chronologies). The matrix must be formatted such that each location is in a column, and each row is a year of data. Note that the number of rows of the matrix must be the same as the number of years provided to calibration.years.
# Extract a matrix of annualized climate data (all cells in the raster)
mvnp_prism.matrix <- mvnp_prism %>%
raster::as.matrix() %>%
t()
# Print to show format
mvnp_prism.matrix %>%
tibble::as_tibble()## # A tibble: 60 × 624
## V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11 V12
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 291 291 291 294 297 301 304 309 314 318 324 333
## 2 364 364 368 371 375 382 389 395 404 409 420 427
## 3 387 388 390 394 398 403 408 413 424 427 436 443
## 4 506 507 510 516 521 530 539 547 556 563 574 582
## 5 254 254 257 258 261 268 269 278 284 285 293 298
## 6 436 437 439 443 451 459 469 478 490 498 508 521
## 7 259 261 263 266 271 275 284 289 297 301 308 315
## 8 290 290 289 292 295 298 299 300 304 308 314 318
## 9 419 419 421 426 431 437 442 449 455 461 470 479
## 10 239 243 245 249 254 260 265 272 279 284 292 297
## # ... with 50 more rows, and 612 more variables: V13 <dbl>, V14 <dbl>,
## # V15 <dbl>, V16 <dbl>, V17 <dbl>, V18 <dbl>, V19 <dbl>, V20 <dbl>,
## # V21 <dbl>, V22 <dbl>, V23 <dbl>, V24 <dbl>, V25 <dbl>, V26 <dbl>,
## # V27 <dbl>, V28 <dbl>, V29 <dbl>, V30 <dbl>, V31 <dbl>, V32 <dbl>,
## # V33 <dbl>, V34 <dbl>, V35 <dbl>, V36 <dbl>, V37 <dbl>, V38 <dbl>,
## # V39 <dbl>, V40 <dbl>, V41 <dbl>, V42 <dbl>, V43 <dbl>, V44 <dbl>,
## # V45 <dbl>, V46 <dbl>, V47 <dbl>, V48 <dbl>, V49 <dbl>, V50 <dbl>,
## # V51 <dbl>, V52 <dbl>, V53 <dbl>, V54 <dbl>, V55 <dbl>, V56 <dbl>,
## # V57 <dbl>, V58 <dbl>, V59 <dbl>, V60 <dbl>, V61 <dbl>, V62 <dbl>,
## # V63 <dbl>, V64 <dbl>, V65 <dbl>, V66 <dbl>, V67 <dbl>, V68 <dbl>,
## # V69 <dbl>, V70 <dbl>, V71 <dbl>, V72 <dbl>, V73 <dbl>, V74 <dbl>,
## # V75 <dbl>, V76 <dbl>, V77 <dbl>, V78 <dbl>, V79 <dbl>, V80 <dbl>,
## # V81 <dbl>, V82 <dbl>, V83 <dbl>, V84 <dbl>, V85 <dbl>, V86 <dbl>,
## # V87 <dbl>, V88 <dbl>, V89 <dbl>, V90 <dbl>, V91 <dbl>, V92 <dbl>,
## # V93 <dbl>, V94 <dbl>, V95 <dbl>, V96 <dbl>, V97 <dbl>, V98 <dbl>,
## # V99 <dbl>, V100 <dbl>, V101 <dbl>, V102 <dbl>, V103 <dbl>, V104 <dbl>,
## # V105 <dbl>, V106 <dbl>, V107 <dbl>, V108 <dbl>, V109 <dbl>,
## # V110 <dbl>, V111 <dbl>, V112 <dbl>, ...
test.matrix <- paleocar_models(predictands = mvnp_prism.matrix,
chronologies = itrdb,
calibration.years = 1924:1983,
prediction.years = 1:1985,
verbose = T)## Calculating PaleoCAR models
##
## Prepare data and calculate CAR scores: 0.08 minutes
##
## Calculating models of with 1 input vectors.
## Define models: 0.02 minutes
## Calculate 9 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.04 minutes
## 69264 cell-years remaining
##
## Calculating models of with 2 input vectors.
## Define models: 0.02 minutes
## Calculate 24 linear models: 0.02 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.05 minutes
## 64246 cell-years remaining
##
## Calculating models of with 3 input vectors.
## Define models: 0.02 minutes
## Calculate 34 linear models: 0.02 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.05 minutes
## 47452 cell-years remaining
##
## Calculating models of with 4 input vectors.
## Define models: 0.02 minutes
## Calculate 36 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.04 minutes
## 24085 cell-years remaining
##
## Calculating models of with 5 input vectors.
## Define models: 0.02 minutes
## Calculate 27 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.03 minutes
## 10839 cell-years remaining
##
## Calculating models of with 6 input vectors.
## Define models: 0.02 minutes
## Calculate 12 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.03 minutes
##
## Total Modeling Time: 0.2388568 minutes
##
## Optimizing models: 0.05 minutes
# Generate predictions and uncertainty (and plot location means in uncertainty)
predict_paleocar_models(models = test.matrix,
meanVar = "chained",
prediction.years = 600:1300) %>%
rowMeans() %>%
plot(x = as.numeric(names(.)),
y = .,
type = "l")## Warning in predict_paleocar_models(models = test.matrix, meanVar
## = "chained", : Chained mean-variance matching requires that the
## prediction.years include the calibration period. Changing prediction years
## to include calibration period.
uncertainty_paleocar_models(models = test.matrix,
prediction.years = 600:1300) %>%
rowMeans() %>%
plot(x = 600:1300,
y = .,
type = "l")
Paleocar can also be performed over a gridded climate dataset such as PRISM, so long as it is a RasterStack or RasterBrick as defined in the raster package for R. Results will be returned in RasterBrick format.
# Print to show format
mvnp_prism## class : RasterStack
## dimensions : 24, 26, 624, 60 (nrow, ncol, ncell, nlayers)
## resolution : 0.008333333, 0.008333333 (x, y)
## extent : -108.5542, -108.3375, 37.15417, 37.35417 (xmin, xmax, ymin, ymax)
## coord. ref. : +proj=longlat +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +no_defs
## names : X1924, X1925, X1926, X1927, X1928, X1929, X1930, X1931, X1932, X1933, X1934, X1935, X1936, X1937, X1938, ...
## min values : 286, 360, 387, 499, 248, 434, 259, 289, 417, 239, 231, 324, 304, 377, 368, ...
## max values : 498, 602, 615, 745, 417, 739, 437, 420, 690, 434, 364, 628, 588, 612, 720, ...
test.raster <- paleocar_models(predictands = mvnp_prism,
chronologies = itrdb,
calibration.years = 1924:1983,
prediction.years = 1:2000,
verbose = T)## Calculating PaleoCAR models
##
## Prepare data and calculate CAR scores: 0.07 minutes
##
## Calculating models of with 1 input vectors.
## Define models: 0.03 minutes
## Calculate 9 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.04 minutes
## 76752 cell-years remaining
##
## Calculating models of with 2 input vectors.
## Define models: 0.03 minutes
## Calculate 24 linear models: 0.02 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.05 minutes
## 71734 cell-years remaining
##
## Calculating models of with 3 input vectors.
## Define models: 0.03 minutes
## Calculate 34 linear models: 0.02 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.05 minutes
## 53999 cell-years remaining
##
## Calculating models of with 4 input vectors.
## Define models: 0.03 minutes
## Calculate 36 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.05 minutes
## 28331 cell-years remaining
##
## Calculating models of with 5 input vectors.
## Define models: 0.03 minutes
## Calculate 27 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.04 minutes
## 13369 cell-years remaining
##
## Calculating models of with 6 input vectors.
## Define models: 0.02 minutes
## Calculate 12 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.03 minutes
##
## Total Modeling Time: 0.2535981 minutes
##
## Optimizing models: 0.05 minutes
# Generate predictions and errors
test.raster.predictions <- predict_paleocar_models(models = test.raster,
meanVar = "chained",
prediction.years = 600:1300)## Warning in predict_paleocar_models(models = test.raster, meanVar
## = "chained", : Chained mean-variance matching requires that the
## prediction.years include the calibration period. Changing prediction years
## to include calibration period.
test.raster.uncertainty <- uncertainty_paleocar_models(models = test.raster,
prediction.years = 600:1300)
# Plot the mean predictions and uncertainty
test.raster.predictions %>%
raster::mean() %>%
raster::plot()
test.raster.uncertainty %>%
raster::mean() %>%
raster::plot()
The paleocar() convenience wrapper returns a list containing the models, reconstructions, and uncertainty. The paleocar() method also automatically saves the output of predict_paleocar_models() and errors_paleocar_models(). Pass variables through this function to other ones (e.g., meanVar = "chained").
# Generate models and perform the reconstruction and error predictions.
mvnp_recon <- paleocar(predictands = mvnp_prism,
label = "mvnp_prism",
chronologies = itrdb,
calibration.years = 1924:1983,
prediction.years = 1:2000,
out.dir = testDir,
meanVar = "none",
floor = 0,
ceiling = NULL,
force.redo = T,
verbose = T)##
## Calculating all models
## Calculating PaleoCAR models
##
## Prepare data and calculate CAR scores: 0.07 minutes
##
## Calculating models of with 1 input vectors.
## Define models: 0.03 minutes
## Calculate 9 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.04 minutes
## 76752 cell-years remaining
##
## Calculating models of with 2 input vectors.
## Define models: 0.03 minutes
## Calculate 24 linear models: 0.02 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.05 minutes
## 71734 cell-years remaining
##
## Calculating models of with 3 input vectors.
## Define models: 0.03 minutes
## Calculate 34 linear models: 0.02 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.05 minutes
## 53999 cell-years remaining
##
## Calculating models of with 4 input vectors.
## Define models: 0.03 minutes
## Calculate 36 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.05 minutes
## 28331 cell-years remaining
##
## Calculating models of with 5 input vectors.
## Define models: 0.02 minutes
## Calculate 27 linear models: 0.01 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.04 minutes
## 13369 cell-years remaining
##
## Calculating models of with 6 input vectors.
## Define models: 0.02 minutes
## Calculate 12 linear models: 0 minutes
## Clean linear models: 0 minutes
## Total modeling time: 0.03 minutes
##
## Total Modeling Time: 0.2556791 minutes
##
## Optimizing models: 0.05 minutes
##
## Generating prediction
##
## Generating uncertainty predictions
##
## The entire reconstruction took 0.49 minutes
# Examine the structure of the output
str(mvnp_recon, max.level = 2)## List of 3
## $ models :List of 5
## ..$ models :Classes 'data.table' and 'data.frame': 5527 obs. of 8 variables:
## .. ..- attr(*, "sorted")= chr [1:2] "cell" "year"
## .. ..- attr(*, ".internal.selfref")=<externalptr>
## ..$ predictands :Formal class 'RasterStack' [package "raster"] with 11 slots
## ..$ predictor.matrix : num [1:60, 1:120] 1.315 0.883 1.354 1.011 1.354 ...
## .. ..- attr(*, "dimnames")=List of 2
## ..$ reconstruction.matrix: num [1:2000, 1:120] NA NA NA NA NA NA NA NA NA NA ...
## .. ..- attr(*, "dimnames")=List of 2
## ..$ carscores :Classes 'data.table' and 'data.frame': 624 obs. of 120 variables:
## .. .. [list output truncated]
## .. ..- attr(*, ".internal.selfref")=<externalptr>
## $ predictions:Formal class 'RasterBrick' [package "raster"] with 12 slots
## $ uncertainty:Formal class 'RasterBrick' [package "raster"] with 12 slots
You can quickly load a prior reconstruction by setting force.redo = FALSE:
# Generate models and perform the reconstruction and error predictions.
mvnp_recon <- paleocar(predictands = mvnp_prism,
label = "mvnp_prism",
chronologies = itrdb,
calibration.years = 1924:1983,
prediction.years = 1:2000,
out.dir = "./",
meanVar = "none",
floor = 0,
ceiling = NULL,
force.redo = F,
verbose = T)##
## Calculating all models
##
## Generating prediction
##
## Generating uncertainty predictions
##
## The entire reconstruction took 0 minutes
mvnp_recon$predictions %>%
raster::mean() %>%
raster::plot()