| Title: | Numerical Calibration of Proxy-Climate Relationships |
|---|---|
| Description: | Bootstrapped response and correlation functions, seasonal correlations and evaluation of reconstruction skills for use in dendroclimatology and dendroecology, see Zang and Biondi (2015) <doi:10.1111/ecog.01335>. |
| Authors: | Christian Zang [aut, cre, cph, trl], Franco Biondi [ctb, cph] |
| Maintainer: | Christian Zang <[email protected]> |
| License: | GPL-3 |
| Version: | 2.0.7.1 |
| Built: | 2024-12-17 08:25:50 UTC |
| Source: | https://github.com/cszang/treeclim |
This function calculates (potentially moving or evolving) response and correlation functions from tree-ring chronologies and monthly climatic data. For the moving case, the calculation is performed repeatedly for consecutive time windows. Function parameters may be bootstrapped to calculate their significance and confidence intervals.
dcc( chrono, climate, selection = -6:9, method = "response", dynamic = "static", moving, win_size = 25, win_offset = 1, start_last = TRUE, timespan = NULL, var_names = NULL, ci = 0.05, boot = "stationary", sb = TRUE, verbose = TRUE )dcc( chrono, climate, selection = -6:9, method = "response", dynamic = "static", moving, win_size = 25, win_offset = 1, start_last = TRUE, timespan = NULL, var_names = NULL, ci = 0.05, boot = "stationary", sb = TRUE, verbose = TRUE )
chrono |
|
climate |
either a |
selection |
either a numeric vector, a modifier, or a chain of modifiers specifying the parameter selection for the model (see Details). |
method |
|
dynamic |
|
moving |
deprecated, use |
win_size |
integer giving the window size for each recalculation in years for a moving analysis, and the initial window size for an evolving analysis. |
win_offset |
integer giving the number of years between each window start in years. |
start_last |
|
timespan |
|
var_names |
|
ci |
|
boot |
|
sb |
|
verbose |
|
This function builds upon and extents the functionality of programme DENDROCLIM2002 (Biondi and Waikul, 2004), and will calculate bootstrapped (and non-bootstrapped) moving and static response and correlation functions in a similar fashion as described in the above mentioned paper. Important extensions include a very flexible parameter selection model (see below), the possibility to use an unlimited number of climate parameters, and the option to use exact bootstrapping.
Input chronology data can be a data.frame such as produced
by function chron of package dplR. It has to be a
data.frame with at least one column containing the
tree-ring indices, and the corresponding years as rownames.
For climatic input data, there are three possibilities: Firstly,
input climatic data can be a data.frame or matrix
consisting of at least 3 rows for years, months and at least one
climate parameter in the given order. Secondly, input climatic
data can be a single data.frame or matrix in the
style of the original DENDROCLIM2002 input data, i.e. one
parameter with 12 months in one row, where the first column
represents the year. Or thirdly, input climatic data can be a
(potentially named) list of one or several of the latter described
data.frame or matrices. If named list is provided,
potentially provided variable names through argument
var_names are ignored. As an internal format dispatcher
checks the format automatically, it is absolutely necessary that
in all three cases, only complete years (months 1-12) are
provided. It is not possible to mix different formats in one go.
Parameters can be selected with the 'selection' parameter in two different ways:
simple selections: as an example -6:9 selects from all
climate variables all months from previous year's June (-6,
previous year's months are specified as negative integers) to
current years September (9, months of the current year are
specified as positive integers) as model parameters. Months from
the previous year and the year before that can be selected using
treeclim-shifters like ..(6) to refer to July of
the year before the previous year.
using modifiers: More complex parameter selections
can be obtained by the modifiers provided in treeclim:
.range, .mean, and .sum. These modifiers
can also be chained to create complex selections. See
treeclim-modifiers for details.
For the exclusion of months, the convenience function
exclude_from (or short exfr) is
provided.
1000 bootstrap samples are taken from the original distributions
of climate and tree-ring data, either using the stationary
bootstrap (Politis and Romano 1994, boot = "stationary") or
classical bootstrap (DENDROCLIM2002-style, boot =
"std"). The stationary bootstrap mimics the stationary properties
of the original time series in the resampled time series by
resampling within blocks. Within each block, the number of
observations is random and has a geometric distribution.
Consequently, the choice of the distribution parameter will affect
the autocorrelation structure of the resampled time
series. Optimal (expected) block length is chosen according to
Politis and White (2004). In the case of response function
analysis, an eigen decomposition of the standardized predictor
matrix is performed. Nonrelevant eigenvectors are removed using
the PVP criterion (Guiot, 1990), principal component scores are
then calculated from the matrices of reduced eigenvectors and
standardized climatic predictors. Response coefficients are found
via singular value decomposition, and tested for significance
using the 95% percentile range method (Dixon, 2001). In case of
correlation function analysis, the coefficients are Pearson's
correlation coefficients. The same method for significance testing
is applied.
There is also the option to use exact bootstrapping like
implemented in seascorr (Meko et al. 2011, boot =
"exact"). In this case, circulant embedding is used to simulate
the tree-ring data 1000 times as time series with the same
frequency characteristics like the original time-series (Percival
& Constantine, 2006). Empirical non-exceedence probabilities are
used to test the coefficients of the response/correlation function
with the original data for significance. For the exact
bootstrapping case, no confidence intervals for the
response/correlation coefficients can be computed.
'dcc' returns an 'object' of class '"tc_dcc"'.
The functions 'summary' and 'plot' are used to obtain and print a summary of the results, and to create a plot. The function 'coef' can be used to extract the coefficients.
An object of class '"tc_dcc"' is a list containing at least the following components:
call |
the call made to function 'dcc' |
coef |
the coefficients, themselves being an object of class 'tc_coef' for the static case, and of class 'tc_mcoef' for the moving case. Objects of class 'tc_coef' are single data.frames, while objects of class 'tc_mcoef' are lists of seperate data.frames for the coefficients (element 'coef'), upper and lower confidence interval (elements 'ci_upper' and 'ci_lower'), and significance flags (element 'significant') |
design |
the design matrix on which this call to 'dcc' operates |
truncated |
the input data truncated to the common timespan or the specified timespan |
original |
the original input data, with the climate data being recast into a single data.frame |
Christian Zang; the original MATLAB code for exact bootstrapping was written by Dave Meko
Biondi, F & Waikul, K (2004) DENDROCLIM2002: A C++ program for statistical calibration of climate signals in tree-ring chronologies. Computers & Geosciences 30:303-311
Dixon, PM (2001) Bootstrap resampling. In: El-Shaarawi, AH, Piegorsch, WW (Eds.), The Encyclopedia of Environmetrics. Wiley, New York.
Guiot, J (1991) The boostrapped response function. Tree-Ring Bulletin 51:39-41
Meko DM, Touchan R, Anchukaitis KJ (2011) Seascorr: A MATLAB program for identifying the seasonal climate signal in an annual tree-ring time series. Computers & Geosciences 37:1234-241
Percival DB, Constantine WLB (2006) Exact simulation of Gaussian Time Series from Nonparametric Spectral Estimates with Application to Bootstrapping. Statistics and Computing 16:25-35
Patton, A. and D.N. Politis and H. White (2009), "CORRECTION TO 'Automatic block-length selection for the dependent bootstrap' by D. Politis and H. White", Econometric Reviews 28(4), 372-375.
Politis, D.N. and H. White (2004), Automatic block-length selection for the dependent bootstrap, Econometric Reviews 23(1), 53-70.
dc_resp <- dcc(muc_spruce, muc_clim)dc_resp <- dcc(muc_spruce, muc_clim)
This is a wrapper around lm for working with the
same data structures and modifiers as dcc does.
dlm( chrono, climate, selection, timespan = NULL, var_names = NULL, param_names = NULL, intercept = TRUE, scale = FALSE, verbose = TRUE )dlm( chrono, climate, selection, timespan = NULL, var_names = NULL, param_names = NULL, intercept = TRUE, scale = FALSE, verbose = TRUE )
chrono |
|
climate |
either a |
selection |
either a numeric vector, a modifier, or a chain of modifiers specifying the parameter selection for the model (see Details). |
timespan |
|
var_names |
|
param_names |
|
intercept |
|
scale |
|
verbose |
|
Input chronology data can be a data.frame such as produced
by function chron of package dplR. It has to be a
data.frame with at least one column containing the
tree-ring indices, and the corresponding years as rownames.
For climatic input data, there are three possibilities: Firstly,
input climatic data can be a data.frame or matrix
consisting of at least 3 rows for years, months and at least one
climate parameter in the given order. Secondly, input climatic
data can be a single data.frame or matrix in the
style of the original DENDROCLIM2002 input data, i.e. one
parameter with 12 months in one row, where the first column
represents the year. Or thirdly, input climatic data can be a
(potentially named) list of one or several of the latter described
data.frame or matrices. If named list is provided,
potentially provided variable names through argument
var_names are ignored. As an internal format dispatcher
checks the format automatically, it is absolutely necessary that
in all three cases, only complete years (months 1-12) are
provided. It is not possible to mix different formats in one go.
In 'dlm', there is no default parameter selection, in contrast to 'dcc'. Parameters can be selected with the 'selection' parameter in two different ways:
simple selections: as an example -6:9 selects from all
climate variables all months from previous year's June (-6,
previous year's months are specified as negative integers) to
current years September (9, months of the current year are
specified as positive integers) as model parameters. Months from
the previous year and the year before that can be selected using
treeclim-shifters like ..(6) to refer to July of
the year before the previous year.
using modifiers: More complex parameter selections
can be obtained by the modifiers provided in treeclim:
.range, .mean, and .sum. These modifiers
can also be chained to create complex selections. See
treeclim-modifiers for details.
For the exclusion of months, the convenience function
exclude_from (or short exfr) is
provided.
With 'dlm' one would usually try to keep the number of predictors low.
For pretty output of the resulting linear model, parameters can be renamed, to e.g. reflect the season they represent.
'dlm' returns an 'object' of class '"tc_dlm"', which is a superclass of 'lm'. Additional elements to what is included in standard 'lm' objects:
call_dlm |
the call made to function 'dlm' |
design |
the design matrix on which this call to 'dlm' operates |
truncated |
the input data truncated to the common timespan or the specified timespan |
original |
the original input data, with the climate data being recast into a single data.frame |
Christian Zang
dlm1 <- dlm(rt_spruce, rt_prec, .sum(6:8), param_names = "summer_prec") summary(dlm1) dlm2 <- dlm(rt_spruce, list(rt_prec, rt_temp), .sum(6:8, "prec") + .mean(6:8, "temp"), var_names = c("prec", "temp"), param_names = c("summer_prec", "summer_temp")) summary(dlm2) anova(dlm1, dlm2)dlm1 <- dlm(rt_spruce, rt_prec, .sum(6:8), param_names = "summer_prec") summary(dlm1) dlm2 <- dlm(rt_spruce, list(rt_prec, rt_temp), .sum(6:8, "prec") + .mean(6:8, "temp"), var_names = c("prec", "temp"), param_names = c("summer_prec", "summer_temp")) summary(dlm2) anova(dlm1, dlm2)
Single months or ranges of months can be excluded from analysis. This is helpful for e.g. excluding winter months without cambial activity.
exclude_from(month, exclude = NULL) exfr(month, exclude = NULL)exclude_from(month, exclude = NULL) exfr(month, exclude = NULL)
month |
range of numeric month ids |
exclude |
range or set of months to exclude |
These convenience function is provided for the exclusion
of months. E.g., .range(exclude_from(-6:10, -11:3)) will
yield the monthly values of all parameters for the months
previous June (-6) to current October (10), but without the
months previous November (-11) to current March (3) in
between. While it is also possible to supply arbitrary vectors
as month specification, and not only ranges as shown in most of
the examples here, this way of excluding e.g. the dormant season
is far more convenient.
a reduced set of numeric month ids
link{.range}, link{.mean},
link{.sum}
exfr(-5:10, -10:3)exfr(-5:10, -10:3)
This function provides a test to decide whether low-frequency modulations in the relationship between climate and tree-growth are significantly stronger or weaker than could be expected by chance.
g_test(x, boot = FALSE, sb = TRUE, check_duration = TRUE)g_test(x, boot = FALSE, sb = TRUE, check_duration = TRUE)
x |
an object of class '"tc_dcc"' as returned from a call to
|
boot |
|
sb |
|
check_duration |
|
This function is a multivariate extension of the test for spurious low-frequency modulations for moving correlations of time series as proposed by Gershunov et al. (2001). In short, 1000 simulations of random data sets are generated, where the climate data is simulated as Gaussian noise, and the tree-data as linear combinations of the climate parameters using the original coefficients of the correlation function, and an error component with a variance equal to the variance unexplained by the individual parameters.
For each iteration, a moving correlation function is calculated with exactly the same settings as the original model. The standard deviation over the individual windows for each parameter is then compared to the bootstrapped distribution of the standard deviation of the simulated data to test for significantly higher or lower low-frequency modulations.
a data.frame with p values for the testing the null
hypothesis that the low-frequency modulation of the
correlations of the variables with tree-growth can be
considered as noise.
Gershunov, A., N. Schneider, and T. Barnett. 2001. Low-frequency modulation of the ENSO-Indian Monsoon rainfall relationship: Signal or noise? Journal of Climate 14:2486-2492.
dc_cor <- dcc(muc_spruce, muc_clim, 3:9, method = "cor", moving = TRUE) g_test(dc_cor)dc_cor <- dcc(muc_spruce, muc_clim, 3:9, method = "cor", moving = TRUE) g_test(dc_cor)
This dataset gives the monthly mean temperature and total precipitation at Forstenrieder Park, Munich, Bavaria, Germany.
data(muc_clim)data(muc_clim)
A data.frame containing four columns with year,
month, temperature and precipitation.
Zang C, Pretzsch H, Rothe A (2012) Size-dependent responses to summer drought in Scots pine, Norway spruce and common oak. Trees - Structure and Function, 26, 557-569.
This dataset gives the modelled tree-ring widths for Picea abies at Forstenrieder Park, Munich, Bavaria, Germany. Tree growth was modeled as a response of low temperatures in previous and current July and August, high temperatures in current February and March, and high precipitation amounts in current July and August.
data(muc_fake)data(muc_fake)
A data.frame containing tree-ring indices and
replication depth with respective years as rownames.
Zang C, Pretzsch H, Rothe A (2012) Size-dependent responses to summer drought in Scots pine, Norway spruce and common oak. Trees - Structure and Function, 26, 557-569.
This dataset gives the tree-ring indices for Picea abies at
Forstenrieder Park, Munich, Bavaria, Germany. The chronology
represents 20 cores from 10 trees. The series were read in using
read.rwl from package dplR, detrended
using a 60a spline with 50% frequency cutoff (function
detrend), and averaged to a chronology using a
robust mean chron.
data(muc_spruce)data(muc_spruce)
A data.frame containing tree-ring indices and
replication depth with respective years as rownames.
Zang C, Pretzsch H, Rothe A (2012) Size-dependent responses to summer drought in Scots pine, Norway spruce and common oak. Trees - Structure and Function, 26, 557-569.
This dataset gives the tree-ring indices for Pinus
sylvestris at a site near Visdalen, Norway. The chronology
represents 34 dated series and spans the period between 1600 and
1983. The raw measurements were obtained from the ITRDB (see
source), and read in, detrended with cubic splines (frequency
cutoff of 50 percent at two thirds of the series length) and
averaged using read.rwl,
detrend, and chron from
package dplR.
data(norw015)data(norw015)
A data.frame containing tree-ring indices and
replication depth with respective years as rownames.
Briffa K. Briffa - Visdalen - PISY - ITRDB NORW015. http://hurricane.ncdc.noaa.gov/pls/paleox/f?p=519:1:::::P1_STUDY_ID:2861. Accessed 2014/07/03.
This data set gives the monthly precipitation sums at country level for Norway in a decadal (DENDROCLIM2002-style) format from the TYN CY 1.1 data set.
data(norway_prec)data(norway_prec)
A data.frame containing thirteen columns with year
and twelve months of temperature data in degree Celsius.
Mitchell TD, Carter TR, Jones PD, Hulme M, New M, et al. (2004) A comprehensive set of high-resolution grids of monthly climate for Europe and the globe: the observed record (1901-2000) and 16 scenarios (2001-2100). Tyndall Centre for Climate Change Research Working Paper 55, 25.
This data set gives the monthly temperature means at country level for Norway in a decadal (DENDROCLIM2002-style) format from the TYN CY 1.1 data set.
data(norway_temp)data(norway_temp)
A data.frame containing thirteen columns with year
and twelve months of temperature data in degree Celsius.
Mitchell TD, Carter TR, Jones PD, Hulme M, New M, et al. (2004) A comprehensive set of high-resolution grids of monthly climate for Europe and the globe: the observed record (1901-2000) and 16 scenarios (2001-2100). Tyndall Centre for Climate Change Research Working Paper 55, 25.
This dataset gives the monthly precipitation sum at Rothenburg ob der Tauber, Bavaria, Germany in a decadal (DENDROCLIM2002-style) format)
data(rt_prec)data(rt_prec)
A data.frame containing thirteen columns with year
and twelve months of precipitation data in mm rainfall.
Zang C, Rothe A, Weis W, Pretzsch H (2011) Zur Baumarteneignung bei Klimawandel: Ableitung der Trockenstress-Anfaelligkeit wichtiger Waldbaumarten aus Jahrringbreiten. Allgemeine Forst- und Jagdzeitung, 182, 98-112.
This dataset gives the tree-ring indices for Picea abies at Rothenburg
ob der Tauber, Bavaria, Germany. The chronology represents 20 cores from 10
trees. The series were read in using read.rwl from
package dplR, detrended using a 60a spline with 50% frequency cutoff
(function detrend), and averaged to a chronology using a
robust mean chron.
data(rt_spruce)data(rt_spruce)
A data.frame containing tree-ring indices and
replication depth with respective years as rownames.
Zang C, Rothe A, Weis W, Pretzsch H (2011) Zur Baumarteneignung bei Klimawandel: Ableitung der Trockenstress-Anfaelligkeit wichtiger Waldbaumarten aus Jahrringbreiten. Allgemeine Forst- und Jagdzeitung, 182, 98-112.
This dataset gives the monthly temperature means at Rothenburg ob der Tauber, Bavaria, Germany in a decadal (DENDROCLIM2002-style) format)
data(rt_temp)data(rt_temp)
A data.frame containing thirteen columns with year
and twelve months of temperature data in degree Celsius.
Zang C, Rothe A, Weis W, Pretzsch H (2011) Zur Baumarteneignung bei Klimawandel: Ableitung der Trockenstress-Anfaelligkeit wichtiger Waldbaumarten aus Jahrringbreiten. Allgemeine Forst- und Jagdzeitung, 182, 98-112.
Calculate seasonal correlation with primary and secondary climate variables and tree-ring data, similar to the seascorr function for MATLAB.
seascorr( chrono, climate, var_names = NULL, timespan = NULL, complete = 9, season_lengths = c(1, 3, 6), primary = 1, secondary = 2, ci = 0.05, verbose = TRUE )seascorr( chrono, climate, var_names = NULL, timespan = NULL, complete = 9, season_lengths = c(1, 3, 6), primary = 1, secondary = 2, ci = 0.05, verbose = TRUE )
chrono |
|
climate |
either a |
var_names |
|
timespan |
|
complete |
|
season_lengths |
|
primary |
position |
secondary |
position |
ci |
|
verbose |
|
This function mimicks the behaviour of the MATLAB function seascorr (Meko et al. 2011), which calculates partial correlations of tree-ring data with a primary and a secondary climatic variable for seasons of different lengths.
Input chronology data can be a data.frame such as produced
by function chron of package dplR. It has to be a
data.frame with at least one column containing the
tree-ring indices, and the corresponding years as rownames.
For climatic input data, there are three possibilities: Firstly,
input climatic data can be a data.frame or matrix
consisting of at least 3 rows for years, months and at least one
climate parameter in the given order. Secondly, input climatic
data can be a single data.frame or matrix in the
style of the original DENDROCLIM2002 input data, i.e. one
parameter with 12 months in one row, where the first column
represents the year. Or thirdly, input climatic data can be a list
of several of the latter described data.frame or
matrices. As an internal format dispatcher checks the
format automatically, it is absolutely necessary that in all three
cases, only complete years (months 1-12) are provided. It is not
possible to mix different formats in one go.
The 'complete' parameter specifies the months of the current year in which tree-growth is assumed to finish. This month marks the last month of the first season, and starting from here, 14 different seasons are computed for each specified season length in one-month steps. E.g., for a starting value of 9 (current September) and season length of 3 months, the first season comprises current July to current September, the second season comprises current June to current August, and the last season comprises previous June to previous August. This results in 14 seasons for a given season length. An arbitrary number of season lengths can be specified.
The choice for primary vs. secondary variable can be made either via numeric selection (the integer value 1 stands for the first variable in the supplied climate data set), or by name ("temp", when one of the variables is named "temp"). The correlation of the primary variable with tree-growth is computed as the simple (Pearson) correlation coefficient, while the influence of the secondary variable on tree-growth is computed with the influence of the primary variable on tree-growth removed.
Like in the original seascorr program, the significance of each (partial) correlation is evaluated using exact bootstrapping by circulant embedding of the tree-ring data (Percival & Constantine, 2006).
'seascorr' returns an 'object' of class '"tc_seascorr"'.
The 'plot' function is used to obtain a plot of the results.
An object of class '"tc_seascorr"' is a list containing at least the following components:
call |
the call made to 'seascorr' |
seasons |
a list of length n, where n is the number of season lengths provided; each list element consists of a data.frame with end month, correlation coefficient and significance flag |
truncated |
the input data truncated to the common timespan or the specified timespan |
original |
the original input data, with the climate data being recast into a single data.frame |
Christian Zang; the procedure incl. exact bootstrapping was implemented first by Dave Meko in MATLAB
Meko DM, Touchan R, Anchukaitis KJ (2011) Seascorr: A MATLAB program for identifying the seasonal climate signal in an annual tree-ring time series. Computers & Geosciences, 37, 1234-1241.
Percival DB, Constantine WLB (2006) Exact simulation of Gaussian Time Series from Nonparametric Spectral Estimates with Application to Bootstrapping. Statistics and Computing 16:25-35
sc <- seascorr(muc_fake, muc_clim) sc plot(sc)sc <- seascorr(muc_fake, muc_clim) sc plot(sc)
This function allows to evaluate the reconstruction skills for a given proxy time series in split-calibration approach.
skills( object, target = NULL, model = "ols", calibration = "50%", timespan = NULL )skills( object, target = NULL, model = "ols", calibration = "50%", timespan = NULL )
object |
an object of class "tc_dcc", "tc_dlm", or "tc_seascorr" |
target |
a treeclim selection modifier specifying the climate target to be reconstructed, see below for details |
model |
one of "ols" or "rma" |
calibration |
which part of the data shall be used as calibration subset? Given as either a range of years, an integer corresponding to the first or last number of observations, or a percentage as character string corresponding to the part of the data set to be used as calibration subset. |
timespan |
timespan to be used to truncate the data |
The result of a call to dcc, dlm, or
seascorr can be used as object for the
function. The required data is then taken from this object and no
further processing of the tree and climate variables has to be
done for by the user. This reflects the flow of analysis, where
first general climate/growth relations are explored, and then the
strongest ones are deployed for reconstruction purposes.
target is an aggregation modifier (one of
.mean, .sum, and
.range). The user should be aware of the fact that
in case the aggregation modifier evaluates to more than one
variable (e.g., summer means for both temperature and
precipiation), a warning message is issued, and only the first
variable is taken into consideration for evaluating the
reconstruction skills. If not specified, the selection from the
original call to dcc, seascorr, or
dlm is used.
The type of regression model (ordinary least squares or errors-in-variables via reduced major axis regression) can be selected.
The part of the data to be used as a calibration subset can be specified in three different ways: 1) as a range of years, these are then taken as calibration period; 2) as a single integer, if positive, this number of observations at the recent end of the data set is taken as calibration set, if negative, this number of oldest observations is taken; and 3) as a character string giving a percentage of values, e.g., "-40%" would select the 40% oldest observations, while "55%" would select the 55% most recent ones.
The relationship between climate and tree-ring data is evaluated for the calibration period and the complete data set. Frequently used verification statistics are computed: reduction of error (RE), coefficient of efficiency (CE), and the Durban-Watson statistic (DW) (Cook et al. 1994, Durbin and Watson, 1951).
'skills' returns an 'object' of class '"tc_skills"'.
An object of class '"tc_skills"' is a list containing at least the following components:
call |
the call made to function 'skills' |
target |
the target used for reconstruction |
r.cal |
the coefficient of correlation for the calibration timespan |
r.full |
the coefficient of correlation for the complete data set |
coef.cal |
regression coefficients for the calibration model |
coef.full |
regression coefficients for the full model |
p.cal |
significance for the calibration model |
p.full |
significance for the full model |
RE |
reduction of error statistic |
CE |
coefficient of efficiency statistic |
DW |
Durbin-Watson statistic |
RMSE |
The root mean squared error for the prediction interval |
cal.model |
the complete calibration model (an object of class 'lmodel2') |
full.mode |
the complete full model (an object of class 'lmodel2') |
Cook E, Briffa K, Jones P (1994) Spatial regression methods in dendroclimatology: A review and comparison of two techniques. International Journal of Climatology, 14, 379-402.
Durbin, J, Watson, GS (1951) Testing for serial correlation in least squares regression. Biometrika 38:159-78.
dc <- dcc(muc_fake, muc_clim, .mean(6:9, "temp") + .sum(6:9, "prec")) sk <- skills(dc) sk plot(sk)dc <- dcc(muc_fake, muc_clim, .mean(6:9, "temp") + .sum(6:9, "prec")) sk <- skills(dc) sk plot(sk)
This dataset gives the tree-ring indices for Pinus
sylvestris at a site near Penota, Spain. The chronology
represents 18 dated series and spans the period between 1763 and
1991. The chronology was obtained from the ITRDB (see source), and
read in using read.crn from package
dplR.
data(spai020)data(spai020)
A data.frame containing tree-ring indices and
replication depth with respective years as rownames.
Fernandez-Cancio A, Genova Fuster M: Fernandez-Cancio - Penota - PISY - ITRDB SPAI020, http://hurricane.ncdc.noaa.gov/pls/paleox/f?p=519:1:::::P1_STUDY_ID:3255, accessed 2014/07/03.
This data set gives the monthly precipitation sums at country level for Spain in a decadal (DENDROCLIM2002-style) format from the TYN CY 1.1 data set.
data(spain_prec)data(spain_prec)
A data.frame containing thirteen columns with year
and twelve months of temperature data in degree Celsius.
Mitchell TD, Carter TR, Jones PD, Hulme M, New M, et al. (2004) A comprehensive set of high-resolution grids of monthly climate for Europe and the globe: the observed record (1901-2000) and 16 scenarios (2001-2100). Tyndall Centre for Climate Change Research Working Paper 55, 25.
This data set gives the monthly temperature means at country level for Spain in a decadal (DENDROCLIM2002-style) format from the TYN CY 1.1 data set.
data(spain_temp)data(spain_temp)
A data.frame containing thirteen columns with year
and twelve months of temperature data in degree Celsius.
Mitchell TD, Carter TR, Jones PD, Hulme M, New M, et al. (2004) A comprehensive set of high-resolution grids of monthly climate for Europe and the globe: the observed record (1901-2000) and 16 scenarios (2001-2100). Tyndall Centre for Climate Change Research Working Paper 55, 25.
Deparse list structure from month specification into single calls, call an aggregating function to collect the data, and return a parameter set for calibration. In general, the months must be specified as a list or nested list structure. Possible cases are: list() list(...) list(list(...)) list(list(...), list(...), ...)
tc_design(selection, climate, check_2 = TRUE)tc_design(selection, climate, check_2 = TRUE)
selection |
the list structure to specify the month selection |
climate |
the climate data as returned by make_pmat |
check_2 |
should there be a check for less than 2 variables in the design matrix? |
a data.frame
This is a plotting function that takes the output from a dynamic ("moving"
or "evolving") dcc run to trace the temporal development of all or
selected variables. The variables can be selected using their name, partial
matching is allowed.
traceplot(x, variables = NULL, facet = FALSE)traceplot(x, variables = NULL, facet = FALSE)
x |
an object returned from a call to |
variables |
a character vector of variable names to be plotted; if NULL (default), all variables will be traced |
facet |
logical: shall the plot be splitted into facets for different variables? |
an object of class "gg".
dc_resp <- dcc(muc_spruce, muc_clim, 4:9, dynamic = "evolving") traceplot(dc_resp, c("prec.curr.may", "prec.curr.jun"))dc_resp <- dcc(muc_spruce, muc_clim, 4:9, dynamic = "evolving") traceplot(dc_resp, c("prec.curr.may", "prec.curr.jun"))
These modifiers are used to select specific months from specific
climate parameters, and potentially transform the selections into
their respective sums or means. The modifiers can be chained
together using '+'.
.range corresponds to using all specified months
separately, while .sum and .mean will use the sums
and means of the specified months. These modifiers also allow to
select specific climatic variables, addressed by name. Thus,
.mean(4:8, "temp") will select the mean for climate
parameter "temp" for the months April to August. Not only ranges,
but also individual vectors can be used for month specification,
like e.g., .range(c(1, 3, 4, 5).
The modifiers can be chained together using the '+' symbol, which
makes it possible to create arbitrarily complex selections of
climate parameters for calibration. E.g., .mean(2:5,
"temp") + .sum(2:5, "prec") will yield the February-to-May mean
for the variable "temp" and the sum of the variable "prec" for the
same time. While there is no limitation for number of lists that
can be chained together, 'dcc' will not check for meaningful
specifications. Testing smart hypotheses is up the researcher.
.range(.months = NULL, .variables = NULL) .mean(.months = NULL, .variables = NULL) .sum(.months = NULL, .variables = NULL).range(.months = NULL, .variables = NULL) .mean(.months = NULL, .variables = NULL) .sum(.months = NULL, .variables = NULL)
.months |
numeric identifiers for the months (-1 for previous January until 12 for current December, with -6 for previous June, etc.) |
.variables |
names of the variables the modifier shall be applied to |
An object of class 'tc_paramlist' defining a list of climate variables for further processing, consisting of specifications for the time frame (months) and the selected variables.
my_calib <- dcc(rt_spruce, list(prec = rt_prec, temp = rt_temp), .range(4:9, "temp") + .mean(-5:-9, "temp") + .sum(3:10, "prec")) plot(my_calib)my_calib <- dcc(rt_spruce, list(prec = rt_prec, temp = rt_temp), .range(4:9, "temp") + .mean(-5:-9, "temp") + .sum(3:10, "prec")) plot(my_calib)
These functions are used to specify months from previous years for
the design matrix. .(1) is the same as -1, and
specifies previous year's January, whereas ..(1) signifies
January of the year prior to the last year.
.(x) ..(x).(x) ..(x)
x |
numeric identifier (of length 1) for the months (1 for January, ..., 12 for December) |
an integer value specifying the month for selection