10.5061/DRYAD.MB605
Ogle, Kiona
Northern Arizona University
Ryan, Edmund
Lancaster University
Dijkstra, Fieke A.
Western Sydney University
Pendall, Elise
Western Sydney University
Dijkstra, Feike A.
University of Sydney
Data from: Quantifying and reducing uncertainties in estimated soil CO2
fluxes with hierarchical data-model integration
Dryad
dataset
2017
diffusion model
soil carbon
soil respiration
2007-2011
global change experiments
soil CO2 flux
2017-11-01T00:00:00Z
en
https://doi.org/10.1002/2016jg003385
1758490 bytes
1
CC0 1.0 Universal (CC0 1.0) Public Domain Dedication
Non-steady state chambers are often employed to measure soil CO2 fluxes.
CO2 concentrations (C) in the headspace are sampled at different times
(t), and fluxes (f) are calculated from regressions of C versus t based a
limited number of observations. Variability in the data can lead to poor
fits and unreliable f estimates; groups with too few observations or poor
fits are often discarded, resulting in “missing” f values. We solve these
problems by fitting linear (steady state) and non-linear (non-steady
state, diffusion based) models of C versus t, within in a hierarchical
Bayesian framework. Data are from the Prairie Heating and CO2 Enrichment
(PHACE) study that manipulated atmospheric CO2, temperature, soil
moisture, and vegetation. CO2 was collected from static chambers bi-weekly
during five growing seasons, resulting in >12,000 samples and
>3100 groups and associated fluxes. We compare f estimates based on
non-hierarchical and hierarchical Bayesian (B vs HB) versions of the
linear and diffusion-based (L vs D) models, resulting in four different
models (BL, BD, HBL, HBD). Three models fit the data exceptionally well
(R2 ≥ 0.98), but the BD model was inferior (R2 = 0.87). The
non-hierarchical models (BL, BD) produced highly uncertain f estimates f
(wide 95% CIs), whereas the hierarchical models (HBL, HBD) produced very
precise estimates. Of the hierarchical versions, the linear model (HBL)
underestimated f by ~33% relative to the non-steady state model (HBD). The
hierarchical models offer improvements upon traditional non-hierarchical
approaches to estimating f, and we provide example code for the models.
Dataset1_HBDDataset1_HBD contains individual observations of CO2 and time,
along with categorical or integer identifiers for day, vegetation type,
plot, session (group). DayN = day since 1/1/2006; Datecode = unique
integer ID for each measurement date; Min = minutes after chamber closure;
Veg = vegetation treatment ID (1 = vegetated; 2 = herbicide); Plot = plot
ID; Group = chamber session ID; CorrCO2 = corrected CO2 concentration
(ppm); SoilT = soil temperature (celcius) in/near temperature in surface
layer.Read_me_HBD.pdfDataset2a_HBDDataset2a_HBD contains the integer
identifiers for date, vegetation type, and plot associated with each
chamber session; dateID, vegID, and PlotID are the same as Datecode, Veg,
and Plot in Dataset1_HBD, but each row in Dataset2a_HBD corresponds to the
chamber session ID (i.e., Group in
Dataset1_HBD).Read_me_HBD.pdfDataset2b_HBDDataset2b_HBD contains various
constants and sample size information required for the the HBD
(hierarchical Bayesian diffusion model). This includes Tmtcode, which is a
list of 30 values giving the treatment ID associated with each of the 30
plots (ordered by plot #), where 1 = ct, 2 = cT, 3 = Ct, 4 = CT, 5 = ctd,
and 6 = cts, for c = ambient CO2, C = elevated CO2, t = ambient
temperature, T= warmed, d = deep irritation, and s = shallow irrigation.
This dataset also contains Ctrt, denoting the CO2 treatment level
associated with each of the 6 treatments; 1 = ambient CO2 and 2 = elevated
CO2.Read_me_HBD.pdfDataset3_HBDDataset3_HBD contains soil water content
data for every plot and date. SWC_DSM1[r,c] is the soil water content
associated with day = r and plot ID = c. Plot ID (columns) correspond to
Plot in Dataset1_HBD; r = 1 (row #1) corresponds to 12/21/2006 or day 355
from 1/1/2006, with rows ordered by date (no missing
days).Read_me_HBD.pdfDataset4_HBDDataset4_HBD contains atmospheric
pressure data for every date, with row #1 corresponding to 3/1/2007 (i.e.,
day 425 since 1/1/2006). Rows are ordered by day (no missing
dates).Read_me_HBD.pdfModelCode_HBD_SIOpenBUGS / JAGS model code that uses
Dataset1_HBD.txt, Dataset2a_HBD.txt, Dataset2b_HBD.txt, Dataset3_HBD.txt,
and Dataset4_HBD.txtRead_me_HBD.pdfInitials_HBDa_SIInitials (starting)
values for first MCMC chain, to be used with the OpenBUGS / JAGS model
code (ModelCode_HBD_SI.txt).Read_me_HBD.pdfInitials_HBDb_SIInitials
(starting) values for second MCMC chain, to be used with the OpenBUGS /
JAGS model code
(ModelCode_HBD_SI.txt).Read_me_HBD.pdfInitials_HBDc_SIInitials (starting)
values for third MCMC chain, to be used with the OpenBUGS / JAGS model
code (ModelCode_HBD_SI.txt).Read_me_HBD.pdfModelScript_HBD_SIOpenBUGS
script file for running the model (ModelCode_HBD_SI.txt) and calling the
corresponding datasets (Dataset1_HBD.txt, Dataset2a_HBD.txt,
Dataset2b_HBD.txt, Dataset3_HBD.txt, and Dataset4_HBD.txt) and initial /
starting values (Initials_HBDa_S1.txt, Initials_HBDb_S1.txt,
Initials_HBDc_S1.txt).Read_me_HBD.pdfDataset2b_SIDataset2b_SI.txt is very
similar to Dataset2b_HBD.txt and contains various constants, sample size
information, and treatment codes required to run the HBL model. See
Dataset2b_HBD.txt description for additional
details.Read_me_HBL.pdfModelScript_HBL_SIModelScript_HBL_SI.txt is the
OpenBUGS script file for running the HBL model (ModelCode_HBL_SI.txt),
reading the correct data files, and loading the initial values for three
MCMC chains (Initials_a … b_S1.txt). Note that the script calls
Dataset1_SI.txt and Dataset2a_SI.txt, which are not provided because they
are exactly the same as the data files called Dataset1_HBD.txt and
Dataset2a_HBD.txt, respectively. In particular, Dataset1_SI.txt can be
created by grabbing for following columns (variables) from
Dataset1_HBD.txt: Datecode, Min, Veg, Plot, Group, and CorrCO2.
Dataset2a_SI.txt can be created by simply making a copy of
Dataset2a_HBD.txt.Read_me_HBL.pdfModelCode_HBL_SIModelCode_HBL_SI.txt is
the OpenBUGS model code for implementing the HBL
model.Read_me_HBL.pdfInitials_a_SIInitials_a_S1.txt contains initial
(starting) values for parameters in the HBL model, for chain
1.Read_me_HBL.pdfInitials_b_SIInitials_b_S1.txt contains initial
(starting) values for parameters in the HBL model, for chain
2.Read_me_HBL.pdfInitials_c_SIInitials_c_S1.txt contains initial
(starting) values for parameters in the HBL model, for chain
3.Read_me_HBL.pdf
shortgrass steppe
southeast Wyoming