7 Exercises
7.1 Fitting Ac-Ci Curves
We are going to follow along with a Vignette that explains how to fit response curves using R
We’ll use PEcAn - you’ll use PEcAn next week to do some data assimilation
We will also look at another package written by Duursma
7.1.1 Setup
7.1.1.1 JAGS
JAGS is a program for analysis of Bayesian hierarchical models
To install go to https://sourceforge.net/projects/mcmc-jags/files/JAGS/4.x/ and select the version appropriate for your operating system.
There may be a link pointing directly to the newest version so you don’t have to look through the file structure.
7.1.1.2 The PEcAn Photosynthesis Package
##### Required for PEcAN
# install packages
install.packages("devtools")
install.packages("rjags")
# load the packages
library(devtools)
library(rjags)# Next Install PEcAN.photosynthesis ... this is not available through CRAN
# - but you can find it on GITHUB - this code will download and install
### This step might take a while
#if (!require("PEcAn.photosynthesis", character.only = TRUE)) {
devtools::install_github("PecanProject/pecan/modules/photosynthesis",
ref = "release/1.5.0")
#}
knitr::opts_chunk$set(cache = TRUE)
library(PEcAn.photosynthesis)7.1.1.3 The plantecophys package
- Modelling and Analysis of Leaf Gas Exchange Data
- Coupled leaf gas exchange model, A-Ci curve simulation and fitting,
- Ball-Berry stomatal conductance models, leaf energy balance using Penman-Monteith,
- Cowan-Farquhar optimization, humidity unit conversions.
https://cran.r-project.org/web/packages/plantecophys/index.html
# IN CASE YOU JUST ABOLUTELY HAVE TO HAVE THE LATEST VERSION
# i nstall_bitbucket("remkoduursma/plantecophys")
# install the package
install.packages("plantecophys")
# load the package
library (plantecophys)
# REFERENCE MANUAL
https://cran.r-project.org/web/packages/plantecophys/plantecophys.pdf
# Some utility packages# if you need to install ... then install them :)
# install.packages("dplyr")
library(dplyr)
# install.packages("tidyr")
library(tidyr)
# install.packages("ggplot2")
library(ggplot2)
# install.packages("grid")
library(grid) #required for 'unit'7.1.2 Loading data
7.1.2.1 Using pre-loaded data
With the PEcAn Photosynthesis package, you have downloaded the LICOR files collected by the FLUXCOURSE in 2012.
If you want to see the files you can go into your R directory on your computer and you can find it in the extdata folder within the PEcAN.photosynthesis package.
For example : C:\Users\dmoore\Documents\R\R-3.4.0\library\PEcAn.photosynthesis\extdata
The command system.file()` in R allows you to call up folders relevant to a particular package. This commant looks in the folder, scans for files with aci or aq in the filename and writes these names \(and their full path\) to the R object “filenames”.
## Get list of LI-COR 6400 file names (ASCII not xls)
filenames <- system.file("extdata", paste0("flux-course-",rep(1:6,each=2),c("aci","aq")), package = "PEcAn.photosynthesis")
# filenames is 12 character variables
# These files are licor files ... not CSV files, not XLSX files
# ... just the file that comes right off the machine ... it has no file extension
# PEcAN.photosynthesis contains a command call read.licor() ...I bet you can guess what it does!
# read.Licor will load an individual licor file ... if you have one.
## Load files to a list
master = lapply(filenames, read_Licor)7.1.2.2 Using your own data
You can download this year’s data as a ZIP file at https://github.com/Fluxcourse/2017_LI-COR
You can do the same process with your own data …. you just need to point R to those data and load em up into a big list.
my_data <- "path_to_your_data"
# for example: my_data <- "~/Downloads/2017_LI-COR-master/07102017PSC0365_CO2response"
master = read_Licor(my_data)7.1.3 QA/QC Checks
The code below performs a set of interactive QA/QC checks on the LI-COR data that’s been loaded.
master[[1]] <- Licor_QC(master[[1]])If you have more than one file loaded into the `master` list, you may only want to run the function for one file, rather than looping over all the files.
master[[1]] <- Licor_QC(master[[1]])7.2 Stomatal Conductance Modeling Exercise:
Below is a dataset containing meteorological observations and tower-derived GPP estimates collected at MMSF during the course of a severe drought occurring in 2012.
The data can be downloaded here
DOY Hour VPD (kPa) SWP (Mpa) VWC (m3/m3) Par (umol/m2/s) CO2 (ppm) GPP (umol/m2/s)
161.00 13.00 1.78 -0.31 24.16 1581.51 378.88 31.17
162.00 13.00 0.81 -0.40 22.18 1039.02 393.49 26.87
163.00 13.00 2.19 -0.35 23.29 1981.02 391.32 14.06
164.00 13.00 1.65 -0.37 22.79 1959.24 390.29 28.01
165.00 13.00 2.28 -0.40 22.08 1975.11 386.57 21.81
166.00 13.00 2.61 -0.43 21.51 1964.63 388.04 17.56
167.00 13.00 2.32 -0.46 21.06 1921.93 384.74 20.38
168.00 13.00 1.34 -0.49 20.65 1620.87 392.81 20.75
169.00 13.00 2.08 -0.53 20.05 1912.18 385.83 22.01
170.00 13.00 2.28 -0.57 19.51 1942.83 389.73 19.74
171.00 13.00 2.51 -0.63 18.85 1817.14 390.25 19.51
172.00 13.00 2.51 -0.67 18.37 1718.25 388.63 20.61
173.00 13.00 2.37 -0.71 17.99 2004.02 390.08 17.98
174.00 13.00 2.53 -0.76 17.61 1597.01 384.12 26.28
175.00 13.00 2.72 -0.80 17.27 1151.81 382.42 22.52
176.00 13.00 2.40 -0.86 16.83 1974.86 389.35 17.76
177.00 13.00 2.02 -0.90 16.54 1992.69 389.01 12.18
178.00 13.00 2.58 -0.96 16.12 1976.11 386.41 16.60
179.00 13.00 3.77 -1.03 15.74 1945.17 386.67 7.43
180.00 13.00 3.20 -1.08 15.48 1858.93 386.64 16.02
181.00 13.00 3.29 -1.12 15.26 1815.67 389.95 12.42
182.00 13.00 2.84 -1.15 15.00 1679.99 394.88 19.59
183.00 13.00 2.91 -1.21 14.83 1743.00 393.03 19.10
184.00 13.00 2.42 -1.24 14.66 1782.24 395.40 20.37
185.00 13.00 3.41 -1.29 14.47 1846.70 399.83 12.69
186.00 13.00 4.39 -1.34 14.26 1653.94 397.45 11.65
187.00 13.00 4.14 -1.38 14.09 1599.68 392.84 8.08
188.00 13.00 4.19 -1.41 13.99 726.34 392.89 7.05
189.00 13.00 2.08 -1.41 14.00 183.91 402.59 NaN
190.00 13.00 2.65 -1.45 13.84 1751.20 383.80 17.21
191.00 13.00 3.23 -1.51 13.64 1964.57 385.53 13.61
192.00 13.00 3.03 -1.54 13.54 2000.73 384.07 16.73
193.00 13.00 3.13 -1.59 13.37 1614.61 384.63 11.51
194.00 13.00 1.91 -1.60 13.35 1801.13 396.36 23.55
195.00 13.00 0.89 -1.58 13.40 1973.77 404.49 21.15
196.00 13.00 1.96 -1.62 13.29 1807.26 398.74 14.46
197.00 13.00 2.14 -1.65 13.19 832.07 400.72 9.93
198.00 13.00 2.52 -1.70 13.05 1703.59 398.55 12.63
199.00 13.00 2.94 -1.70 13.04 1531.01 397.72 11.89For the purposes of this exercise, the data are limited to midday \(i.e. hour 1300\) values. Use the data to generate estimates of stomatal conductance using the Leuning model, and plant hydraulic model, and the Medlyn optimality model. As a reminder, here are the model forms, and some suggested parameter values.
Use the data to generate estimates of stomatal conductance using the Leuning model, and plant hydraulic model, and the Medlyn optimality model. As a reminder, here are the model forms, and some suggested parameter values:
7.2.1 Leuning
\[ g_s = \frac{m_2 A}{s_s - \Gamma}(1 + \frac{D}{D_o})^{-1} + b_2 \]
Let:
\[D_o\] = 1.1 kPa
\[b_2\] = .001 mol/m2/s
\[m_2\] = 6.5 dimensionless
\[\Gamma\] = 50 ppm
7.2.2 Hydraulic Model
\[g_s = \frac{K(\Psi_s - \Psi_L -pgh)}{VPD}\]
Let:
\[K\] = 0.3 mol/m2/s
\[pgh\] = 0.3 \(appropriate for a 30-m-tall tree\)
Assume:
constant \[\Psi_L\] = -1.8 MPa \(isohydric species\)
7.2.3 Optimality Model
\[g_s^* \approx g_0 + 1.6(1+ \frac{g_1}{\sqrt{D}})\frac{A}{C_a}\]
Let:
\[g_0\] = .001 mol/m2/s
\[g_1\] = 2