This Francis_Analysis_Readme.txt file was generated in November 2019 by Jordi Francis Clar
-------------------
GENERAL INFORMATION
-------------------
Title of the script: FrancisAnex2019_analysis.R
Author Information:
Author:Jordi Francis Clar
Email: francisclar@wisc.edu
University: University of Wisconsin - Madison
Departament: Biological Systems Engineering
Address: 115 Agricultural Engineering Building, 460 Henry Mall, Madison, WI 53706
Principal Investigator: Dr. Robert P Anex
Associate or Co-investigator: Jordi Francis Clar
Alternate Contact(s):
Funding sources:
This material is based upon work that was supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture, Hatch projects under accession numbers 1001805 and 1009785
--------------------------
SHARING/ACCESS INFORMATION
--------------------------
Licenses/restrictions placed on the data, or limitations of reuse:
Recommended citation for the data: 'Measuring frequently during peak soil N2O emissions is more important than choosing the time of day to sample'.
Francis Clar, J. and Anex, R. P.: Measuring frequently during peak soil N2O emissions is more important than choosing the time of day to sample, doi:https://doi.org/10.5194/bg-2019-396
Citation for and links to publications that cite or use the data: 'Measuring frequently during peak soil N2O emissions is more important than choosing the time of day to sample'.
Francis Clar, J. and Anex, R. P.: Measuring frequently during peak soil N2O emissions is more important than choosing the time of day to sample,
doi: https://doi.org/10.5194/bg-2019-396
--------------------
CODE OVERVIEW
--------------------
File list (1 file): FrancisAnex2019_Analysis.R
Script format: .R (native file system of RStudio)
Sorftare:
RStudio Team (2018). RStudio: Integrated Development for R. RStudio, Inc., Boston, MA URL:http://www.rstudio.com/.
R version 3.5.3 (2019-03-11) -- "Great Truth"
Copyright (C) 2019 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (64-bit)
Script file size: 27 KB
--------------------------
METHODOLOGICAL INFORMATION
--------------------------
General: the script uses linear regression (Ordinary Least Squares) to study the relationship between the starting time of a short flux measurment (hourly.flux) and
the mean daily flux (mean.daily.flux). The goal of the analysis is to determine if there is a time of the day at which the flux measured is not significantly different
form the mean daily flux.
The code runs the same analysis at differnt levels of standarized flux intensity (i.e. cumulative contribution).
Methods for processing the data: 'Measuring frequently during peak soil N2O emissions is more important than choosing the time of day to sample'.
Francis Clar, J. and Anex, R. P.: Measuring frequently during peak soil N2O emissions is more important than choosing the time of day to sample,
doi: https://doi.org/10.5194/bg-2019-396
Description of methods used for analyzing the data: See materials and methods section of 'Measuring frequently during peak soil N2O emissions is more important than choosing the time of day to sample'.
Francis Clar, J. and Anex, R. P.: Measuring frequently during peak soil N2O emissions is more important than choosing the time of day to sample,
doi: https://doi.org/10.5194/bg-2019-396
--------------------------
CODE DESCRIPTION
--------------------------
Input data: FrancisAnex2019_FluxData.csv (can be found in this same repository)
Output data (in publication):Figures 1 and 2
Output data (usable as supplementary data): Summaries of the regression (e.g.'SummaryRegression_top50') analysis between houlry and daily mean fluxes
for each hour (0-23) and level of cumulative contribution (i.e HCC 75%, HCC 50%, HCC 30% and, LCC 50%).
Each summary data frame contains 24 rows (i.e. 0-23 hours) and 11 columns, these are:
hour: hour at which the flux measurment started
Neg_bound: distance from the regression estimate (beta) to the lower bound of the 95% confidence interval
Pos_bound: distance from the regression estimate (beta) to the upper bound of the 95% confidence interval
term: regression coefficient between the log10_hourly.flux and the log10_daily.mean.flux, refered in the text as beta coefficient.
estimate: value of the regression coefficient
std.error = Residual Standard Error divided by the square root of the sum of the square of the estimate.
statistic: value of the F-Statistic computed as ((SSyy-SSE)/k) / (SSE/(n-(k+1)))
p.value: P_value (significance level) attributed to the F-Statistic, teting that there is a linear relationship between the independet and the response variable, Ho:no relationship between log10_hourly.flux and log10_mean.daily.flux
pvalueDiff1: Significance level of the coefficient regression being equal one, ho: coefficient regression is equal to one (hourly flux measurment is equal to mean daily flux)
r.squared: coefficient of determination between dependent (log10_hourly.flux) and response (log10_mean.daily.flux) variables
n: number of data points included in the analysis
--------------------------
CODE-SPECIFIC INFORMATION
--------------------------
Packages needed:
Package Version
grid "grid" "3.5.3"
gridExtra "gridExtra" "2.3"
car "car" "3.0-3"
carData "carData" "3.0-2"
reshape "reshape" "0.8.8"
mlr "mlr" "2.15.0"
ParamHelpers "ParamHelpers" "1.12"
MESS "MESS" "0.5.5"
geeM "geeM" "0.10.1"
Matrix "Matrix" "1.2-17"
geepack "geepack" "1.2-1"
dplyr "dplyr" "0.8.3"
broom "broom" "0.5.2"
devtools "devtools" "2.1.0"
usethis "usethis" "1.5.1"
lubridate "lubridate" "1.7.4"
magrittr "magrittr" "1.5"
ggplot2 "ggplot2" "3.2.1"
tidyr "tidyr" "0.8.3"
ascii "ascii" "2.1"
stats "stats" "3.5.3"
graphics "graphics" "3.5.3"
grDevices "grDevices" "3.5.3"
utils "utils" "3.5.3"
datasets "datasets" "3.5.3"
methods "methods" "3.5.3"
base "base" "3.5.3"