AHW

The scripts and data used to determine common air-sea states during MHWs along the coast of South Africa.

File description

• The raw text of the paper may be found in "LaTeX/Schlegel_et_al.tex"
• The code used to calculate the MHWs and fit the SOM may be found in "2.Model_fitting.R"
• The code used to create the figures seen in the paper may be found in "3.Figures.R"
• The code used to create the tables seen in the paper may be found in "4.Tables.R"
• The code used to generate the specific values given in the text may be found in "5.Text.R"
• The "data" folder contains the processed data used in the analyses as well as the output of "2.Model_fitting.R"
• The "func" folder contains custom made functions used in the analyses
• The "graph" folder contains the files necessary to create maps as well as the output of "3.Figures.R"
• The "setupParams" folder contains the SACTN site list and analysis period data
• The "LaTeX" folder contains the files used to compile "/LaTeX/...", the body of the text
• The "old" folder contains scripts and data from previous versions of the analysis/ methodology

2016/09/18

• All for loops removed from extreme event calculation work flow
• A for loop is still used to "grow" the SAWS data, allowing for it to be compared against each SACTN time series
• All possible extreme events combinations etc. have been calculated as per current agreement on methodology

2016/09/19

• All for loops removed from the co-occurrence calculations
• Co-occurrence calculated for all combinations of SACTN and SAWS sites including tmean, tmin and tmax

2016/09/20

• Corrected error in calculation of co-occurrence proportions
• Produced a medley of initial figures to visualise first round of results

2016/09/28

• Removed for loop used to "grow" the SAWS data
• All extreme events re-calculated with same results

2016/09/29

• Calculated extreme events for SAWS time series based on a static analysis period (1981 - 2010)
• Changed file naming convention for extreme event results to differentiate the results calculated with the static period
• Created function that calculates distance AND bearing between any two sites
• Calculated distance and bearing between all possible sites within and among datasets

2016/09/30

• Improved co-occurrence workflow to allow for calculation of extreme events between AND within datasets
• Expanded results of co-occurrence analysis to also include the statistics of the co-occurring events
• No longer constraining co-occurrence rates by set periods of days
• Now the difference in days between the latest occurring events is recorded

2016/10/03

• Calculated latest occurring events between AND within all sites AND datasets
• Added distance, bearing and coastal section metrics for compared sites
• Created a .Rmd file to explain the co-occurrence step of the work-flow

2016/10/18

• Created figures for the largest (int_cum) heat wave and cold spell for each time series
• Explored relationships between SAWS and SACTN heat waves occurring within 7 and 2 days of one another

2016/10/20

• A small bug was detected in the original SACTN data set
• Improved screening methods were implemented at the source to smooth this issue
• These improved methods were applied to the load step in this project as well
• All scripts were re-run with the newly screened data and all calculations/ figures compiled anew
• The final co-occurrence rates are ever so slightly higher than previously

2016/10/22

• Fix issues in which missing SACTN values were saved as NaN instead of NA
• This had no effect on the calculations

2016/10/31

• Wind data for Port Nolloth added to project
• Function created that can calculate mean daily wind vectors
• First draft of a figure that shows wind vectors with temperatures

2016/11/01

• Worked out a couple of preliminary techniques for how best to visualise air, sea and wind data together during co-occurrence

2016/11/02

• The preliminary round of air, sea, wind figures

2016/11/29

• Updated exploratory analyses
• More work done with wind data
• Two new reports written up and knitted

2017/02/07

• Script created to download BRAN data
• These data are saved in an external folder as they are to multitudinous for GitHub
• First steps taken for the creation of the LaTeX version of the manuscript

2017/02/08

• Function created to load BRAN data into R in long format
• Additional text added to manuscript

2017/02/13

• Function created to load ERA Interim data into R in long format
• First proof of concept figure for synoptic air-sea view during events created

2017/02/14

• Reanalysis data processing pipeline improved
• Wind vectors visualised

2017/02/17

• Summary data boxes added to synoptic air-sea figure
• Rough draft of introduction complete

2017/02/20

• Outline of synoptic air-sea figure complete

2017/02/23

• Calculation of daily climatologies for air-sea state
• The data frames are not stored here as they are too large

2017/02/23

• Temperature anomalies added to synoptic figure

2017/02/25

• Current and wind anomalies added

2017/02/28

• Daily climatologies for girded reanalysis temperatures have now all been smoothed
• This allows for smoother temperature anomaly values to be calculated

2017/03/01

• Cosmetic adjustments to synoptic figure

2017/03/01

• Synoptic figure output now also saves the temp and uv values in a separate list for SOMs

2017/03/02

• All synoptic figures created

2017/03/08

• Proof of concept for SOMs

2017/03/09

• SOM code automated
• Initial results visualised

2017/03/10

• Seasonality added to output
• Individual SOMs run for each coastal section

2017/03/13

• All functions created for SOM work flow moved to a separate script
• Alternative rescaling function for SOM results added

2017/03/16

• Ran hierarchical cluster analysis for air-sea states during MHWs
• Began investigation into clustering of air-sea states for daily climatologies

2017/03/16

• Hierarchical cluster analysis for air-sea state during daily climatologies

2017/03/30

• First step towards running the three agreed upon cluster analyses on all data

2017/04/03

• More work towards the cluster analyses

2017/04/08

• More work towards the cluster analyses

2017/04/10

• All variables that could affect the results of a clustering technique have been analysed

2017/04/11

• Updates to cluster results code

2017/05/11

• Now using principal component initialisation (PCI) rather than random initialisation (RI) for SOMs
• All SOM modelling in "LaTeX/cluster_results.Rmd" updated accordingly

2017/05/12

• "LaTeX/cluster_results.Rmd" completed

2017/05/15

• Additions to methods section

2017/05/16

• Edits to intro and methods sections
• Additions to methods and results sections
• Began cleaning workflow for publication

2017/05/17

• "1.Data_assembly.R" complete
• Removed data from GitHub not cleared for public access
• Began moving unused scripts and data to the "old" folder

2017/05/18

• Began "2.Model_assembly.R"
• More files moved to "old" folder

2017/05/19

• Re-created data packets for use with SOMs and synoptic figures

2017/05/22

• Smoothing out Git push issues
• Complete documentation provided for all files

2017/05/23

• More modeling, figures, and one table
• Additional intro, methods, and results edits

2017/05/24

• References for clustering added to methods section
• More results written up

2017/06/05

• Writing for results, discussion and conclusion added

2017/06/08

• Table 2 modified
• Two new figures included
• Additional writing in manuscript

2017/06/09

• Minor edits
• First draft complete

2017/06/13

• First round of co-author edits

2017/06/15

• More literature added in support of the introduction

2017/07/11

• Additional co-author comments incorporated into text

2017/07/18

• Code for Figure 1 added

2017/07/19

• All comments for the Introduction section incorporated

2017/07/20

• Progress made on methods section

2017/07/21

• Comments for methods section wrapped up

2017/07/26

• Phasing out BRAN in favour of AVISO and OISST
• Recalculated ERA daily clims based on new longer period
• Calculated OISST daily clims

2017/07/27

• More phasing out of BRAN
• Calculated AVISO daily clims

2017/07/28

• Analysis with new OISST+AVISO data complete
• New data packets created
• New synoptic atlas figures created
• New SOM figures created

2017/07/30

• New colour palette used for SOM nodes
• SOM nodes described in the results

2017/07/31

• Editing of results and discussion sections

2017/08/01

• Calculated MDS and HCA with new data
• Created biplot and dendrogram
• Edits to figure 1
• Some edits to text

2017/08/03

• More edits to text

2017/08/04

• Notes on atmosphere

2017/08/05

• Edits throughout texts

2017/08/06

• Second draft ready for co-author revisions

2017/08/07

• Figure tweaks
• Some few revisions from co-authors

2017/08/08

• More text edits

2017/08/10

• A few edits

2017/08/11

• AJ's direct .tex edits
• Tweaks to most figures

2017/08/013

• All coments on figures incorporated
• All coments not requiring re-writes incorporated

2017/08/014

• All minor repetitive text changes made
• LaTeX file converted to Frontiers standard

2017/08/015

• All final co-authors comments incorporated
• Several minor tweaks
• Manuscript ready for submission
• Created word document for abstract
• Created word document for cover letter
• Created the required 300dpi .tiff files for figures

2017/09/015

• Most of round one reviewer comments addressed

2017/09/017

• Several edits from AJ
• More reviewer comments addressed

2017/09/019

• All reviewer comments addressed
• Manuscript re-submitted

2017/10/03

• Minor tweaks for author proofs