This repository is home to the data reconciliation of the Gulf of Maine CPR obtained from NOAA Fisheries (NEFSC) and the Sir Alister Hardy Foundation (SAHFOS), as well as analyses looking at patterns in abundance anomalies.
Zooplankton analyses contained here focus on identifying changes in plankton abundance over time, and their relationship with physical properties of the environment.
The analysis pipeline has been documented using the {targets} package and can be seen in the _targets.R file.
Full documentation of the data provenance for this project have been made available on github using bookdown and github pages here:
Various rendered reports have been made available using github pages. Links to the various reports can be found at the index page
README.md - Introduction to the project and project navigation
index.md - Quick links to reports, presentations, & web apps
R - Folder containing R code for data cleanup and analyses
matlab - Folder with matlab CalProc routine for estimating seasonal splines
00_cprdata_first_look.R: Primary investigation of bi-monthly cpr anomaly dataset
00b_cprdata_first_look.R: Primary investigation of bi-monthly cpr anomaly dataset
01_cpr_eda.R: Exploratory data analysis of bi-monthly cpr data
02_cpr_pca.R: Recreation of figures from Pershing et al. 2005 PCA analysis
02b_cpr_sst_pca.R: Incorporation of bi-monthly SST data with CPR dataset. Creation of processed CPR and lagged SST datasets used in later analyses: cpr_with_SSTlags.csv, SST_with_lags.csv
02c_quarterly_cpr_sst_regressions.R: Regression analyses between quarterly CPR and SST datasets with and without period-lags. Creation of processed quarterly CPR+SST dataset: quarterly_cpr_sst.csv
03_regime_shifts.R: Identification of regime shifts in bi-monthly datasets
03b_bestelmeyer2011_code.R: Threshold detection code from Bestelmeyer et al. 2011
03c_Bestel_Test.R: Test of Bestelmeyer et al. 2011's methods
04_bimonthly_sst_corrplots.R: Correlation plots for bi-monthly CPR and SST data. Full timeseries and late-period PCA timelines following Pershing et al. 2005 methods
04b_quarterly_sst_corrplots.R: Quarterly correlation plots of CPR and SST data
05_cluster_analysis.R: k-means clustering analysis of bi-monthly CPR and SST data
05_cpr_clustering.Rmd: Rmarkdown document synthesizing research steps taken up to this point
05_cpr_clustering.html: Rendered output of Rmarkdown document
06_buoy_data_mgmt.R: Data management steps for synthesizing NERACOOS buoy data. Creation of buoys_aggregated.csv & buoys_daily.csv
07_buoy_regressions.R: Regression analysis of quarterly bouy aggregate data and CPR data. Creation of cpr_quarters_buoys.csv
08_buoy_anomaly_calculations.R: Calculation of daily buoy anomalies for temperature and salinity. Creation of buoy_anomalies.csv
09_buoy_quarterly_pca.R: Quarterly buoy aggregate corrplots and PCA analysis
10_buoy_daily_interpolations.R: Estimation of interpolated values for daily NA values in the buoy sensor data. Export of the non-interpolated data used to generate daily buoy PCA buoy_pcadat_raw.csv & the interpolated dataset resulting from multiple-regression interpolation buoy_pcadat_interpolated.csv
11_buoy_daily_pca.R: PCA on daily buoy sensor measurements, then applied to interpolated buoy dataset
12_paper_layout.Rmd: Consolidation report of analyses run thus far
12_paper_layout.html: Rendered output of Rmarkdown document
13_sahfos_cpr_cleanup.R: Reads in the full CPR datasets from NOAA and from SAHFOS. Splits both into a key created from the header space, and the abundances/concentrations by station.
14_taxa_check_heatplots.R: Visualization tools for determining which taxa-stage groupings are in use and how that changes over time.
15_NOAA_CPR_Cleanup.R: Script for reducing the NOAA dataset down to groupings that are more similar to the SAHFOS data, drops taxa that are not ever observed.
16_SAHFOS_CPR_Cleanup.R: Partner script to the NOAA CPR Key. Converts SAHFOS data from concentration per 10cm silk mesh to concentration per
17_noaa_sahfos_eda.R: Exploratory Data Analysis of the NOAA/SAHFOS data, post-cleanup.
18_karen_s_analysis_funs.R: Re-coding of MATLAB code obtained from Karen Stamieszkin. Splits taxa proportionally into contributing stages. Determines the predicted length for each stage at a given temperature.
19_cpr_splines.R: Seasonal abundance anomaly calculations for combined NOAA/SAHFOS CPR data
cpr_buoy_DE: Shiny App Files for exploring correlations between buoy sensor data and quarterly CPR data
cpr_helper_funs.R: Shortcut to commonly used functions
NAO_import: Code for importing north atlantic oscillation data
new_anom_analyses: Scripts for refined workflow using NOAA/SAHFOS combined dataset
presentations: Code & images for xaringan presentations
stylesheets: CSS stylesheets
01_new_anoms_eda.R: Verificaation that new anomalies resemble matlab anomaly routine with exploratory analyses.
02_new_anoms_pca.R: Recreation of PCA analyses using R routine anomaly data
03_new_anoms_quarterly_sst.R: Quarterly SST anomaly regression figures for all taxa in new dataset
04_new_anoms_sst_corrplots.R: Correlograms between CPR taxa anomalies and SST anomalies
05_new_anoms_buoys.R: Quarterly buoy sensor data correlations
figures: Figure recreations created using the new R anomaly routine from the combined NOAA/SAHFOS dataset