#---------------------#
# General information #
#---------------------#

Atmospherically Driven Seasonal and Interannual Variability in the Lagrangian Transport Time Scales of a Multiple-inlet Coastal System
Authors:
 J.M. Fajardo-Urbina
 M. Duran-Matute
 T. Gerkema
 H. Clercx 
 U. Grawe
 G. Arts

Corresponding author:
  M. Duran-Matute (m.duran.matute@tue.nl) 
  Fluids and Flows group, Department of Applied Physics, Eindhoven University of Technology
  P.O. Box 513, 5600 MB Eindhoven, The Netherlands.


#--------------#
# Introduction #
#--------------#

The data in this repository can be used for reproducing the figures from the manuscript "Atmospherically Driven Seasonal and Interannual Variability in the Lagrangian Transport Time Scales of a Multiple-inlet Coastal System".

This data was used to:
 - Explore and understand the seasonality and interannual variability of the Lagrangian Transport Time Scales (LTTS), in particular the resindence and exposure times.
 - Understand the impact of the local wind on the LTTS.
 - Study the role of the large-scale atmospheric circulation and patterns on the LTTS.

The data is also storage at the GitHub repository (https://github.com/JeancarloFU/paper_Winds_AtmPatterns_Seasonal_Interannual_TTS_MultipleInlet). In this repository, scripts and notebooks (based on Python v3.8) used to reproduce the figures of this study are archived.


#----------------#
# Data structure #
#----------------#

There are in total 9 NetCDF files which can be used to reproduce Figure2 to Figure10 from the main manuscript.

------
Fig2a-d requires the file: 
 * maps_climatology_Tr_Te.nc 
This file has the following data:

 - Dimensions: yc: 115, xc: 286, yhc: 486, xhc: 820, np_dws: 1797, xy_dws: 2
 - Coordinates: 
    x(yc, xc): x-position (Easting) for the residence and exposure times (km). 
    y(yc, xc): y-position (Northing) for the residence and exposure times (km). 
    xh(yhc, xhc): x-position (Easting) for the bathymetry (km). 
    yh(yhc, xhc): y-position (Northing) for the bathymetry (km). 
 - Variables:
    Tr_autwin(yc, xc): mean autumn-winter(Sep-Feb) residence time (days).
    Tr_sprsum(yc, xc): mean spring-summer(Mar-Aug) residence time (days).
    Te_autwin(yc, xc): mean autumn-winter(Sep-Feb) exposure time (days).
    Te_sprsum(yc, xc): mean spring-summer(Mar-Aug) exposure time (days).
    h(yhc, xhc): bathymetry (m)
    bdr_dws(np_dws, xy_dws): xy coordinates of the boundary of the DWS (km).

------
Fig3 requires the file: 
 * annual_cycle_Tr_Te_E.nc
This file has the following data:

 - Dimensions: time: 872, time_E = 872
 - Coordinates: 
    time(time): dates of deployments in datetime64 format with a resolution of M2=44714s.
 - Variables:
    time_E(time_E): dates for the 15-day-mean wind energy.
    Tr_annual(time): annual cycle of the spatially-averaged residence time T_hat (days).
    Te_annual(time): annual cycle of the spatially-averaged exposure time T_hat (days).
    E_annual(time): annual cycle of the sum of the 15-day-mean wind energy of the dominant sectors W+SW+S (MJ).

------
Fig4 and Fig5 require the files:   
 * spatially-averaged_15-day-mean_Tr_Te.nc
This file has the following data:

 - Dimensions: time: 872
 - Coordinates: 
    time(time): time in datetime64 format with a resolution of 15 days.
 - Variables:
    Tr_hat(time): spatially-averaged 15-day-mean residence time (days).
    Te_hat(time): spatially-averaged 15-day-mean exposure time (days).

------
Fig6 requires the files:   
 * wind_based_model_Tr.nc
 * wind_based_model_Te.nc
These files have the following data:

 - Dimensions: time: 872
 - Coordinates: 
    time(time): time in datetime64 format with a resolution of 15 days.
 - Variables:
    Tr_tilde(time): half-year low-pass filter of the spatially-averaged 15-day-mean residence time (days).
    Tr_tilde_windmodel(time): Tr_tilde predicted with the wind-based model (days).
    Tr_tilde_windmodel_lag(time): Tr_tilde predicted with the wind-based model with 15-day lag (days).
    Te_tilde(time): half-year low-pass filter of the spatially-averaged 15-day-mean exposure time (days).
    Te_tilde_windmodel(time): Te_tilde predicted with the wind-based model (days).
    Te_tilde_windmodel_lag(time): Te_tilde predicted with the wind-based model with 15-day lag (days).
    E_tilde(time): Sum of the half-year low-pass filter of the 15-day-mean wind energy of the dominant sectors W+SW+S (MJ).

------
Fig7 requires the file: 
 * ncep_ncar_climatology_slp_wind.nc
This file has the following data:

 - Dimensions: lat: 29, lon = 53, season = 2
 - Coordinates: 
    lat(lat): Latitude (degrees north, from 10N-80N).
    lon(lon): Latitude (degrees, from 80W-50E).
    season(season): mean autumn-winter and mean spring-summer seasons.
 - Variables:
    slp(season, lat, lon): sea level pressure (hPa).
    u10(season, lat, lon): zonal wind at 10m (m/s).
    v10(season, lat, lon): meridional wind at 10m (m/s).

------
Fig8 requires the file: 
 * ncep_ncar_monthly_eofs_pcs_nao_eap_scan_from_detrend_deseasonalized_mslp.nc.nc 
This file has the following data:

 - Dimensions: time: 792, mode: 3, lat: 21, lon: 53
 - Coordinates: 
    time(time): time in datetime64 format with a monthly resolution.
    mode(mode): Empirical Ortogonal Functions EOFs mode number.
    lat(lat): Latitude (degrees north, from 30N-80N).
    lon(lon): Latitude (degrees, from 80W-50E).
 - Variables:
    pc(time, mode): Principal Components PCs, data is standardized (dimensionless).
    eof(mode, lat, lon): EOFs (hPa).
    var_exp(mode): variance explained per mode.
    ug_from_eof(mode, lat, lon): zonal geostrophic wind computed from eof (m/s).
    vg_from_eof(mode, lat, lon): meridional geostrophic wind computed from eof (m/s).
 - Attributes:
    general_info: 1) Analysis from monthly ncep-ncar 1950-2015. 2) Data was detrended and deseasonalized before getting EOFs and PCs.
    mode : 0=NAO, 1=EAP, 2=SCAN

------
Fig9 and Fig10 require the files:   
 * large_scale_model_Tr.nc
 * large_scale_model_Te.nc
These files have the following data:

 - Dimensions: time: 872, predictors: 3
 - Coordinates: 
    time(time): time in datetime64 format with a resolution of 15 days.
    predictors(predictors): object with names of the PCs 'SCAN', 'NAO', 'EAP'.
 - Variables:
    Tr_tilde(time): half-year low-pass filter of the spatially-averaged 15-day-mean residence time (days).
    Tr_tilde_seasonal(time): annual cycle of Tr_tilde (days).
    Tr_tilde_pcs_model(time, predictors): deseasonalized Tr_tilde predicted with each component of the PCs model (days).
    Tr_tilde_pcs_model_lag(time, predictors): deseasonalized Tr_tilde predicted with each component of the PCs model with 15-day lag (days).
    Te_tilde(time): half-year low-pass filter of the spatially-averaged 15-day-mean exposure time (days).
    Te_tilde_seasonal(time): annual cycle of Te_tilde (days).
    Te_tilde_pcs_model(time, predictors): deseasonalized Te_tilde predicted with each component of the PCs model (days).
    Te_tilde_pcs_model_lag(time, predictors): deseasonalized Te_tilde predicted with each component of the PCs model with 15-day lag (days).


#--------------------------------------#
# Information about raw numerical data #
#--------------------------------------#

The 9 netCDF files used to generate the figures, were obtained from the following raw data:

* Eulerian data from the GETM/GOTM model, and its set-up is described in:
    - Duran-Matute et al. (2014): https://doi.org/10.5194/os-10-611-2014
    - Grawe et al. (2016): https://doi.org/10.1002/2016JC011655
* Lagrangian data from the model Parcels v2.1.1, which can be installed from: 
    - https://anaconda.org/conda-forge/parcels
    - https://oceanparcels.org
* Monthly mean sea level pressure from NCEP-NCAR Reanalysis 1 (used for the large-scale analytical model):
    - https://psl.noaa.gov/data/gridded/data.ncep.reanalysis.html

