Work Description

Title: Rotary spectra of surface kinetic energy in drifters and high-resolution global ocean models Open Access Deposited

h
Attribute Value
Methodology
  • The data are rotary spectra computed from the HYCOM and MITgcm (LLC4320) simulations subsampled to 1/4 degree. FOr each model, time series are generated based on the zonal (u) and meridional velocities (u) for a full year. These time series are split into 11 segments of 60 days each overlapping by 50%. 1D discrete Fourier transform of u + iv is computed. The Fourier coefficients are mutliplied by their complex conjugates and averaged over spegments to produce spectra. These spectra are integrated over defined frequency bands to produce kinetic energy values.
Description
  • For Drifters, HYCOM, MITgcm: Spectra and kinetic energy files. Please see readme.txt for a description of all data and code contained here.

  • - Compare kinetic energies (KE) of high-resolution global ocean models estimated from rotary spectra to KE in surface drifter observations. - Near-inertial KE is closer to drifter observations in models with frequently updated wind forcing - Internal tide KE is closer to drifter observations in models with topographic wave drag
Creator
Depositor
  • jbrasch@umich.edu
Contact information
Discipline
Funding agency
  • National Aeronautics and Space Administration (NASA)
  • Department of Defense (DOD)
  • National Science Foundation (NSF)
ORSP grant number
  • This research was supported by funding from US National Science Foundation (NSF) grant OCE-1851164 as well as a Research Experience for Undergraduates Supplement to that grant. BKA also gratefully acknowledges funding from Office of Naval Research (ONR) grant N00014-18-1-2544 and NASA grants NNX16AH79G, NNX17AH55G, and 80NSSC20K1135.
Keyword
Date coverage
  • 2020-05-01 to 2020-08-31
Citations to related material
  • Elipot, S., Lumpkin, R., Perez, R. C., Lilly, J. M., Early, J. J., & Sykulski, A. M. (2016). A global surface drifter data set at hourly resolution. Journal of Geophysical Research: Oceans, 121(5), 2937–2966. https://doi.org/10.1002/2016JC011716
Related items in Deep Blue Documents
Resource type
Curation notes
Last modified
  • 11/18/2022
Published
  • 09/01/2020
Language
DOI
  • https://doi.org/10.7302/r8q1-g224
License
To Cite this Work:
Brasch, J. M., Elipot, S., Arbic, B. (2020). Rotary spectra of surface kinetic energy in drifters and high-resolution global ocean models [Data set], University of Michigan - Deep Blue Data. https://doi.org/10.7302/r8q1-g224

Relationships

This work is not a member of any user collections.

Files (Count: 7; Size: 40.7 GB)

Thumbnaildsfsdfsdf Title Original Upload Last Modified File Size Access Actions
readme.txt 2020-09-01 2020-09-01 7.51 KB Open Access
figures.zip 2020-08-13 2020-08-13 522 KB Open Access
ke.zip 2020-08-13 2020-08-13 4.71 KB Open Access
spectra.zip 2020-08-13 2020-08-13 12.2 KB Open Access
Drifters.zip 2020-09-01 2020-09-01 11.4 MB Open Access
HYCOM.zip 2020-09-01 2020-09-01 21.9 GB Open Access
MITgcm.zip 2020-09-01 2020-09-01 18.8 GB Open Access

Date: 24 August, 2020

Dataset Title: Factors Influencing Surface Kinetic Energy in Global High Resolution Ocean Models

Dataset Creators: Jonathan Brasch, Shane Elipot, Brian Arbic

Dataset Contacts: Jonathan Brasch jbrasch@umich.edu; Brian Arbic arbic@umich.edu

Funding: OCE-1851164 (NSF), N00014-18-1-2544 (ONR), NNX16AH79G (NASA), NNX17AH55G, (NASA), & 80NSSC20K1135 (NASA)

Key Points:
- We compare kinetic energies (KE) of high-resolution global ocean models estimated from rotary spectra to KE in surface drifter observations.
- Near-inertial KE is closer to drifter observations in models with frequently updated wind forcing
- Internal tide KE is closer to drifter observations in models with topographic wave drag

Methodology:
The data are rotary spectra computed from the HYCOM and MITgcm (LLC4320) simulations subsampled to 1/4 degree. FOr each model, time series are generated based on the zonal (u) and meridional velocities (u) for a full year. These time series are split into 11 segments of 60 days each overlapping by 50%. 1D discrete Fourier transform of u + iv is computed. The Fourier coefficients are mutliplied by their complex conjugates and averaged over spegments to produce spectra. These spectra are integrated over defined frequency bands to produce kinetic energy values.

Data-set information:
MITgcm model: https://data.nas.nasa.gov/ecco/data.php?dir=/eccodata/llc_4320
Drifter observations: https://www.aoml.noaa.gov/phod/gdp/hourly_data.php
Hycom model: HYCOM output can be accessed via the OSiRIS infrastructure, please contact for details on access.

Instrument and/or Software specifications: Python 3.7.6, Matlab

Files contained here:
The folders show divisions based on each simulation or observation analyzed. The folders and files are described below:

HYCOM: results from HYCOM subsampled to 1/4 degree

Rotary Spectra: spectra_hycom_0m.nc, spectra_hycom_15m.nc

X: horizontal grid spacing (1500x1)
Y: vertial grid spacing (1176x1)
ke_uv: spectra (1500x176x1440)
freq_time: spectral frequencies (1440x1)
freq_time_spacing: spacing between points in freq_time (1x1)
lat: model latitude in degrees north (1500x1176)
lon: model longitude in degrees east (1500x1176)

Rotary Spectra averaged to 1x1 degree bins: spectra_hycom_0m_bin.nc, spectra_hycom_15m_bin.nc

ke_uv_bin: spectra averaged to 1x1 degree bins (360x160x1440)
freq_time: spectral frequencies (1440x1)
lat: latitude range of bin-averaged data (360x1)
lon: longitide range of bin-averaged data (1x160)

Kinetic Energy Results: ke_hycom_15m.mat, ke_hycom_0m.mat

lat_grid: model latitude averaged to 1x1 degree bins (360x160)
lon_grid: model longitide averaged to 1x1 degree bins (360x160)
xmid: latitude at bin centers (160x1)
ymid: longitide at bin centers (360x1)

For each energy band:
[energy-band]: ke results on model grid (1500x1176)
[energy-band]_bin: ke results averaged to 1x1 degree bins (360x160)
[energy-band]_avg: ke results zonally averaged from 1x1 degree bins (160x1)

Grid Data: topo.mat

plat: latitude in degrees north (9000x7055)
plon: longitude in degrees east (9000x7055)
depth: ocean depth in meters (9000x7055)

MITgcm: results from MITgcm subsampled to 1/4 degree

Rotary Spectra: spectra_llc4320_0m.nc, spectra_llc4320_15m.nc

i: horizontal grid spacing (1440x1)
j: vertial grid spacing (1080x1)
ke_uv: spectra (1440x1080x1440x1)
freq_time: spectral frequencies (1440x1)
freq_time_spacing: spacing between points in freq_time (1x1)

Rotary Spectra averaged to 1x1 degree bins: spectra_llc4320_0m_bin.nc, spectra_llc4320_15m_bin.nc

ke_uv_bin: spectra averaged to 1x1 degree bins (360x162x1440)
freq_time: spectral frequencies (1440x1)
lat: latitude range of bin-averaged data (360x1)
lon: longitide range of bin-averaged data (1x162)

Kinetic Energy Results: ke_llc4320_0m.mat, ke_llc4320_15m.mat

lat_grid: model latitude averaged to 1x1 degree bins (360x162)
lon_grid: model longitide averaged to 1x1 degree bins (360x162)
xmid: latitude at bin centers (162x1)
ymid: longitide at bin centers (360x1)

For each energy band:
[energy-band]: ke results on model grid (1440x1080)
[energy-band]_bin: ke results averaged to 1x1 degree bins (360x162)
[energy-band]_avg: ke results zonally averaged from 1x1 degree bins (162x1)

Grid Data:

XC.nc
XC: longitude East of center of grid cell (1440x1080)

YC.nc
YC: latitude North of center of grid cell

Depth.nc
Depth: ocean depth in meters

(all grid files contain same i and j as MITgcm spectra files in addition to their grid variable)

Drifters: (more information on how this folder is organized is found at a readme within the folder)

Kinetic Energy Results (bin averaged): ke_drifters.mat

mz*: mean kinetic energy values on 1 degree by 1 degree bins
numz*: number of elements to form averages in each bin (168x360)
xmid*: longitide at bin centers (360x1)
ymid*: latitude at bin centers (168x1)

Kinetic Energy Results (zonally averaged): ke_drifters_avg.mat

ymid: latitude at bin centers (168x1)
xmid: longitide at bin centers (360x1)

For each energy band:
[energy-band]_[drouged|undrouged]: ke results averaged to 1x1 degree bins (168x360)
[energy-band]_[drouged|undrouged]_avg: ke results zonally averaged from 1x1 degree bins (168x1)

Rotary Spectra averaged to 1x1 degree bins, zonally averaged: spectra_drifters.mat

msnn: negative spectra for all drifters (120x721)
mspp: positive spectra for all drifters (120x721)
msnn_d: negative spectra for drogued drifters (120x721)
mspp_d: positive spectra for drogued drifters (120x721)
msnn_u: negative spectra for undrogued drifters (120x721)
mspp_u: positive spectra for undrogued drifters (120x721)
ymid: latitude at bin centers (120x1)
xmid: frequency at bin centers (721x1)

spectra: code to produce spectra

hycom:
hycom.py: Create zarrs, subsample model, compute rotary spectra & total KE
spectra_bin.m: average spectra to 1x1 degree bins

mitgcm:
subsample.ipynb: Subsample model
concat_rechunk.py: Concat and rechunk subsampled timeseries
spectra.ipynb: compute rotary spectra & total KE
spectra_bin.m: average spectra to 1x1 degree bins

ke: code to produce kinetic energy results from spectra

[hycom|mitgcm|drifter]_ke.m: organize integration, bin-average, shift grid
[hycom|mitgcm]_shift_grid.m: shift grid for easy mapping
ke_integrate.m: integrate frequency bands
ke_integrate_wide.m: integrate wide frequency bands

figures: code to produce figures

FIG_freq_lat.ipynb: visualize zonally averaged spectra
FIG_zonal_avgs.m: visualize zonally averaged kinetic energy for the frequency bands
FIG_maps.m: visualize kinetic energy for the frequency bands.

Related publication(s):
Brasch, J.M., et al. (2020). Factors Influencing Surface Kinetic Energy in Global High Resolution Ocean Models. Forthcoming.

Use and Access:
This data set is made available under a Attribution 4.0 International (CC BY 4.0).

To Cite Data:
Brasch, J.M., et al. (2020). Factors Influencing Surface Kinetic Energy in Global High Resolution Ocean Models [Data set]. University of Michigan - Deep Blue. https://doi.org/10.7302/r8q1-g224

Download All Files (To download individual files, select them in the “Files” panel above)

Total work file size of 40.7 GB is too large to download directly. Consider using Globus (see below).



Best for data sets > 3 GB. Globus is the platform Deep Blue Data uses to make large data sets available.   More about Globus