Propagation of internal tides on the Northwest Australian Shelf studied with time-augmented empirical orthogonal functions

J.W. Book, Nicole Jones, Ryan Lowe, Gregory Ivey, C.R. Steinberg, R.M. Brinkman, A.E. Rice, Cynthia Bluteau, S.R. Smith, T.A. Smith, S. Matt

Research output: Chapter in Book/Conference paperConference paperpeer-review


A series of collaborations between U.S. and Australian funded
projects resulted in the deployment of 30 moorings at 23 different
sites between November 2011 and August 2012 along
the Northwest Australian Shelf (NWS), a region of strong internal
tide oscillations. The measurement program peaked during
April 2012.

A time-augmented empirical orthogonal function (EOF) technique
was developed and implemented that allowed the correlated
variability, structure, and phasing of semidiurnal internal
tide packets to be evaluated at multiple mooring sites. This
technique allows the EOF analysis to preserve the non-linear
character of the internal wave packets without requiring a large
number of EOF modes.

In three different examples from the NWS, between 45% and
65% of the variance of complex internal tide dominated temperature
structures were explained using only two EOF modes.
Using only the first mode of the EOFs our analysis also identified
a propagation blocking effect that was found to be associated
with the mesoscale coastal circulation and the wind.
Original languageEnglish
Title of host publicationThe Proceedings of the 20th Australasian Fluid Mechanics Conference
EditorsGreg Ivey, Nicole Jones, Tongming Zhou
PublisherAustralian Fluid Mechanics Society
Number of pages4
ISBN (Print)9781740523776
Publication statusPublished - 2016
Event20th Australasian Fluid Mechanics Conference - University of Western Australia, Perth, Australia
Duration: 5 Dec 20168 Dec 2016
Conference number: 20


Conference20th Australasian Fluid Mechanics Conference
Abbreviated titleAFMC
Internet address


Dive into the research topics of 'Propagation of internal tides on the Northwest Australian Shelf studied with time-augmented empirical orthogonal functions'. Together they form a unique fingerprint.

Cite this