Defence Applications of Operational

Oceanography

An Australian Perspective

Commander Robert Woodham BSc(Hons) MSc Royal Australian Navy

Director of Oceanography and Meteorology, Royal Australian Navy

Abstract

Oceanographic conditions can affect naval operations in a variety of ways, and

for this reason navies around the world have traditionally used oceanographic ob-
servations, and climatologies derived from them, for operational decision making.
Rapid advances in global ocean observing systems since the 1990s, and more re-
cently in operational ocean forecasting systems, offer substantial opportunities for
improved decision making. The recent focus of many defence forces on informa-
tion superiority has coincided with the availability of high resolution forecasts of
oceanic physical properties. These oceanic data sets are being used to assess and
forecast such properties as sea surface height, temperature and salinity, for acous-
tic applications to undersea warfare, and oceanic currents and tidal streams, for
Search and Rescue (SAR), mine warfare and amphibious applications. The Royal
Australian Navy (RAN) is using ocean forecasts from the BLUElink global ocean
modelling system and a limited area ocean model, and is developing a very high
resolution model for applications in the littoral zone, as well as integrating high
resolution oceanographic data to sonar range prediction models. These military
applications of operational oceanography are reviewed, and illustrated with exam-
ples from an Australian perspective.

1.1 Introduction

It is perhaps self-evident that Navies around the world are interested in ocean con-
ditions. What may not be so obvious, however, is the variety of ways in which the
ocean can affect naval operations. This chapter aims to describe oceanographic
impacts on maritime operations, and how these can be assessed and forecast using
operational oceanographic capabilities which have become available in recent
years. Whilst its content is generally applicable to naval forces, it is presented
2

from the viewpoint of the Royal Australian Navy (RAN), which has been closely
involved in the establishment of oceanographic observation and forecasting in
Australia, as a partner in the ‘BLUElink’ project [Brassington et al., 2007]. Jacobs
et al. have recently published a more general overview of how operational ocean-
ography is being used by navies throughout the world, which includes examples
from the United States, the United Kingdom, France and Australia [Jacobs et al.,
2009]. Harding and Rigney have previously published an overview of operational
oceanography in the United States Navy [Harding and Rigney, 2006].
In general terms, military forces around the world have given increasing atten-
tion, in recent times, to the importance of basing decision making on the most
comprehensive and up to date information available. This is partly due to the in-
creasing capabilities offered by information and communications technologies
(ICT), driven primarily by the commercial sector, and partly due to a change in
emphasis to manoeuvre warfare, rather than positional (or attritional) warfare. The
focus on manoeuvre warfare has its roots in the latter part of the Cold War, when
NATO realised that it must use force-multipliers, including information superior-
ity and the manoeuvrist approach, if it was to overcome the numerical superiority
of Soviet forces. The related concept of ‘Network Centric Warfare’ (NCW), as
distinct from a platform centric approach, envisages the rapid collection and dis-
semination of actionable information, using the latest technologies, to achieve in-
formation superiority throughout the battlespace.
Environmental information, including Meteorological and Oceanographic
(METOC) information, is regarded by modern navies as a vital component of in-
formation superiority and NCW, allowing naval forces to optimise their weapons,
sensors and manoeuvre for the prevailing and forecast environmental conditions.
For these reasons, the more technologically advanced world navies have been
quick to take advantage of the recent rapid developments in operational oceanog-
raphy, which have been described elsewhere in this volume. Improved oceanic ob-
servations, from both satellite and in situ sensors, data management and oceanic
analysis and forecast systems have all been applied to naval operations, in order to
contribute to the goal of information superiority.
This approach is particularly applicable in Australia, because oceanographic
conditions in the region are so complex. The East Australian Current affects the
Tasman Sea, spawning numerous warm- and cold-core eddies [Ridgway and
Dunn, 2003]. The Leeuwin Current flows down the west coast and across the
Great Australian Bight. The Pacific-Indonesian Throughflow affects the Timor
and Arafura Seas and the Northwest Shelf. The Antarctic Circumpolar Current af-
fects waters to the south of the continent. Other oceanographic phenomena in the
region include upwelling events (particularly along the Queensland coast, and the
Bonney coast of South Australia), internal waves, solitons, extreme tidal ranges
and abundant freshwater inflows, providing strong buoyancy forcing during the
Northwest monsoon. Faced with the need to operate successfully in such complex
 3

waters, the RAN has been quick to appreciate the need to maintain a state-of-the-
art oceanographic capability, and has worked closely with partners, notably the
Australian Bureau of Meteorology (BoM) and the Commonwealth Scientific and
Industrial Research Organisation (CSIRO), to develop such a capability.

1.2 Impacts of the ocean on operations

1.2.1 Anti-Submarine Warfare (ASW)

The need for specialist oceanographic expertise first came to be recognised by

the RAN in the mid 1950s, when the Fairey Gannet Anti-Submarine Warfare
(ASW) aircraft was first operated from the aircraft carrier HMAS MELBOURNE.
Meteorological officers onboard MELBOURNE provided tactical oceanographic
advice to the Gannet squadrons, using bathythermographic observations of the
ocean as the basis for sonar performance predictions. This advice was used by the
Gannet crews to determine the optimum deployment of buoys fitted with hydro-
phones (‘sonobuoys’), which they used in the acoustic detection and tracking of
submarines.
In making these acoustic assessments, the effects of the ocean’s thermohaline
structure on the propagation of sound in water must be considered. The effects on
sound speed of temperature, salinity and pressure are as follows:
• Temperature – sound speed is higher in warmer water (4 ms-1 per 1oC)
• Salinity – sound speed is higher in more saline water (1.4 ms-1 per 1 PSU)
• Depth – sound speed is higher at greater pressure (1.7 ms-1 per 100m)

Acoustic propagation in water can be understood, to a first approximation, by

imagining that sound propagates through a homogeneous medium in straight lines
(the ‘raytrace’ approach) [Urick, 1983]. The refraction of sound ‘rays’ is described
by Snell’s Law, which states that, when a ray crosses a boundary between two
media in which its speed of propagation is different, the ratio of the sines of the
angles of incidence and refraction is equal to the ratio of those speeds. This means,
in simple terms, that sound in the sea is refracted towards areas of lower sound
speed. The degree of refraction is also frequency dependent, being greater for
higher frequencies. Snell’s Law can be applied qualitatively, to understand the
acoustic properties of the water column, and hence determine optimum tactics,
such as search or evasion plans. It can also be applied quantitatively, in sonar
range prediction models, such as the RAN’s ‘Tactical Environmental Support Sys-
tem version 2’ (TESS 2). These models estimate detection ranges, based on ocean
4

acoustics, the performance characteristics of sonar systems (such as operating fre-

quencies, transmitted power, pulse length, processing losses and gains, etc), and a
knowledge of target characteristics (such as target strength, depth, aspect, etc).
Ray-tracing models are generally found to give good results at medium and high
frequencies (above 1-2 kHz). These frequencies are typically used by active
sonars, which transmit a pulse of acoustic energy, and detect its echo (as distinct
from passive sonars, which detect radiated noise from a target). Active sonars are
fitted in ships and submarines, and can be deployed from aircraft as sonobuoys or,
in the case of helicopters, on winches (‘dipping’ sonars).
Consider a typical thermal profile of the ocean, such as the one taken from the
central Tasman Sea shown in figure 1.1. This profile is from the ‘Ship Of Oppor-
tunity Programme’ (SOOP) dataset, and has been extracted from the Integrated
Marine Observing System (IMOS) Ocean Portal. The top 20 to 30 m shows an
isothermal profile in the mixed layer. Here, temperature and salinity are constant,
but pressure will increase as depth increases, giving rise to a slight increase in
sound speed. This will have the effect of refracting sound waves upwards towards
the surface. If the sound frequency is sufficiently high, in comparison to the depth
of the mixed layer, acoustic rays travelling through the water at small angles to the
horizontal will be refracted upwards towards the surface, where they will be re-
flected. After reflection, the rays will again be refracted upwards towards the sur-
face. This has the effect of trapping acoustic energy, provided its frequency ex-
ceeds a critical value for the particular layer depth, in the surface ‘duct’, which can
give rise to low acoustic losses, and therefore long ranges. If the surface wind is
light, surface losses due to scattering on reflection will be low, and very long
ranges are possible. In order to take advantage of this effect, hull-mounted active
sonars in ASW frigates are generally designed to operate at frequencies which are
high enough to be trapped by the surface duct, in order to maximise detection
ranges against shallow submarines.
Below the mixed layer, the water gets colder in the thermocline zone. Between
the base of the mixed layer at around 30 m, and the base of the thermocline at
around 100 m, the temperature has fallen by about 7oC (figure 1.1). This means
that the sound speed will have increased by around 1.7 ms-1, due to the increasing
pressure, but fallen by around 28 ms-1 due to the decreasing temperature, so over-
all there is a large decrease in sound speed, which means that acoustic energy will
be refracted downwards. Between 100m and 800m, there is a drop in temperature
of around 1oC per hundred metres, which means that the sound speed will de-
crease by about 2.3 ms-1 per hundred metres. Overall, in the water column below
the mixed layer, there is a downward-refracting profile, stronger in the main ther-
mocline region, which means that acoustic energy will be refracted down towards
the sea bed. If the sea bed is a good absorber of acoustic energy at the relevant
frequency, such as a muddy or silty bottom, acoustic propagation will be generally
poor.
 5

The profile shown in figure 1.1 extends to around 850 m. Below this depth,
however, there will come a point where the decrease in sound speed caused by the
temperature lapse is negated by the increase in sound speed due to the pressure in-
crease. In isothermal water, clearly the sound speed will increase with depth, and
acoustic energy will start to be refracted up towards the surface.

Fig. 1.1 Typical thermal profile of the ocean, taken from the central Tasman Sea. Data is from
the Ship of Opportunity Programme (SOOP), and was obtained from the IMOS Ocean Portal

The consequences of an increasing sound speed at depth can be seen in figure

1.2, which shows a sound speed cross section through the Tasman Sea at longitude
155oE, from 30oS to 40oS. Temperature and salinity data has been obtained from
the Ocean Forecast Australia Model (OFAM) for 31 March 2008, and converted
to sound speed using Mackenzie’s equation [Mackenzie, 1981]. There is a sound
speed minimum at a depth of around 1,200m, whilst below this depth, sound speed
increases due to the small temperature lapse combined with the increasing pres-
sure. The value of sound speed increases to be similar to that at the surface. The
sound speed minimum at 1,200m is associated with an acoustic channel, which is
a very low-loss path. Above 1,200m, sound tends to be refracted downwards, to-
wards the channel axis. Below 1,200m, sound tends to be refracted upwards, again
towards the channel axis. A depth of 1,200m would therefore be a good depth at
which to position a hydrophone, in order to take advantage of this low-loss path in
the acoustic detection of submarines.
6

Where the sound speed at depth exceeds the value at the surface, a ‘conver-
gence zone’ may be experienced. This is a ring around the acoustic source, typi-
cally with a radius of around 25 miles, where sound is focussed by a caustic effect.
This focussing of acoustic energy near the surface provides opportunities for
greatly increased ranges, and it is even possible for multiple convergence zones to
be present, giving even longer range detections.

Fig. 1.2 Sound Speed profile through the Tasman Sea at 155 degrees East on 31 March 2008.
Sound speed (colour stretch at right hand side) is in metres per second. Data is from the Ocean
Forecast Australia Model (OFAM). An anticyclonic eddy is evident at around 32oS.

In addition to the vertical gradients of sound speed discussed so far, horizontal

gradients of sound speed are caused by temperature and salinity gradients associ-
ated with fronts and eddies, and these can also have a large effect on the acoustic
properties of the ocean. For example, an anticyclonic (warm core) eddy will have
warmer water towards its centre, therefore there will be a lateral gradient of sound
speed associated with the eddy. An anticyclonic eddy can be seen in figure 1.2, at
around 32oS, with an associated sound speed maximum at a depth of around 200
m. If a surface ship outside the eddy is searching for a submarine in the centre of
the eddy, using active sonar, the sound will be refracted away from the submarine,
reducing the probability of detection. Similarly, if the ship and submarine are on
the opposite sides of an oceanic front, detection ranges will be much reduced.
Acoustic effects such as the ones described in this section have been well
known, and have been the principal concern of naval oceanographers, for a long
time. Recent advances in operational oceanography are starting to provide the
 7

highly detailed oceanic data required to enable acoustic assessments and forecasts
to be made at greatly increased spatial and temporal resolutions, suitable for tacti-
cal applications. For example, a submarine wishing to evade acoustic detection
can use such oceanographic data to identify a location in the thermocline beneath
the mixed layer, where the water is not too deep and the bottom is a good absorber
of low frequency noise. This will ensure that its radiated noise is directed down to
the sea bed, where it is absorbed, hence minimising counter-detection ranges.
ASW aircraft can use high-resolution oceanographic data to identify near-surface
sound channels, deploying the hydrophones on their sonobuoys or dipping sonars
in the channel in order to achieve the greatest possible detection ranges. Knowl-
edge of the location of fronts and eddies enables ASW frigates to design the most
effective search plans, armed with an accurate assessment of detection ranges.
These are just a few examples of how the wealth of oceanographic data now avail-
able presents abundant opportunities for the ingenuity of naval oceanographers
and tacticians to be stimulated.
As well as temperature and salinity, ocean currents also have an effect on
ASW, and should be considered by naval forces. Submarines can take advantage
of currents to increase their speed over the ground, whilst keeping their engines at
low power (and therefore operating quietly). In some cases, particularly in the
Australian region, ocean currents can run at 3 or 4 knots [Roughan and Middleton,
2002], so this effect can be significant. Ocean currents can also be taken into ac-
count in sonar range prediction systems, such as the RAN’s TESS 2, since they af-
fect sound speed.

1.2.2 Amphibious Warfare

Amphibious operations can be very sensitive to weather and oceanographic

conditions. The offloading of troops and equipment from specialist amphibious
shipping to a beachhead involves transfers from ships to landing craft, and from
landing craft to the beach itself. Most navies possess a range of relatively small
watercraft for use during amphibious operations. Such activities are sensitive to
sea state, swell and surf conditions, tidal streams, longshore currents and rips,
which must all be assessed and forecast in order to ensure mission success.
Many navies use sea, swell and surf models to predict oceanic conditions in the
littoral environment, and hence assess their impact on amphibious operations. Fig-
ure 1.3 shows the output from an experimental implementation of the ‘Simulating
Waves Nearshore’ (SWAN) wave model, and the US Navy’s ‘Surf’ model, which
displays model output using a Geospatial Information System (GIS). The model
has been run over North Beach, Cronulla, which is on the east coast of New South
Wales to the south of Sydney. Figure 1.3 shows: significant wave height (grey
contours) and direction (vectors); significant wave period (blue rasters); littoral
8

currents (closely spaced arrows along the approach to the beach); wave trains (dis-
played as representative wave crests); and breaker percentage (displayed as green
for <1%, amber for 1-15% and red (surf zone) for >15%). This information can be
used in the planning phase of an amphibious assault, to compare the suitability of
various beaches for the operation, or to predict conditions at the beach at the time
of the assault. Depending on the nature of the assault, a suitable beach may be re-
quired to have negligible surf, although a single line of low, spilling surf may be
tolerated, and manageable longshore currents. The location of the beach centre
and approach lanes can also be chosen, using model output of this type, to avoid
rips. A knowledge of the location and strength of longshore currents in the boat
lanes, at the time of the assault, will help the landing craft crews to make a
successful approach and beaching.

Fig. 1.3 Sea and surf conditions forecast for North Beach, Cronulla. See text (sec-
tion 1.2.2) for an explanation of the symbology.
The RAN is developing a high resolution forecasting system, called the ‘Litto-
ral Ocean Modelling System’ (LOMS), which will provide sea, swell and surf pre-
dictions at greater resolution and fidelity, and over larger domains, than the
SWAN / Surf implementation described above. It will provide a three dimensional
characterisation of the wave conditions, at resolutions in the order of tens of me-
tres.
 9

1.2.3 Mine Warfare

Mine Warfare operations include mine hunting (using specialist sonars and re-
motely operated vehicles), mine sweeping, and mine clearance diving. These op-
erations are generally conducted in littoral environments, which can be challeng-
ing due to the complexity of ocean conditions. Tidal streams are often strong,
turbidity can affect visibility, and variations in the bottom type and thermohaline
structure can make acoustic detection difficult.
In order to achieve good detection and resolution of small objects, mine hunt-
ing sonars typically operate at relatively high frequencies (hundreds of kHz). This
means that typical detection ranges are quite low, and so ocean models with hori-
zontal resolutions in the order of tens of kilometres are unable to provide adequate
resolution of the oceanic structure for these applications. The RAN uses a limited
area oceanic model, called the ‘Relocatable Ocean Atmosphere Model’ (ROAM),
which is described in section 1.4.2 below, to generate forecasts at resolutions
down to 1 or 2 km, and the LOMS model will provide even higher resolution in
the near future. A mine warfare variant of the TESS 2 sonar range prediction
software, called TESS 2M, provides acoustic assessments at the scales required by
mine warfare applications.
The main demand on naval oceanographers supporting mine warfare operations
is often to assess and predict currents, both at the surface and at depth. These de-
pend on the tidal regime, wind driven flow and influence of the current structure
in the adjacent deep ocean basin, all of which can be modelled by systems such as
ROAM. Remotely Operated Vehicles (ROVs) and divers may be limited by the
strength of these currents, and forecasts are used to identify windows of opportu-
nity, when current strengths are low enough that such activities will not be unduly
hampered. The thermohaline structure of the littoral water mass is of interest to
mine hunting operations, and can have a substantial impact, particularly in the
case of a salt wedge estuary or where there is strong tidal modulation.

1.2.4 Submarine operations

Submarine operations require a knowledge of the locations of fronts and eddies

(see section 1.2.1 above), and the general thermohaline structure of the ocean, in
order to identify the best tactics for detection, attacks and evasion. The strength
and direction of ocean currents is also required, for the purposes of manoeuvre. A
knowledge of conditions at the surface, such as wind waves and swells, enables
the risk of counter-detection to be assessed, informing decisions on whether it is
safe to raise a periscope or communications mast, or to recharge batteries by
‘snorting’.
10

1.2.5 Search and Rescue (SAR)

Analyses and forecasts of ocean currents are invaluable in the assessment of

drift during Search and Rescue (SAR) operations, by informing the design of ef-
fective search plans. Perhaps the most complex aspect of such calculations are as-
sociated with the drift of the object being searched for under the influence of the
wind (or ‘leeway’) [Hackett et al., 2006]. Objects with different shapes, such as
persons wearing lifejackets, survival rafts and lifeboats, experience different lee-
way effects. Even without the assistance of algorithms which account for leeway
effects, a good approximation can often be obtained from ocean models which in-
clude currents, tidal streams and Ekman flow. In addition, knowledge of sea sur-
face temperatures (SST) allows survival times to be estimated. Ocean modelling
systems can even be used to investigate historical problems of this type, such as
the search for the location of HMAS SYDNEY II, which was greatly assisted by
oceanic drift calculations using BLUElink reanalysis data [Mearns, 2009]. The
SYDNEY wreck site was located off Western Australia in April 2008, 66 years af-
ter the ship sank, with the tragic loss of her entire ship’s company.

1.2.6 Maritime interdiction operations

The bulk of the chapter so far has concentrated on high-end warfighting appli-
cations of operational oceanography, such as prosecuting submarines, clearing
minefields and conducting amphibious assaults. Oceanographic products are also
used, however, to provide routine support to lower tempo operations, such as
maritime interdiction, patrol tasks and constabulary activities, which may be con-
strained by high sea states or heavy swells.
A current example is the use of real-time satellite observations, and forecasts,
of significant wave height to identify the risk of pirate attacks off the coast of So-
malia. After establishing the correlation between pirate attacks and satellite obser-
vations of significant wave height, using historical data, a ‘stoplight’ product is
routinely provided to naval forces, including an Australian frigate, in the interna-
tional Combined Task Force 150 (CTF 150), operating off the Somali coast. This
product shows the risk of pirate attack in three categories: ‘probable’, ‘possible’
and ‘unlikely’ (figure 1.4).
 11

Fig. 1.4 36 hour Forecast of risk of pirate attacks off the Somali coast, based on significant wave
height, valid 28 August 2009.

1.3 Forecast methods – their strengths and weaknesses

1.3.1 Climatology

Until the recent advent of operational oceanography, Navies have had to rely
on climatologies or point observations to make operational decisions [Greggs et
al., 2009]. Climatologies can be useful for planning purposes, but they are of lim-
ited use where oceanic variability is high. In the extreme case of a bimodal sys-
tem, climatology shows the mean of the two modes, which may be a physical
situation which never arises in reality (e.g. south or north of a front, inside or out-
side an eddy). Figure 1.5 illustrates the limitations of climatology, by showing
SST in the Tasman Sea as depicted by the World Ocean Atlas 2001, and a snap-
shot of SST from the BLUElink forecasting system.
12

Fig. 1.5 Comparison of SST as depicted by climatology (left) (World Ocean Atlas 2001, Sep-
tember) and an oceanic model (right) (BLUElink Reanalysis, 16 September 2009).

Conversely, where variability is low, or where it occurs on timescales longer

than the averaging period (normally monthly), climatology can give a very good
indication of expected conditions. Furthermore, the expected error of a forecast
based on climatology is independent of the lead time of the forecast, and climatol-
ogy gives better guidance than other forecast methods, such as persistence or de-
terministic models, at long lead times. Forecasts based on persistence or determi-
nistic models asymptote to twice the climatological monthly anomaly at long time
periods, because once these forecasts are completely decorrelated from reality,
they have errors from having anomalies in the wrong places, as well as errors from
not having anomalies in the right places [Kalnay, 2000]. For this reason, defence
forces normally use climatological oceanographic data when conducting long-
range planning.

1.3.2 Persistence

Point observations, such as temperature profiles from eXpendable Bathy Ther-

mograph (XBT) systems, have been used by navies for many decades to infer the
acoustic properties of the water column. These observations are relatively simple
to make, and do not require assistance from ashore. This approach amounts to a
persistence forecast, that is, an assumption that the water properties will not
change during the period for which the assessment is required. Persistence fore-
casts can be expected to have lower errors than climatology at the start of the fore-
cast period, but as the oceanic flow evolves from this initial state, the errors grow
rapidly. Spatial variations, caused by the physical movement of the ship or aircraft
making the observation, as well as temporal variations due to ocean dynamics,
contribute to these errors. Nonetheless, a persistence forecast may be useful where
uniform conditions can reasonably be expected, such as in high latitudes where
mixed layers are very deep, or over continental shelves, provided the water is well
 13

mixed, and advection is minimal. Persistence is a more valid approach for short
lead time forecasts, such as may be required for a Co-ordinated Anti-Submarine
EXercise (CASEX) lasting two or three hours, than for longer lead times. Never-
theless, where spatial and temporal variability is great, such as in the waters
around Australia, a persistence forecast may be misleading even on very short
timescales.

1.3.3 Deterministic forecasts

Deterministic forecasts of the ocean have only become available in relatively

recent times [Bell et al., 2000]. In Australia, the BLUElink ocean forecasting sys-
tem commenced routine forecasting operations in August 2007 [Brassington et al.,
2007]. Provided sufficient observational data is available, deterministic forecasts
should have relatively small errors at the start of the forecast period. These errors
will grow more slowly than persistence forecast errors, because the deterministic
model is able to keep up with changes in the state of the ocean, by modelling its
dynamic processes. Deterministic forecasts from systems such as BLUElink are
highly detailed, providing variables such as temperature, salinity, current and sea
surface height at high spatial resolution for forecast periods of several days. They
represent a huge advance on the persistence and climatological forecasts used by
navies for many decades. In one sense, however, their strength is also their weak-
ness, since it is difficult to transmit the high volumes of oceanic data now avail-
able from deterministic ocean forecasting systems from shore to ships and subma-
rines.

1.3.4 Ensemble forecasts

Ensemble forecasting is well established for Numerical Weather Prediction

(NWP), but less so for oceanographic forecasting. Ensemble techniques are used
to generate covariance matrices for oceanic data assimilation applications [Oke et
al., 2008], and some ocean models have tangent linear and adjoint versions, which
can be used to generate ensembles of initial conditions. The computational burden,
however, of running oceanic ensembles operationally probably remains hard to
justify, at least for the time being. A further difficulty arises from the size of initial
errors in oceanic models, which tend to be larger than in NWP, because the ocean
is less well observed than the atmosphere. This means that the assumption of line-
arity, which is generally required for ensemble techniques, is much less valid in
the ocean than for the atmosphere.
14

1.4 Naval applications of deterministic forecasts

The ocean analysis and forecasting capability of the BLUElink system has been
described elsewhere in this volume. This section will describe how forecasts from
the BLUElink system, including ROAM, are used by the RAN for operational de-
cision making.

1.4.1 The BLUElink global / regional model (OceanMAPS)

The BLUElink Ocean Modelling, Analysis and Prediction System (Ocean-

MAPS) is implemented at the Bureau of Meteorology (BoM) in Melbourne
[Brassington et al., 2007]. It produces an analysis and 6-day forecast of ocean tem-
perature, salinity, currents, sea surface height and mixed layer depth twice per
week. Model output graphics are available from the BoM public website, and the
model data itself is available to the RAN, and more generally for research pur-
poses, from the BoM’s ‘Thematic Realtime Environmental Distributed Data Serv-
ices’ (THREDDS) server, in Network Common Data Form (NetCDF) format.
The OceanMAPS system is currently configured to give eddy-resolving resolu-
tion (10 km horizontally) over the Australian region (90oE – 180oE and 16oN –
75oS). Within this domain, OceanMAPS data is routinely used by the RAN to cre-
ate oceanographic charts, which are available to naval personnel for a range of ap-
plications, including ASW, amphibious and mine warfare, passage planning and
spatial awareness. An example of a ‘METOC Oceanographic Forecast Summary’
(MOFS) chart is shown in figure 1.6.
From the MOFS chart shown in figure 1.6, it is readily apparent that the Jervis
Bay (JB) exercise areas, shown as polygons in the lower portion of the graphic, is
dominated by a large, anticyclonic eddy, centred over its eastern margin. There is
a sharp temperature contrast between 14oC water in the southwest of the JB areas,
and 18oC water on the western flank of the eddy, where reduced sonar ranges may
be expected. For an ASW exercise in this location, the ASW commander might
decide to allocate search assets either side of the temperature gradient, in order to
achieve an efficient search. The submarine commander may chose to remain in the
core of the current associated with the eddy, in order to evade detection, moving to
either side for brief periods in order to improve sonar performance so that the tac-
tical picture can be compiled. The currents can also be exploited by the submarine
to increase speed over the ground. Air assets deploying lines of sonobuoys can ap-
ply knowledge of the current field to ensure that the buoy patterns are not sheared
out of shape by the flow. In addition, specialist oceanographers (‘METOC’ offi-
cers) may be available to provide further insights into the acoustic properties of
the area, using OceanMAPS data, and hence assist decision making.
 15

Fig. 1.6. ‘METOC Oceanographic Forecast Summary’ (MOFS) chart, showing SST and currents
in the Tasman Sea for 29 August 2009. MOFS charts are routinely produced twice weekly,
showing forecasts out to 6 days, and are used by a variety of naval personnel.
16

1.4.2 Relocatable Ocean Atmosphere Model (ROAM)

The Relocatable Ocean Atmosphere Model (ROAM) is used by the RAN to

generate high resolution oceanic and atmospheric forecasts over limited domains
of interest to the Australian Defence Force (ADF). ROAM is designed to be set up
by non-expert users, with minimal input, anywhere in the Australian region
[Herzfeld, 2009], and is used routinely by RAN forecasters. The ROAM ocean
model is initialised and forced by data from OceanMAPS, and is typically imple-
mented at resolutions of 1-2 km. Figure 1.7 shows Sea Surface Temperature (SST)
and currents calculated by ROAM for a domain in the vicinity of Hobart, Tasma-
nia, which was used for the recent RAN mine warfare exercise ‘DUGONG’. Exer-
cise ‘DUGONG’ involved the Mine Hunter Coastal (MHC) vessels HUON and
DIAMANTINA, which provided mine sweeping and hunting capabilities, the aux-
iliary minesweeper BANDICOOT, clearance diving teams and US Navy salvage
divers. It took place over two weeks in October 2009 in the Derwent River, and
the approaches to Hobart.
In this example, the current characteristics were of primary importance to the
exercise, which involved an underwater survey of the historic wreck of MV Lake
Illawarra in the Derwent River. The water temperature was also of interest to the
diving teams, to ensure that they were suitably prepared for the prevailing condi-
tions. ROAM was used to generate current forecasts at intervals down to one hour,
and in addition, the ROAM atmospheric model provided high resolution forecasts
of the wind strength and direction, also at one hour timesteps, which allowed
changes in the sea state to be anticipated.
 17

Fig. 1.7. Sea Surface Temperature (SST) and current forecast produced by the ROAM system for
the mine warfare Exercise DUGONG in October 2009.

As well as providing oceanographic data for graphical products, the output

from ocean forecasting systems can be used in sonar range prediction systems, in
order to produce assessments and forecasts of acoustic conditions which take ac-
count of the spatial and temporal variability of the ocean environment. Figure 1.8
shows a series of sonar range predictions, which have been generated by the
RAN’s Tactical Environmental Support System (TESS 2) using ROAM data at 1
km resolution. The domain is in the vicinity of Jervis Bay, which is on the south
coast of New South Wales (NSW), adjacent to the RAN air station at Nowra. It is
an area where the RAN frequently conducts ASW and MW exercises. The sonar
range predictions are displayed as ‘Probability of Detection’ plots (PODgrams),
where a 90% or greater probability of detection is shown in red. Figure 1.8 shows
18

ROAM sea surface temperature and currents as the background. The three POD-
grams are for an ASW frigate leaving Jervis Bay and tracking to the northeast,
searching for a submarine at Periscope Depth (PD). The capabilities of the sonar
used for the calculation are fictional. The PODgrams seem to make sense intui-
tively, since they show the greatest ranges inshore, where the water is shallow and
with a relatively homogeneous thermohaline structure, and bottom losses are low
from the sandy sea bed. Offshore, where the temperature gradient is greater, detec-
tion ranges are less. The scale of figure 1.8 can be gauged by considering that the
current vectors are shown at the ROAM resolution of 1 km. The PODgrams have
hollow centres because echoes cannot be received whilst the sonar is transmitting,
which gives rise to a ‘dead zone’ of varying radius, depending on the duration of
the transmitted pulse, and the speed of sound in water.

Fig. 1.8 Sonar performance predictions produced by ROAM and TESS 2 for 1000 UTC on 06
October 2009, in the vicinity of Jervis Bay, NSW. The background shows sea surface tempera-
ture (colour stretch and contours) and current vectors. The three ‘Probability of Detection’ plots
(PODgrams) are over-plotted, with a probability of detection of 90% or more shown in red.
 19

1.5 Summary

Oceanographic data has been collected by the world’s navies for many years, and
used to inform the planning and conduct of a range of naval operations. Perhaps
the main preoccupation of the naval oceanographer is with the acoustic properties
of the ocean, because acoustic detection is of great importance in Anti-Submarine
Warfare (ASW) and Mine Warfare (MW). The effects of oceanic temperature and
salinity, and the depth of water, on sound speed are well known, so oceanographic
conditions can be used to infer acoustic properties, both qualitatively by naval per-
sonnel, and quantitatively in sonar range prediction systems. The relatively recent
advent of operational oceanography, spanning ocean observation systems, data as-
similation and deterministic forecasting models, has made available a wealth of
observational and forecast data at high spatial and temporal resolutions. These
new datasets are being used by naval oceanographers to provide much improved
characterisations of the physical structure of the ocean, in order to inform opera-
tional and tactical decision making. The time and space scales which are starting
to be resolved allow oceanographic support to be provided in complex littoral en-
vironments, where there is a demand from amphibious and mine warfare opera-
tions. This new oceanographic capability is timely, given a broader trend towards
information superiority in the more technologically advanced defence forces.
This chapter has described oceanographic effects on ASW, Amphibious War-
fare, Mine Warfare, submarine operations and lower tempo activities such as
Search and Rescue (SAR) and maritime interdiction operations. The acoustic
properties of the ocean have been outlined in some detail, using examples from the
Tasman Sea. The strengths and weaknesses of various forecasting methods,
namely climatology, persistence, deterministic and ensemble forecasts, have been
described, from the perspective of naval forces. Finally, some examples have been
given of the use of deterministic forecasts of the ocean, including ASW activities
in the Tasman Sea, a Mine Warfare exercise in the approaches to Hobart, Tasma-
nia, and the use of high resolution oceanographic data to generate range-dependent
sonar predictions in the Jervis Bay exercise areas.
The maturing international capability for operational oceanography presents a
remarkable opportunity for the world’s navies and maritime forces, and this has
been seized on by the RAN and other top-rank navies. As the resolvable time and
space scales continue to reduce, and progress is made with the downscaling of
global systems to coastal scales, the complexities of the littoral environment,
which are of great interest to navies around the world, will continue to be unrav-
elled. It is an exciting time to be a naval oceanographer.
20

Acknowledgments

I would like to thank the International GODAE Summer School organizing

committee, for their kind invitation to present a lecture on defence applications of
operational oceanography at the International GODAE Summer School in Perth,
Western Australia, during January 2010, on which this chapter is based. Sincere
thanks also to Lieutenant Commander Aaron Young, RAN, for his kind assistance
with the figures.

References

Bell, Michael J., Forbes, Richard M. and Hines, Adrian, ‘Assessment of the FOAM global data
assimilation system for real-time operational ocean forecasting’, 2000, Journal of Marine
Systems, Vol 25, pp 1-22

Brassington, G.B., Pugh, T., Spillman, C., Shulz, E., Beggs, H., Schiller, A. and Oke, P.R.,
‘BLUElink – development of operational oceanography and servicing in Australia’, 2007,
Journal of Research and Practice in Information Technology, Vol 39, No 2, pp 151-164
Hackett, Bruce, Breivik, Oyvind and Wettre, Cecilie, ‘Forecasting the drift of objects and sub-
stances in the ocean’, in ‘Ocean weather forecasting’, 2006, Eds. Chassignet, E.P. and Ver-
ron, Jacques, Springer
Harding, John and Rigney, James, ‘Operational oceanography in the US Navy: a GODAE per-
spective’, in ‘Ocean weather forecasting’, 2006, Eds. Chassignet, E.P. and Verron, Jacques,
Springer
Herzfeld, Mike, ‘Improving stability of regional numerical ocean models’, 2009, Ocean Dynam-
ics, Vol 59, pp 21-46
Jacobs, Gregg A., Woodham, Robert, Jourdan, Didier and Braithwaite, Jez, ‘GODAE applica-
tions useful to navies throughout the world’, 2009, Oceanography, Vol 22, No 3, pp 182-189
Mackenzie, Kenneth V., ‘Nine term equation for sound speed in the oceans’, 1981, J. Acoust.
Soc. Am., Vol 70, Issue 3, pp 807-812
Mearns, David L., ‘The search for the Sydney’, 2009, HarperCollins Australia
Oke, P.R., Schiller, A., Griffin, D.A. and Brassington, G.B., ‘Ensemble data assimilation for an
eddy-resolving ocean model of the Australian region’, 2005, Q. J. R. Met. Soc., Vol 131, No
613, pp 3301-3311.
Ridgway, K.R. and Dunn, J.R., ‘Mesoscale structure of the mean East Australian Current system
and its relationship with topography’, 2003, Progress in Oceanography, Vol 56, pp 189-222

Roughan, M. and Middleton, J.H., ‘A comparison of observed upwelling mechanisms off the
east coast of Australia’, 2002, Continental Shelf Research, Vol 22, No 17, pp 2551-2572
 21

Urick, Robert J., ‘Principles of Underwater Sound’, 1983, McGraw-Hill Book Company, USA.