Argo Profilers missions, sampling rates, accuracy etc

1
Argo Profilersmissions, sampling rates, accuracy
etc
A

• Howard Freeland
• D.F.O. Science/Pacific Region, Canada
• Tel [phone]
• Email FreelandHj_at_pac.dfo-mpo.gc.ca

2
Argo Profilersmissions, sampling rates, accuracy
etc
A

• Howard Freeland
• D.F.O. Science/Pacific Region, Canada
• Tel [phone]
• Email FreelandHj_at_pac.dfo-mpo.gc.ca

3
Argo Profilersmissions, sampling rates, accuracy
etc
B
16th Nov. 2006 2658 floats

• Howard Freeland
• D.F.O. Science/Pacific Region, Canada
• Tel [phone]
• Email FreelandHj_at_pac.dfo-mpo.gc.ca

4
The Argo community plans to deploy 3000 floats
5
And we very nearly are there!
6
How does an Argo float work?
SOLO Float Courtesy of the animation wizards at
Woods Hole and Scripps.
PROVOR Float
APEX Float
7
How many of each float type?
20th Nov. 2006
8
How are floats being launched?
Or deployed from container vessels
Floats are most commonly launched from research
vessels
Dropped from aircraft.
9
The Air Launch is so neat, youve got to see
this
10
Finally, what does a mission look like
X
11
Alternate missions exist.
1) Launch
6) Transmit data
Ocean surface
5) Profile to 0
2) Dive
1000 db
Etc.
3) 10-d drift
4) Dive to 2000
2000 db
The standard mission repeats 1) to 6) until
batteries expire.
12
Alternate missions exist.
1) Launch
6) Transmit data
Ocean surface
9) Profile to 0
5) Profile to 0
2) Dive
Etc.
7) Dive
1000 db
3) 10-d drift
8) 10-d drift
4) Dive to 2000
2000 db
This is Park and Profile, PnP 2 I use PnP 2,
Riser is using 3 or 4 a recent addition is DPF
Deep Profile First
13
Alternate missions exist.
4) Transmit data
1) Launch
Ocean surface
7) Profile to 0
3) Profile to 0
Etc.
5) Dive
1000 db
2) Dive to 2000
6) 10-d drift
2000 db
a recent addition is DPF Deep Profile
First this enhances quality control if the float
was launched from a research vessel doing CTDs
14
New sensors are appearing
Water intake and platinum resistance
thermometer. Conductivity cell Aanderaa optode
sensor, for dissolved oxygen which has gained
acceptance because of perceived high value for
the data return, and has minimal impact on the
primary Argo mission.
15
New sensors are appearing
Wet Labs FLNTU fluorometer turbidity sensor.
16
How do we get the critical information?
17
Two critical items about float performance1)
Accuracy, from manufacturer
Resolution Sensor Accuracy Reported Pressu
re 2.4 dbar 0.1 Temperature 0.002 0.001 Salinit
y 0.005 0.001
Analysis of floats recovered with various times
at sea shows that these are being met and that
sensors usually show little drift.
18
Two critical items about float performance2)
Float energetics
Energy stored on an APEX 2350 kJ (alkaline
cells) gt 220 profiles via Argos
19
The messages are broadcast to a satellite
20
The messages are broadcast to a satellite
02442 25410 65 32 J 3 2006-11-17 060937
44.963 233.944 0.000 401652016 2006-11-17
060337 1 72 05 11
A7
85 E2 1B 51
12 6B
85 BB
19 5D 12
B0 85
87 17 6A
14 00 85
79
15 73 14 A2
85 59
13 81
This is message number 5, messages were received
in the order- 7, 8, 7, 9, 10, 11, 1, 10, 2, 5,
7, 9, 10, 8 etc
11A7 (hex) 4519 (decimal) pressure 451.9
85E2 (hex) 34274 (decimal) salinity 34.274
1B51 (hex) 6993 (decimal) temperature 6.993
21
The messages are broadcast to a satellite
Floats in the Gulf of Alaska get 24 passes/day
and it takes 8 hours to get a profile home.
Floats in the Gulf of Guinea get 12 passes/day
and it takes 16 hours to get a profile home, also
costs kJ and total profiles.
22
The consequence is-
An easy exercise, count the floats in the Gulf of
Mexico!
23
A tightly integrated system
A typical Canadian float uses the same parameters
that Steve Riser uses, he has advised many
nations. Average latitude 50N 8 hours at sea
surface We use PnP 2 (i.e. drift at 1000 dbars,
profile from 2000 every second
profile) Transmit data from 72 levels Accuracy as
specified by the manufacturer We have floats
running with gt170 profiles completed we think 200
profiles/float is likely. What can we improve or
change? Very little it seems.
24
How can we escape these constraints?

• Energy stored on an APEX 2350 kJ (alkaline
  cells)
• 220 Argos profiles or 208 Iridium profiles
• 
  this may change too!

25
From Webb Research, and I assume Martec/Metocean
will follow -

• Energy stored on an APEX 2350 kJ (alkaline
  cells)
• 220 Argos profiles or 208 Iridium profiles
• WRC is experimenting with a carbon-fibre hull for
  their floats.
• Lighter material means more batteries can be
  carried.
• Slightly compressible hull implies less energy
  needed to move from 2000 dbars to the surface.
• We can use lithium batteries, but that poses
  serious operational issues.
• In Canada we are not rushing to adopt Iridium
  communications or lithium batteries, for a
  variety of reasons, including shipping
  inflexibility.

26
How can we escape these constraints?

• Switching to Iridium has immediate benefits-
• Only 3 minutes at the sea surface, so grounding
  risk is greatly reduced and fouling risk is
  greatly reduced.
• Much larger files are transferred easily.
• The fraction of profiles passing DMQC will
  increase.
• But, switching to Iridium is not without costs-
• Iridium does not supply positions, so increased
  capital cost.
• With only 3 min at the surface we lose surface
  trajectories.
• Adding Iridium to a standard float does cost
  energy and, therefore, profiles.

27
An Iridium antenna needs care
Standard patch configuration
Mai-tai configuration this IS sub-optimal for
communications, but.
28
An Iridium antenna needs care
Standard patch configuration in tow tank dye test
Mai-tai configuration in tow tank dye test
29
What of the future?

• One obstacle for the objective of global coverage
  is the difficulty of operating in ice-infested
  waters.
• Three approaches have been designed
• JAMSTEC/Metocean
• Germany ice detection algorithms store
  profiles for future transmission
• US (WHOI Breck Owens) approach of probing for
  openings in the ice.

30
JAMSTEC Under-Ice sampling.
31
Arctic Argo started this year
32
German Under-Ice sampling system.
33
German Under-Ice sampling.

• Each float has a built-in ice-detection
  algorithm. If a float detects close approach to
  the freezing point of sea-water then the ascent
  is aborted and the profile saved for future
  transmission.
• When ice clears the float delivers a set of
  profiles. This sometimes happens during the
  ice-covered season if a float is lucky to find a
  hole in the ice.
• Profiles are of no value without positions, so
  each float carries a Rafos receiver that allows
  it to determine its position underwater.

34
WHOI Under-Ice sampling system.

• Use Iridium communications to detect the presence
  of ice, algorithm is simple-
• Have I stopped rising, if yes, can I hear Iridium
  satellites?
• If yes then the float is in water, if no it
  is under ice.
• 2) If the float is not under ice, transmit 1 or
  more profiles.
• 3) If the float is under ice, descend 50 metres,
  wait 1 hour and try again, then again, then
  again.
• 4) After 50 probes descend to drift depth and
  wait 10 days.
• This system is still experimental but shows great
  promise. It will work best in marginal ice
  zones.

35
Argo is a global ocean sampling program available
to anyone anywhere. It is yours, to use as you
see fit, but please, please we have written a
data manual, read it
A

• Howard Freeland
• D.F.O. Science/Pacific Region, Canada
• Tel [phone]
• Email FreelandHj_at_pac.dfo-mpo.gc.ca