INFORMATION CONTENT IN LOW DENSITY

1
? INFORMATION CONTENT IN LOW DENSITY XBT
TRANSECTS RELATIVE TO OCEANIC GYRES ROBERT L.
MOLINARI FORMERLY NOAA/AOML PRESENTLY
UNIVERSITY OF MIAMI- CIMAS CARMEN
ALEX UNIVERSITY OF MIAMI- CIMAS
2

• OUTLINE
• OBJECTIVES
• 2) SUMMARY OF LOW DENSITY NETWORK RELATIVE TO
  GYRE STRUCTURE INCLUDING BASIC CHARACTERISTICS
  AND HYPOTHESIS TESTING
• 3) IN VIEW OF RECENT QUESTIONS ABOUT STABILITY
• OF FALL-RATE EQUATION ARE RESULTS
  REPRESENTATIVE OF ACTUAL GYRE CHARACTERISTICS?
• 4) SUMMARY AND RECOMMENDATIONS

3

• OBJECTIVES (LIMITED BY TIME OF TALK TO NORTH
  ATLANTIC SUBTROPICAL GYRE)
• WHAT INFORMATION RELATIVE TO THE TIME DEPENDENT
  STRUCTURE OF THE NORTH ATLANTIC SUBTROPICAL GYRE
  CAN BE DISCERNED FROM LINES PROVIDING 40 YEARS
  OF OBSERVATIONS (E.G., IS THE GYRE SLOWING?)
• IN VIEW OF THE QUESTIONS RECENTLY RAISED ABOUT
  THE FALL RATE EQUATION OF XBTS, IS THE
  INFORMATION VALID?
• WHAT FUTURE STEPS SHOULD BE TAKEN?

4

• TYPES OF XBT AVAILABLE AT NODC
• DEPTH FIX RECOMMENDED
• Insufficient meta-data available 16,442
• Hanawa fix applied 106,267
• Hanawa fix required 649,941
• DEPTH EQUATION APPLIED
• Original Sippican equation 658,795
• Hanawa 113,557
• Unknown 0
• INSTRUMENT TYPE
• Unknown XBT type (I.e., Sippican, Spartan, etc)
  421,742
• Sippican T7 75,672
• Sippican T4 200,064

5
(No Transcript)
6
DATA Depth range MBT 150 to 200m
Early XBTs 450m
Post 1970 XBTs 750 m Sampling Pre
WOCE-TOGA regional oceanographic surveys
WOCE-TOGA occupation of
low-density network (LDN) shown in Figure
1. Types of LD lines for purposes of this poster
Real lines, VOS transects regularly (once/month)
occupied from one end to the other (blue lines
Figure 1) Imaginary lines given a WOCE-TOGA
number but not regularly occupied by a VOS (red
lines Figure 1). Extensive data can exist along
an imaginary line because of other transects
crossing the section (e.g., A6 is crossed by A5,
A7, A11, A12, A15, A20). Imaginary lines will
not be considered in this presentation.
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
anomaly
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
Wijffels et al. (2008)
18
Wijffels et al. (2008)
19
ISHII AND KOMOTO (2008)
20
GOURETSKI (2008)
21
Gouretski(2008)
22
Wijffels et al. (2008)
23
ISHII AND KIMOTO (2008)
24

• EXAMPLES OF FALL RATE CORRECTIONS
• Sippican depth Hanawa Ishii
  Wijffels
• Depth Error (2) For
  Sippican T7 For all T7s
• 
  1976 1987 1976 1987
• /-9m 465 465
  455 459 460
• 750 /-15m 775 775
  758 765 768

25

• 
  SUMMARY
• SEVERAL DIFFERENT DATA SETS AND METHODS HAVE BEEN
  USED TO DERIVE A VARIETY OF NEW FALL RATE
  EQUATIONS RANGING FROM PROBE SPECIFIC (I.E., FOR
  SIPPICAN T4S AND T7S AND SPARTON PROBES) TO ONE
  GLOBAL EQUATION.
• DATA SET LIMITED PARTICULARLY IN THE 1970S TO
  DETERMINE WHICH EQUATIONS ARE BEST.
• PRELIMINARY RESULTS SUGGEST THAT MOST FIXES ARE
  CAPABLE OF PRODUCING CORRECTIONS SUITABLE FOR
  STUDIES OF LARGE-SCALE OCEAN FEATURES, BUT NOT
  FOR CLIMATE VARIABILITY AND CHANGE REQUIREMENTS
  (PART OF PROBLEM THESE HAVE NOT BEEN DEFINED FOR
  TOTAL OCEAN VOLUME)

26

• Additional Problems
• Temperature biases Kizu and Hanawa (2002), up to
  1C
• Recorder biasesSnowden et al. (2008), below
  manufacturers error
• Limited data Wijffels et al. (2008),
  particularly southern oceans
• Incomplete meta-dataWijffels et al. (2008),
  particularly pre-1990
• Data location biases Thadathil et al. (2002),
  cold water bias

27

• RECOMMENDATIONS
• CONTINUE PRESENT METHOD OF STORING XBT DATA AT
  NODC (I.E., DO NOT APPLY NEW CORRECTIONS, EXCEPT
  HANAWA)
• IMPRESS ON DATA COLLECTORS THE IMPORTANCE OF
  PROVIDING REQUIRED META-DATA
• ASSEMBLE A WORKING GROUP OF MODELERS AND
  OBSERVATIONSALISTS TO DEVELOP A PLAN FOR DEFINING
  CLIMATE CRITERIA FOR SUBSURFACE TEMPERATURE DATA
  AND FOR CORRECTING HISTORICAL XBT DATA TO MEET
  THESE CRITERIA
• (SIMILAR TO APPROACH USED FOR SST)