Report on Delayed Mode for Argo float WMO

6900765

ARGO ESPANA - IEO / 17 - 08

Delayed Mode Quality Control for Argo floatWMO 6900765

February 8, 2017

A. Gonzalez-Santana - P. Velez-BelchıInstituto Espanol de Oceanografıa

1 Introduction

The Delayed Mode Quality Control (DMQC) has been developed for float WMO 6900765and delivered on 22/12/2016 to ifremer. No anomalous profiles were detected during itsinitial analysis in any of the measured variables in the 199 profiles carried out.

Transmision system ARGOSTransmission ID 88344 2685Platform Model APEX APF9A 6315Platform ID 4400Sensors SBE41CP SBE41CP SBE41CPSensores s/n n/a n/a n/aData Centre (Format Version) IF (3.1)Project Name ARGO SPAIN (Malaspina)Data Centre (Format Version) IF (3.1)Project Name ARGO SPAIN (Malaspina)Float OwnerPI Name Pedro Joaquin VELEZ BELCHIParking Depth (dbar) 1000Profile depth (dbar) 2000Number of Profiles 199Status InactiveDeployment Date 03-Feb-2011 00:00:00Deployment Position Lat 24.82 Lon -22.47Last Surfacing Date 07-Jul-2016 02:02:49Deployed Position Lat 23.05 Lon -34.69Age (years) 5.4Voltage (v) 10.265Positioning SystemSensors CTD-PRES,CTD-TEMP,CTD-CNDC

Table 1. Technical information of the float.

Several checks were performed: Pressure values were studied to avoid possible TNDPanomalies. The Thermal Mass Error was also calculated in order to avoid possible errorsdue to the temperature gradients. The Owens and Wong Objective Mapping Analysis(2003) was applied to achieve an optimum calibration of the salinity.

2

2 Salinity correction from the OW method

Owens and Wong Objective Mapping Analysis (2003):

This calibration model assumes that salinity measurements drift slowly overtime. To correct possible salinity drifts, the model makes use of adjacent profiles (atime series) to estimate a time-varying multiplicative correction term ”r” by fitting tothe estimated climatological potential conductivities on theta surfaces. The inclusionof contemporary high quality calibrated hydrographic data with regional temperature -salinity relationships (by using nearby historical hydrographic data) helps to determinewhether a measured trend is due to sensor drift or due to natural variability.

Drift or bias evidence cannot be seen in the salinity measurement for WMO6900765 float. Therefore after the manual evaluation and inspection, no adjustment isneeded according to Argo Quality Control Manual: PSAL ADJUSTED = PSAL (originalvalue), PSAL ADJUSTED ERROR = Uncertainty provided by PI, PSAL ADJUSTEDQC = 1, 2 or 3.

PSAL ADJUSTED = PSAL (original value), PSAL ADJUSTED ERROR =Uncertainty provided by PI, PSAL ADJUSTED QC = 1, 2 or 3.

The following parameters has been set up for the Owens and Wong ObjectiveMapping Analysis method:

Config max casts 175use pv 0scale long large 2scale lat large 2scale long small 1scale lat small 1scale phi small 0scale phi large 0scale age 10p delta 250p exclude 200

Table 2. Owens and Wong Objective Mapping Analysis method parameters .

3

Figure 1: Argo float trajectory (a). T-S Diagram (b). Potential Temperature profiles (c).

Salinity profiles (d).

4

Figure 2: Potential temperature and salinity sections.

5

Figure 3: Pressure record (a). Voltage record (b).

6

315 320 325 330 335 340 345 35010

15

20

25

30

Longitude

Latitude

6900765 profile locations with historical data

1

112131

4151

61

718191

101111121131141

151161171

float

historical points

Figure 4: Historical points around the current ARGO float trajectory. These historical

points are used by Owens and Wong Objective Mapping Analysis to make a

model for an ARGO float data calibration.

7

5

10

15

20

25

35 35.5 36 36.5 37 37.5

θ ° C

Salinity (PSS−78)

6900765 uncalibrated float data (−) and mapped salinity (o) with objective errors

1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97103109115121127133139145151157163169175

(a) T-S Diagram

5

10

15

20

25

35 35.5 36 36.5 37 37.5

θ ° C

Salinity (PSS−78)

6900765 calibrated float data (−) and mapped salinity (o) with objective errors

1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97103109115121127133139145151157163169175

(b) T-S Diagram after a potential calibration

Figure 5: Both graphs show T-S diagrams before and after a potential calibration. This

is useful to identify water masses, to detect some possible offsets or to identify

some anomalous profiles.

8

0 20 40 60 80 100 120 140 160

0.9992

0.9994

0.9996

0.9998

1

1.0002

1.0004

1.0006

1.0008

r

6900765 potential conductivity (mmho/cm) multiplicative correction r with errors

2 x cal error

1 x cal error

1−1 profile fit

0 20 40 60 80 100 120 140 160−0.04

−0.03

−0.02

−0.01

0

0.01

0.02

0.03

float profile number

∆ S

6900765 vertically−averaged salinity (PSS−78) additive correction ∆ S with errors

2 x cal error

1 x cal error

1−1 profile fit

Figure 6: Salinity variation between each profile. Owens and Wong Objective Mapping

Analysis builds its model based in a programmed number of break points.

9

0 20 40 60 80 100 120 140 160

35.5

35.52

35.54

35.56

35.58

35.6

35.62

35.64

35.66

35.68

6900765 salinities with error on θ=11.9018 °C

PS

S−

78

0 20 40 60 80 100 120 140 16035.6

35.62

35.64

35.66

35.68

35.7

35.72

35.74

35.76

35.78

6900765 salinities with error on θ=12.713 °C

PS

S−

78

float profile number

uncal float

mapped salinity

cal float w/1xerr.

Figure 7: This figure gives a rough idea how uncalibrated (blue line) and calibrated (green

line) signals fit each other. Bear in mind that mapped salinity depends on the

historical hydrographic points of the area (Figure 1). The less historical points,

the less approximated is the model.

10

Salinity anom on theta. 6900765

0 20 40 60 80 100 120 140 160 1805

10

15

20

profile number

the

ta

0 20 40 60 80 100 120 140 160 1802

2.5

3

3.5

4

4.5

5

−0.1 −0.05 0 0.05 0.1

320 340 360

10

20

30

(a) Original salinity variation

Calibrated salinity anom on theta. 6900765

0 20 40 60 80 100 120 140 160 1805

10

15

20

profile number

the

ta

0 20 40 60 80 100 120 140 160 1802

2.5

3

3.5

4

4.5

5

−0.1 −0.05 0 0.05 0.1

320 340 360

10

20

30

(b) Calibrated salinity variation

Figure 8: Brians King plots. Both show the salinity variation for an each level of theta

per profile. A colored scale indicates the salinity variation (white color indicates

no varation). Comparing both uncalibrated and calibrated plots, significant

salinity variations can be identified.11

34 35 36 37 380

5

10

15

20

25

30

PSS−78

OW chosen levels −6900765

0 2 4 6

x 10−3

2

4

6

8

10

12

14

16

18Salinity Variance on Theta

Po

ten

tial t

em

p (

° C)

salinity variance

0 10 20 30−2000

−1800

−1600

−1400

−1200

−1000

−800

−600

−400

−200

0

°C

Pre

ssu

re (

db

ar)

OW chosen levels −6900765

35 36 37−2000

−1800

−1600

−1400

−1200

−1000

−800

−600

−400

−200

0

PSS−78

OW chosen levels −6900765

Figure 9: Theta levels are chosen by Owens and Wong Objective Mapping Analysis. The

model identifies automatically the theta levels where the salinity variations are

smaller.

12