4 th steering committee Amsterdam, Netherlands 2-3 December 2014 DATA QC AND PRODUCTS ON ATLANTIC AREA With IFREMER, RIHMI-WDC, IHPT and IEO contributions
4 th steering committee Amsterdam, Netherlands 2-3 December 2014 DATA QC AND PRODUCTS ON ATLANTIC AREA With IFREMER, RIHMI-WDC, IHPT and IEO contributions.
Dec 28, 2015
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
4th steering committee Amsterdam, Netherlands 2-3 December 2014
DATA QC AND PRODUCTS ON ATLANTIC AREA
With IFREMER, RIHMI-WDC, IHPT and IEO contributions
4th steering committee – Amsterdam 2-3 December 2014
2/19
• Data collected by MARIS robot (28/10/2014) :
DATA INVENTORY (1)
CDI typeVertic
al profile
s
Time Series Total
New CDIS 13613 99 13712
Updated CDIs 2601 27 2628
Old CDIs 7040 231 7271
Total 23254 357 23611
4th steering committee – Amsterdam 2-3 December 2014
3/19
• Data collected by MARIS robot (28/10/2014) :
DATA INVENTORY (1)
CDI typeVertic
al profile
s
Time Series Total
New CDIS 13613 99 13712
Updated CDIs 2601 27 2628
Old CDIs 7040 231 7271
Total 23254 357 23611
More than twice the first harvesting
4th steering committee – Amsterdam 2-3 December 2014
4/19
First harvesting
DATA INVENTORY (2)Second harvesting
Old Updated
New
Enlargeddata coverage
4th steering committee – Amsterdam 2-3 December 2014
5/19
DATA INVENTORY BY PARAMETERS (3)
2nd harvestingCDI type
Phos
Amon
Silicate Nox
New CDIs 2230 250 762 835
Updated CDIs 2551 1835 2241 2416
Old CDIs 6538 2916 4107 4221
Total 11319 5001 7110 7472
Not as much data as expected
4th steering committee – Amsterdam 2-3 December 2014
6/19
DATA INVENTORY BY PARAMETERS (3)
2nd harvestingCDI type
Phos
Amon
Silicate Nox
New CDIs 2230 250 762 835
Updated CDIs 2551 1835 2241 2416
Old CDIs 6538 2916 4107 4221
Total 11319 5001 7110 7472
Phos
Amon
Silicate Nox
1st Harvesting 8832 4994 7071 7272
Old + Updated < 1st harvesting
4th steering committee – Amsterdam 2-3 December 2014
7/19
• P35 aggregation• Deletion of empty lines in a station : some levels with
no measurements for the parameter• Deletion of lines with negative depth (incorrect
missing value)• QC using ODV– Search for out of range values with Flag = 1– Set a QC for all measurements with Flag=0– Set default values with Flag=0 or Flag=1 to Flag=9– Replace zeros with Flag=1 by ½ LOD and flag 6 ?– Use N:P ratio ?
• Write a report on QC and send results to data providers
QUALITY CHECKS (1)
4th steering committee – Amsterdam 2-3 December 2014
8/19
LOQ (flag 6) LODAMON 0.96 – 0.7 - 1
LOQ/3
SLCA 0.1 – 0.3 PHOS 0.03 – 0.07 – 0.2NTRA+NTRI
0.24 – 0.5 – 0.51
NTRI0.027 – 0.3 –
0.1NTRA 0.7
FLAG 6 – PORTUGUESE DATA
Replace value with Flag=6 by 1/6 LOQ
• Values under Detection/Quantification Limit : LOD/LOQ
4th steering committee – Amsterdam 2-3 December 2014
9/19
• Many zero values, notably in russian data :
• Some entire profiles and several half-profiles with zeros -> missing data or LOD?-> use the LOQ/3 values from IHPT ?
ZERO VALUES
PHOS_oldNb 0 value
Flag 1 4877Flag 3 13Flag 4 90Total 4980
4th steering committee – Amsterdam 2-3 December 2014
10/19
QUALITY CHECKS – PROBLEMS FACED• Problem of unit conversion : ug/l to umol/l
-> reload russian data and some spanish data• Problem in CDI updating
-> high spanish Phosphate to delete• New QC flags have been changed compared to
the 1st QC, but also some of the reported errors to data centres have not been corrected :-> have to work on the three packages and not only the new CDIs!!-> we cannot just import the new collection in the collection built up after the first harvesting. We have to re-make the whole QC work.
4th steering committee – Amsterdam 2-3 December 2014
11/19
• Global Nox QC done :
• Data QC not yet finished but in good progress for : Phosphate, Silicate and Ammonium!
NB : no NTOT nor PTOT
NUTRIENTS PARAMETERS
4th steering committee – Amsterdam 2-3 December 2014
12/19
• Data from the first harvesting : Phosphate_final• Period : summer 2000-20009• Data with the mean as first guess (not yet a
refence field as background)
DIVA PRODUCTS
4th steering committee – Amsterdam 2-3 December 2014
13/19
• Results – analysis & errors field at surface
DIVA PRODUCTS
Poor data coverage and big error
4th steering committee – Amsterdam 2-3 December 2014
14/19
• Results – masked analysis with error
DIVA PRODUCTS
Only a coastal product
4th steering committee – Amsterdam 2-3 December 2014
15/19
All data 2000-2009 2000-2009 / summer
DIVA PRODUCTS• Phosphate from 1st harvesting
8832 CDI5591 meas
2811 CDI1051 meas
889 CDI263 meas
4th steering committee – Amsterdam 2-3 December 2014
16/19
All data 2000-2009 2000-2009 / summer
DIVA PRODUCTS• Phosphate from 2nd harvesting
Is it reasonnable to make a 10-year periods with yearly step from 1960 to 2014 (so 46 periods of 10 year), by seasons ?
11319 CDI6924 meas
3389 CDI 1206 meas
1221 CDI310 meas
4th steering committee – Amsterdam 2-3 December 2014
17/19
DIVA PRODUCTS
• Next steps within three weeks:-> install diva4.6.7
-> mask the Mediterranean Sea
-> using a refrence field as background
-> split Atl NE in two/three regions ?
-> a global product rather than 10 year moving windows ?
4th steering committee – Amsterdam 2-3 December 2014
18/19
MFSD ATLANTIC AREAS
• Bay of Biscay + Iberian coast• Macaronesia
4th steering committee – Amsterdam 2-3 December 2014
19/19
OTHER DIVA PRODUCTS• Phosphate at 50 m : summer 1960-2014
Related Documents
