MRI measurements of partial pressure of CO2 in surface waters of the Pacific during 1968 to 1970: re-evaluation and comparison of data with those of the 1980s and 1990s
Hy. Inoue et al., MRI measurements of partial pressure of CO2 in surface waters of the Pacific during 1968 to 1970: re-evaluation and comparison of data with those of the 1980s and 1990s, TELLUS B, 51(4), 1999, pp. 830-848
During 1968 to 1972, scientists of the Meteorological Research Institute/Ja
pan Meteorological Agency measured CO2 in the surface seawater and overlyin
g air in the Pacific Ocean quasi-continuously to examine CO2 exchange betwe
en the sea and the atmosphere. From the data remaining in our laboratory fr
om that time, we re-evaluated the partial pressure of CO2 in surface seawat
er (pCO(2)(s)) by taking into account pressure broadening effects due to th
e use of CO2-in-N-2 standards, the use of chemical desiccant (Mg(ClO4)(2)),
calibration using background air data, seawater temperature rise between t
he equilibrator and sea surface, the WMO CO2 mole fraction scale, and diffe
rences in pressure during the oceanic CO2 measurements from those of standa
rds and background air in the nondispersive infra-red gas analyzer cell. Th
e overall uncertainty of pCO(2)(s) measurements for the cruises from 1968 t
o 1970 was estimated to be less than 3.6 mu atm, which allowed us to evalua
te temporal variations in the carbonate system over a few decades, but unce
rtainty was higher (>10 mu atm) for the cruise from 1971 to 1972 because of
serious malfunctions of the system. The re-evaluated pCO(2)(s) data from 1
968 to 1970 exhibit patterns similar to those observed at the same place an
d time of year (within 30 days) in 1982/83, 1988, 1995 and 1996, but compar
ed with levels in the 1980s and 1990s the pCO(2)(s) level was clearly lower
in the wide area of the Pacific except south of the Subtropical Front (STF
; 47 degrees S) in the Australian sector. The observed pCO(2)(s) increased
by 34 +/- 5 mu atm (n = 133) for the area 7 degrees N to 35 degrees N, 138
degrees E to 147 degrees E from February 1969 to February 1995, 29 +/- 5 mu
atm (n = 247) for the area 9 degrees N to 35 degrees N, 138 degrees E to 1
65 degrees E from February 1970 to February/March 1996, 26 +/- 7 mu atm (n
= 224) for the area 29 degrees N to 51 degrees N, 170 degrees W from April
1970 to April 1988, and 41 +/- 9 mu atm (n = 165) for the area 10 degrees S
to 45 degrees S, 148 degrees E to 166 degrees E from January/February 1969
to January/February 1995. In the northern subtropics (7 degrees N to 35 de
grees N, 138 degrees E to 147 degrees E), we estimated the long-term increa
se (35 +/- 6 mu atm, n = 133) after removing seasonal variations that were
obtained from the pCO(2)(s)-sea surface temperature (SST) relationship. Obs
erved and seasonally adjusted increases were nearly equal to those of the p
artial pressure of CO2 in the air (1.4 mu atm yr(-1)) over the same time in
tervals. South of the STF, pCO(2)(s) increase as found in the subtropics wa
s not detected, mostly due to the large variability of pCO(2)(s) (250 to 38
0 mu atm in 1968/69) on small spatial scales. The average pCO(2)(s) south o
f the STF showed large variations on time scales of months and years that a
ffect the estimation of the growth rate of atmospheric CO2.