Auxiliary material for Paper 2010GB004009



Impacts of temporal CO2 and climate trends on the detection of ocean anthropogenic CO2 accumulation


Nathalie F. Goodkin
Department of Earth Sciences, University of Hong Kong, 
Hong Kong, China

Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, 
Woods Hole, Massachusetts, USA


Naomi M. Levine
Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, 
Woods Hole, Massachusetts, USA

Department of Organismic and Evolutionary Biology, Harvard University, 
Cambridge, Massachusetts, USA


Scott C. Doney
Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, 
Woods Hole, Massachusetts, USA


Rik Wanninkhof
Atlantic Oceanographic and Meteorological Laboratory, NOAA, 
Miami, Florida, USA

Goodkin, N. F., N. M. Levine, S. C. Doney, and R. Wanninkhof (2011), Impacts of temporal CO2 and climate trends on the detection of ocean anthropogenic CO2 accumulation, Global Biogeochem. Cycles, 25, GB3023, doi:10.1029/2010GB004009.

Introduction
The supplementary material includes a description of the multiple water mass calculation, one supplemental table, and six supplemental figures.




2010gb004009-txts01.pdf
Text containing a description of the multiple water mass eMLR calculation and results. 

2010gb004009-ts01.pdf
       
Table S1: Cross-­correlation (r) for biogeochemical and physical parameters used in the eMLR.
	

2010gb004009-fs01.pdf
       
Figure S1: Transient-no feedback MLR residual (μmol kg-1) for the ocean P16S transect.



2010gb004009-fs02.pdf
       
Figure S2: Control MLR residual (μmol kg-1) for the ocean P16S transect.



2010gb004009-fs03.pdf
       
Figure S3: Comparison between MLR DIC residuals (μmol/ kg) along P16S (150°W) for an MLR applied to the whole water column and an MLR applied to two alkalinity defined water masses.



2010gb004009-fs04.pdf
       
Figure S4: Average percentage of grid points used in each longitudinal band for the percentage error analysis.

2010gb004009-fs05.pdf
       
Figure S5: Comparison between error estimate for delta Canthro (eMLR) for an MLR applied to the whole water column and an MLR applied to two alkalinity defined water masses.



2010gb004009-fs06.pdf
       
Figure S6: Error estimate for the eMLR using monthly model output (January) along the P16S transect.