Satellite Estimation of Coastal pCO₂ and Air-Sea Flux of Carbon Dioxide in the Northern Gulf of Mexico

Authors

Steven E. Lohrenz, University of Massachusetts--DartmouthFollow
Wei-Jun Cai, University of DelawareFollow
S. Chakraborty, University of Massachusetts--DartmouthFollow
W.J. Huang, National Sun Yat-Sen UniversityFollow
X. Guo, Xiamen UniversityFollow
R. He, North Carolina State UniversityFollow
Zuo Xue, Lousiana State UniversityFollow
Katja Fennel, Dalhousie UniversityFollow
Stephan Howden, University of Southern MississippiFollow
Hanqin Tian, Auburn UniversityFollow

Document Type

Article

Publication Date

3-15-2018

Department

Marine Science

Abstract

Satellite approaches for estimation of the partial pressure of CO₂ (pCO₂) and air-sea flux of CO₂ in coastal regions offer the potential to reduce uncertainties in coastal carbon budgets and improve understanding of spatial and temporal patterns and the factors influencing them. We used satellite-derived products in combination with an extensive data set of ship-based observations to develop an unprecedented multi-year time-series of pCO₂ and air-sea flux of CO₂ in the northern Gulf of Mexico for the period 2006–2010. A regression tree algorithm was used to relate satellite-derived products for chlorophyll, sea surface temperature, and dissolved and detrital organic matter to ship observations of pCO₂. The resulting relationship had an r² of 0.827 and a prediction error of 31.7 μatm pCO₂ (root mean-squared error of the relationship was 28.8 μatm). Using a wind speed and gas exchange relationship along with satellite winds, estimates of air-sea flux of CO₂ were derived yielding an average annual flux over the period 2006–2010 of − 0.8 to − 1.5 (annual mean = − 1.1 ± 0.3) mol C m⁻² y⁻¹, where the negative value indicates net ocean uptake. The estimated total annual CO₂ flux for the study region was − 4.3 + 1.1 Tg C y⁻¹. Relationships of satellite-derived pCO₂ with salinity were consistent with shipboard observations and exhibited a concave relationship with low values at mid- and low salinities attributed to strong biological drawdown of CO₂ in the high productivity river-mixing zone. The time-series of satellite-derived pCO₂ was characterized by a seasonal pattern with values lower during winter and spring, low to intermediate values during fall, and higher and more variable values during summer. These findings were similar to simulations from a coupled physical-biogeochemical model. A seasonal pattern was also evident in the air-sea flux of CO₂ with generally more negative fluxes (i.e., ocean uptake) during winter and spring, and positive fluxes during summer months with fall being a period of transition. Interannual variations in annual means of both air-sea flux of CO₂ and DIN loading were significant, with higher DIN loading coinciding in some cases with more negative air-sea flux of CO₂ (i.e., net ocean uptake). Spatial patterns of pCO₂ reflected regional environmental forcing including effects of river discharge, wind forcing, and shelf-slope circulation. Our study also illustrates the utility of satellite extrapolation for highlighting areas that may contribute significantly to regional signals and for guiding prioritization of locations for acquiring further observations. The approach should be readily applicable to other regions given adequate availability of in situ observations for algorithm development.

Publication Title

Remote Sensing of Environment

Volume

207

First Page

71

Last Page

83

Recommended Citation

Lohrenz, S. E., Cai, W., Chakraborty, S., Huang, W., Guo, X., He, R., Xue, Z., Fennel, K., Howden, S., Tian, H. (2018). Satellite Estimation of Coastal pCO₂ and Air-Sea Flux of Carbon Dioxide in the Northern Gulf of Mexico. Remote Sensing of Environment, 207, 71-83.
Available at: https://aquila.usm.edu/fac_pubs/15038