An Improved Algorithm for Estimating Total Alkalinity in the Northern Gulf Of Mexico

• Main Author: Devkota, Madhur
• Format: Text
• Language: unknown
• Published: Scholars Junction 2018
• Subjects: ocean acidification; total alkalinity; spatial modeling; GWR; gulf of mexico; modeling; Ocean acidification
• Online Access: https://scholarsjunction.msstate.edu/td/561; https://scholarsjunction.msstate.edu/cgi/viewcontent.cgi?article=1560&context=td

Mississippi River affects the carbon dynamics in the northern Gulf of Mexico (N-GoM) significantly. Hence, total alkalinity (TA) algorithms developed for major ocean basins produce inaccurate estimations for this region. A TA algorithm was developed, which addresses the local effects of coastal processes and complex spatial influences. In-situ data collected during numerous previous research cruises in the N-GoM were compiled and used to calculate the efficiency of an existing TA algorithm that uses Sea-Surface-Temperature (SST) and Sea-Surface-Salinity (SSS) as explanatory variables. To improve this algorithm, statistical analyses were performed to improve the coefficients and functional form of this algorithm. Then, chlorophyll-a (Chla) was included as an additional explanatory variable. Chla worked as a proxy for addressing the organic carbon pump’s pronounced effects on coastal waters. Finally, a Geographically Weighted Regression (GWR) algorithm was developed to address spatial non-stationarity, which apparently could not be addressed in the previously developed global algorithm.