Forecasting Ocean Chlorophyll in the Equatorial Pacific

• Main Authors: Cecile S. Rousseaux, Watson W. Gregg
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2017-07-01
• Series: Frontiers in Marine Science
• Subjects: enso; chlorophyll; phytoplankton; forecast; biogeochemical modeling
• Online Access: http://journal.frontiersin.org/article/10.3389/fmars.2017.00236/full

Using a global ocean biogeochemical model combined with a forecast of physical oceanic and atmospheric variables from the NASA Global Modeling and Assimilation Office, we assess the skill of a chlorophyll concentrations forecast in the Equatorial Pacific for the period 2012–2015 with a focus on the forecast of the onset of the 2015 El Niño event. Using a series of retrospective 9-month hindcasts, we assess the uncertainties of the forecasted chlorophyll by comparing the monthly total chlorophyll concentration from the forecast with the corresponding monthly ocean chlorophyll data from the Suomi-National Polar-orbiting Partnership Visible Infrared Imaging Radiometer Suite (S-NPP VIIRS) satellite. The forecast was able to reproduce the phasing of the variability in chlorophyll concentration in the Equatorial Pacific, including the beginning of the 2015–2016 El Niño. The anomaly correlation coefficient (ACC) was significant (p < 0.05) for forecast at 1-month (R = 0.33), 8-month (R = 0.42) and 9-month (R = 0.41) lead times. The root mean square error (RMSE) increased from 0.0399 μg chl L−1 for the 1-month lead forecast to a maximum of 0.0472 μg chl L−1 for the 9-month lead forecast indicating that the forecast of the amplitude of chlorophyll concentration variability was getting worse. Forecasts with a 3-month lead time were on average the closest to the S-NPP VIIRS data (23% or 0.033 μg chl L−1) while the forecast with a 9-month lead time were the furthest (31% or 0.042 μg chl L−1). These results indicate the potential for forecasting chlorophyll concentration in this region but also highlights various deficiencies and suggestions for improvements to the current biogeochemical forecasting system. This system provides an initial basis for future applications including the effects of El Niño events on fisheries and other ocean resources given improvements identified in the analysis of these results.