Resolving the Influences of Climate Variability and Change on the State of Marine Ecosystem
Racault, Marie-Fanny1; Sathyendranath, Shubha1; Saux Picart, Stephane1; Raitsos, Dionysios E.1; Platt, Trevor1; Buitenhuis, Erik2; Le Quere, Corinne2
1Plymouth Marine Laboratory, UNITED KINGDOM; 2University of East Anglia, UNITED KINGDOM

The state of marine ecosystems is driven largely by large-scale patterns of climate variability. Short-term variability (6-18 months) in El-Nino Southern Oscillation (ENSO) index can have profound effects on the functioning of the marine ecosystem. Using decade-long (1998-2007) ocean-colour remote sensing observations at global scale, we demonstrate rapid responses in phytoplankton biomass and phenology to positive and negative phases of Multivariate ENSO Index (MEI). Over large regions of the oceans, the phytoplankton biomass and duration of growing season increased during the negative phase of MEI compared with the positive phase. The ecosystem response is greater in the tropics and sub-tropics where the influence of ENSO on upwelling and nutrients availability is more marked. In these regions, the state of marine ecosystem is tightly coupled to short-term variability in physical forcing. Long-term change in the state of ecosystem may then be elucidated only during coherent phases of climate variability.
To date, the record of ocean-colour observations has been enriched by a number of sensors spanning more than four decades. ESA Ocean-Colour Climate Change Initiative has advanced seamless merging of chlorophyll observations from sensors over the 1990s to 2010s period. However, comparison of these more recent observations with records from 1970s to 1980s remains a complex undertaking, particularly for absolute values of phytoplankton biomass, primarily due to differences in the sensors. A further impediment to the analysis of the past record is the gaps in the observations, both in time and space dimensions. We used relative indicators to synthesize observations from CZCS and SeaWiFS sensors. By comparing coherent phases of MEI anomalies, we unravel long-term change from short-term variability in the state of marine ecosystem. Phenological differences between positive MEI during the 1978-1983 and 2002-2006 periods reveal large increases in duration of phytoplankton growing period in the North Atlantic, southern sub-tropical Atlantic, western North Pacific, and western Indian oceans in the latter period. To assess the relative error in duration estimates due to sampling gaps in the CZCS sensor, we developed an original resampling method allowing comparison of duration estimates between gap-free and non-gap free time-series. In the global oceans, we find that phenology estimates from the CZCS record are subject to, on average, a relative error of 25%. This error is taken into account when assessing long-term change in the state of marine ecosystem.