Date of Award
Doctor of Philosophy (PhD)
A five-component data assimilative ecosystem model is developed in order to investigate the effects of physical processes encompassing a wide range of space and time scales, on the lower trophic levels of the highly dynamic central equatorial Pacific (140°W). Many of the biological processes included in the ecosystem model respond to environmental fluctuations with time scales between one and ten days, which are not typically resolved by basin to global scale circulation models. Therefore, the ecosystem model is forced using daily observations from the TAO mooring array. Model simulations successfully reproduce data collected both during and after the 1991–92 El Niño, suggesting that species composition changes are not of first order importance when examining the effects of this El Niño on the equatorial Pacific ecosystem. Simulations also highlight the importance of higher frequency mesoscale events, such as tropical instability waves and equatorially-trapped internal gravity waves.
The feasibility of improving simulation skill by assimilating biogeochemical observations, is also examined. The adjoint method, which minimizes model/data misfits by adjusting model parameters, provides a viable option for assimilating data into marine ecosystem models. When assimilating synthetic data subsampled using the resolution of the JGOFS EqPac cruises, parameters governing processes such as grazing, growth, mortality and recycling can be recovered exactly. The assimilation of phytoplankton time-series data, as are currently available from SeaWiFS, additionally requires a small amount of supplemental in situnutrient data. As expected, deterioration in simulation skill is more severe when assimilating biased data, as compared to data containing random noise.
When actual EqPac cruise data and SeaWiFS ocean color data are assimilated, one can objectively determine whether or not a given model structure is consistent with specific sets of observations. For instance, a brief period of macro-nutrient limitation during the 1997–98 El Niño, as well as changes in species composition observed during the passage of a tropical instability wave, are found to violate key model assumptions. Thus, although the assimilation of biological data into a marine ecosystem model cannot necessarily overcome inappropriate model dynamics, data assimilation can often serve to guide model reformulation.
Friedrichs, Marjorie A..
"Physical Control of Biological Processes in the Central Equatorial Pacific: A Data Assimilative Modeling Study"
(1999). Doctor of Philosophy (PhD), dissertation, Ocean/Earth/Atmos Sciences, Old Dominion University, DOI: 10.25777/kmqn-1k61