My postdoctoral research here at WHOI involves several projects. Initially, I was hoping to compare the output of the ECCO data-assimilating model with the data from the Kuroshio Extension System Study (KESS) project in the Northwest Pacific. However, analysis showed that the one-degree horizontal resolution cannot adequately represent dynamics in a western boundary current region. Although this project was not initially successful, it shows the importance of developing data assimilation systems with higher resolution. Another potential follow-up project would be the comparison of the KESS data with results of a high-resolution forward model (without data assimilation). I am hoping to collaborate on this in the future with Julie McClean, who is currently working on a 1/10-degree run of the Parallel Ocean Program (POP) model for the KESS region and time period.
Currently, I am involved in an analysis of a 100-year POP run with climatological atmospheric forcing. The model output shows interannual and decadal variability despite the lack of atmospheric forcing at those time scales, suggesting intrinsic oceanic variability. Analysis of this has led me to interesting features in both the path of the Kuroshio, and the characteristics of subtropical mode water. In the case of the Kuroshio, the large meander in the pathway shown in the model results is also seen in observations, indicating that it is not a model artifact and that it is intrisically forced. On the other hand, there are some observed features, such as interannual variability of the Kuroshio Extension, which are completely absent from the model output. This indicates such variability is likely to be atmospherically forced. This analysis is another example of comparisons between models and observations leading to a better understanding of the system and its dynamics. Similarly, the formation and erosion of mode water in the model reflects that observed in nature. The full dynamical picture only a model can provide allows for a quantification of characteristics such as age of mode water that are very hard to observe. However, this analysis of mode water is one aspect that would really be enhanced by comparison with a model with realistic forcing during the KESS time period, to allow direct comparison with data, as opposed to general characterization of the water mass.