My research efforts to date have focused on mechanisms of tropical Pacific climate change, the sensitivity of ENSO to climate change, mid-latitude low frequency ocean variability, and correction and calibration of subsurface ocean data for monitoring of ocean heat uptake. Throughout my graduate research under the guidance of Prof. Amy Clement I have advanced the understanding of the coupled response of the tropical Pacific to past and future climate changes. The following studies resulting from this research have bridged long-standing scientific controversies and presented new and exciting hypotheses.

During El Nino/Southern Oscillation (ENSO) the climate of the tropical Pacific experiences dramatic reorganizations in precipitation, atmospheric circulation, and ocean conditions with far-reaching impacts on human activities and marine ecosystems. Because both the climate of the tropical Pacific and ENSO result from coupling between winds and the equatorial thermocline, it is generally thought that permanent ENSO-like changes could result from the amplification of small east-west asymmetries generated in response to global warming. This interpretation pervades the literature on tropical climate change and has lead to the implication that the tropical Pacific could abruptly change as the Earth warms. These changes could result in dramatic patterns of climate change over the tropical Pacific and beyond, impacting for instance, precipitation over Indonesia, the Amazonia, and the U.S. Southwest; Peruvian fisheries; and Atlantic hurricanes. For this reason improving our understanding of this key mechanism of tropical Pacific climate change is fundamental for constraining model projections of future climate change.

Tropical Climate Change

My graduate research shows that the climate of the tropical Pacific is more stable than expected from the ENSO analogy. This is because during climate changes, such as future anthropogenic global warming (AGW), the Last Glacial Maximum (LGM) 21,000 years ago, or decadal variability, the coupling between the winds and thermocline is much weaker, if not absent. These studies have resulted in several research articles published, under review, and in preparation. The fundamental ideas leading to this line of research are developed in two published research articles (DiNezio et al. 2009b; DiNezio et al. 2010). The first of these studies (DiNezio et al. 2009b) reconciled the two leading theories for tropical climate change despite the apparent conflict between them. Using model projections of AGW performed for the IPCC AR4, this study shows that the response of the tropical Pacific to a warmer climate results fundamentally from the superposition of the weaker Walker and ocean dynamical thermostat mechanisms. The balance between these two opposing mechanisms determines the robust changes simulated by the IPCC AR4 models. However, in this study we did not provide a solid explanation to why the weakening of the Walker circulation is not amplified by a relaxation of the equatorial thermocline like during El Nino events, and instead is damped by the ocean. This issue is discussed in the second article (DiNezio et al. 2010), which presents arguments based on equatorial ocean dynamics to explain why the ocean does not amplify the response of the equatorial Pacific to a weaker Walker circulation. This has implications both for detection of climate change in observations and proxies, because the spatial pattern of the sea surface temperature is not a good constrain on these mechanisms. Instead, these mechanisms have an unmistakable signature in the subsurface ocean.

Paleoclimatology

Conflicting interpretations of climate reconstructions and numerical simulations of the LGM abound in the scientific literature. Analyzing output from the first coordinated set of coupled climate model experiments of the LGM, we have showed the response of the Pacific Walker circulation is not only sensitive to tropical-mean cooling, but also sensitive to changes in land-sea configuration due to lower sea level (DiNezio et al., submitted to Paleoceanography). According to our analysis of the models output, the ascending branch of the Walker circulation strengthens driven by a constrain between precipitation and humidity in response to tropical cooling, but also weakens due to reduced convection over land areas that were exposed during the LGM due to lower sea level. Current generation GCMs exhibit conflicting responses in the tropics because of diverging simulation of the balance between these two processes. However, the wide range of LGM tropical Pacific climates simulated by the models in response to the same forcing can be used to interpret proxies. We show that each mechanism is associated with robust ocean changes that could be unambiguously detected by thermocline and salinity proxies. To conclude we discuss how progress in robust multi-model and multi-proxy comparisons could ultimately lead to constraining the sensitivity of the Walker circulation to tropical cooling, and in this way shed light on its sensitivity to global warming.

Climate Variability

These ideas have influenced further research on decadal Pacific variability (PDV). Research lead by Prof. Clement has showed that the atmosphere can internally generate decadal variability in the absence ocean-atmosphere dynamical coupling (Clement et al. 2011, submitted). In this study, we also showed that on decadal and longer time scales the ocean acts to damp variability, opposite to on interannual time scales, i.e. ENSO, when coupled ocean-atmosphere feedbacks amplify variability. In a second paper currently in preparation (DiNezio and Clement 2011, in preparation), we are evaluating whether there is evidence for these mechanisms in observations.

Despite the fact that IPCC AR4 models simulate robust ocean changes in response to AGW (DiNezio et al. 2009b), their simulated ENSO variability shows diverging responses to the same forcing. This lack of agreement has been widely reported in the literature, but a cogent explanation remains elusive. Myriad ocean or atmospheric processes that affect ENSO can all be altered as the Earth warms, thus leading to diverging responses in the models. I am currently investigating the origin of this discrepancy using a budget analysis of the models’s output to estimate the changes in the thermocline, upwelling, and advection feedbacks that lead to growth of ENSO events. Preliminary results indicate that changes in each of these three feedbacks are robust and intimately linked to changes in the time-mean ocean climate, however, the balance among these three feedbacks is different in each models, explaining the wide range of ENSO responses (DiNezio et al. 2011b, in preparation).

In addition to the research on the tropical Pacific, I am interested in others aspects of the low frequency variability of the climate system. During the Large Scale Ocean Circulation course (MPO612) taught by Professor Bill Johns during the 2007 spring semester I lead a term project that resulted in an additional publication (DiNezio et al. 2009a). In this article, we showed that the year-to-year changes in the 25-year record of observations of the Florida Current, the name commonly used for the Gulf Stream where it flows through the Straits of Florida, can be explained in terms of the Sverdrup response to changes in winds over the subtropical North Atlantic. While this result is not unexpected from the theoretical point of view, our study is the first analysis explaining observed changes of an ocean current over long-term climate time-scales. This study also provides a physical explanation for a previously not understood link between the Florida Current and the North Atlantic Oscillation.