Accurately predicting ENSO -- the single biggest predictor of global climate variability -- beyond six months is desirable but challenging. One potential improvement lies in the linkages between ENSO events and certain tropical and extratropical phenomena, known as ENSO precursors, which occur 6-18 months prior to the mature stage of ENSO. However, the extent to which these precursors are independent of one another and their ability to point to specific ENSO event types remains murky. Our lab is actively exploring these complexities.

The last few generations of state-of-the-art climate models unanimously predict a long-term warming trend in the equatorial Pacific in response to increasing anthropogenic CO2. This is crucial because the equatorial Pacific, home to ENSO, plays a vital role in regional and global climate variability. Interestingly, equatorial SST trends in observational datasets oppose this predicted warming trend. Are the models wrong or is this divergence due to internal variability? What are some of the potential implications if the former is true? 

Recent research has shown that the three tropical oceans are more interconnected than previously thought. For example, SST warming in the tropical Indian and Atlantic Oceans can influence the Pacific Walker circulation, strengthening trade winds and cooling the equatorial Pacific, with significant implications for ENSO and both regional and global climate. Our research aims to better characterize these interbasin relationships, particularly in response to climate change.

A mismatch between historical climate model simulations and observations reduces confidence in future projections. Even when models replicate past patterns, limited instrumental records constrain our ability to distinguish natural variability from anthropogenic climate change. We are engaged in several projects focused on the continental United States, Southeast Asia, and West Africa that leverage long proxy records, specifically tree-ring data, to bridge this gap. 

We also do climate change attribution for extreme weather and climate events and assess future risks. Our ongoing projects address various extremes, including tornadoes, extreme precipitation, and tropical cyclones (TCs). Our work on the latter is currently the most extensive, encompassing several strands, i.e., downscaling CMIP6 projections of TC frequency, integrating the results with an industry catastrophe model to estimate future financial risks and to further downscale TC-induced precipitation and model storm surge. This comprehensive framework allows us to analyze TC-driven spill dynamics for aboveground petrochemical storage tanks along the U.S. Gulf Coast and assess their potential human health impacts.

Land use and land cover changes (LULCCs) are significant drivers of climate change, yet their impacts remain one of the most uncertain aspects of climate projections. This uncertainty stems partly from fewer studies and experimental frameworks dedicated to characterizing land-driven uncertainties in global climate models in comparison to those emanating from the atmosphere and oceans. To help address this need, our group is engaged in research that emphasizes distinguishing anthropogenic versus natural land use dynamics and evaluating their effects on hydroclimate variability across different spatial scales.