Future projects

NOTE:  I am currently looking for students (graduate or postdoctoral) to work on these projects. Contact me for further information.

The evolution of mean tropical Pacific climate over the late Holocene

<>

We plan to collect hundreds of fossil coral cobbles (pictured at left) on a cruise to the Line Islands in August, 2005. Preliminary dates on existing samples suggest that most of the samples will fall within the last several millennia. By producing dozens of short geochemical climate proxy records from the fossil corals, we hope to compile a quantitative reconstruction of mean climate over the last several millennia. The benefit of using dozens of individuals is that we can produce a quantitative estimate of the error of the reconstruction, instead of relying on the assumption that a handful of fossil coral records are representative of large periods of geologic time. Depending on the length of the average record, we might also be able to constrain the evolution of seasonal to interannual climate variability over the same time period. Prospective postdocs should contact me about working on this project.

Late 20th century trends in tropical Pacific climate and upwelling

Figure 2b of McPhaden & Zhang, Nature, 2002, showing a decrease in convergence at the equatorial Pacific (green crosses) and a regional warming (black line).

An abundance of evidence now suggests that climate patterns throughout the entire Pacific ocean underwent a marked change in the mid-1970's towards an El Nino-like pattern. The so-called "regime shift" likely reflects the combined effects of natural Pacific climate variability (the Pacific Decadal Oscillation) and anthropogenic climate change. The dynamics underlying this transition, however, remain ambiguous, as pre-1970 oceanographic observations do not exist for most of the Pacific ocean. Several lines of evidence suggest that dynamics in the tropical Pacific, specifically wind-driven equatorial upwelling, may have played a key role in shaping the evolution of late 20th century climate patterns (see graph at left). As tropical Pacific upwelling plays a key role in the global carbon cycle, such circulation changes have direct consequences for the fate of anthropogenic carbon emissions.

We intend to reconstruct the climate, circulation, and carbonate geochemistry of the surface waters of the equatorial Pacific from 1950 to present, using corals from the central tropical Pacific. Existing coral-based climate proxy records (see panel at right) hint at regional-scale warming and freshening which has accelerated since the mid 1970's. By applying additional geochemical proxies to corals from the central tropical Pacific, we can quantify the magnitude of sea-surface temperature (Sr/Ca and Mg/Ca) and salinity (d18Ow) changes over this period. Furthermore, we can constrain changes in upwelling intensity (Cd/Ca and Ba/Ca) and associated changes in the carbonate chemistry of the surface ocean (boron isotopes) since 1950.

Figure 1a of Cobb et al., GRL, 2001, showing a trend towards lighter oxygen isotopic compositions in a central tropical Pacific coral, likely reflecting a combination of regional warming and freshening, possibly due in part to a reduction in wind-driven upwelling since the mid-1970's.

ENSO-like centennial-scale variability:  data-model synthesis

Paleoclimatic reconstructions of low-frequency climate variability are invariably interpreted with respect to ENSO dynamics, which govern surface temperature, hydrological, and wind patterns throughout the Pacific basin on interannual timescales. While coupled ocean-atmosphere interactions in the tropical Pacific are the largest source of interannual climate variability, their potential to determine global climate patterns on lower frequencies (decadal to millennial) has not yet been well-established. Addressing this problem requires answers to the following two questions:  1. How much does the Pacific zonal SST gradient change on decadal to millennial timescales? and 2. What impact do those changes have on global climate patterns? Figure from Zhang et al., "ENSO-like interdecadal variability:  1900-93", J. Climate, 1997. The first (to my knowledge) use of the increasingly popular phrase "ENSO-like" to describe low-frequency Pacific climate variability. We hope to answer these questions for the last millennium, when data global paleoclimate data coverage is sufficiently rich to make this research feasible. A variety of Pacific paleoclimatic indicators suggest that the early centuries of the last millennium (ie 900-1200A.D., a.k.a. The Medieval Warm Period) may have been characterized by a La Nina-like mean state, while the period from 1500-1800A.D., a.k.a. The Little Ice Age may have seen the Pacific in an El Nino-like mean state. This hypothesis can be tested by quantitatively reconstructing centennial-scale climate changes from ENSO-sensitive areas (using the figure below as a conceptual guide, for example) over the last millennium. The results of the quantitative reconstruction will then be used as boundary conditions for coupled modelling runs using NCAR's coupled GMC (in collaboration with Caspar Amman, NCAR), in order to a) test whether our assumption that centennial-scale ENSO-like teleconnections resemble those of interannual ENSO, and b) constrain the nature of the climate system's response to low-frequency tropical Pacific climate variability.   Figure from Dai and Wigley, GRL, 2001, showing precipitation anomalies associated with ENSO (pink = drought, blue = flooding). Wee hope to take advantage of this relationship to assess the global impact of possible centennial-scale ENSO activity over the last millennium.