Columbia's IGERT program sponsored Princeton Geosciences undergraduate Dalia Bach as a 2003 summer intern at Lamont-Doherty Earth Observatory. Her research was overseen by Doherty Senior Research Scientists Yochanan Kushnir and Richard Seager.

Using observed and general circulation model data, Dalia studied anomalous warming trends in the southern oceans and their relationship to a multi-decadal wetting and drying period in the Sahel region of sub-Saharan Africa.

Dalia's study, "Sahel Precipitation Variability from Sea Surface Temperature Forcing in Isolated Ocean Basins," found that sea surface temperature forcing in the Indian Ocean was the major factor driving Sahel drought conditions in the 1970s and 1980s. However, she also concluded that political and socioeconomic factors were much more influential in determining the extent to which a drought causes either a mere food shortage or a widespread famine.

Dalia was accepted as a graduate student in Columbia's Department of Earth & Environmental Sciences for Fall 2004.

IGERT sponsored College of William & Mary undergraduate Marley Bice, a Geology and Public Policy double major, as our 2006 Lamont-Doherty summer intern. Her research was overseen by Associate Professor Peter de Menocal and IGERT fellow Gemma Kirkwood.

Reconstructing sea-surface salinity of Mid-Atlantic using Mg/Ca and d 18 O ratios of G. ruber

This project was a test of how well using coupled Mg/Ca and d 18 O measurements on Globigerinoides ruber could isolate the salinity signal, or d 18 O of sea water. I picked approximately 100 white rubers from a total of 78 trigger-weight top cores along the mid-Atlantic ridge from 45 o N to 35 o S. The first data I got was SST from Mg/Ca and this data showed a deviation from the expected bell-shaped curve of several degrees Celsius in both the Northern and Southern sub-tropics. However, the data align well from about 10 o S to 10 o N. This led to a study of possible connections between Mg/Ca ratios and sedimentation rate, ruber abundance and habitat, and carbonate chemistry of the oceans. The d 18 O values plotted against Latitude showed a better correlation to the expected bell-shaped curve than the Mg/Ca data. The true test of the process was how well the calculated d 18 O of seawater from Mg/Ca SST and d 18 O of calcite reflected the expected salinity. This data correlates generally with the expected salinity with a R 2 value of 0.31 meaning 31 % of the variance correlates to salinity. Using Mg/Ca ratios of G. ruber as a proxy for SST does not appear to work in the sub-tropics and cannot be fully explained now, suggesting that another variable may be affecting the way Mg/Ca is incorporated into the foraminifera shell. For example, when the data is corrected for a salinity effect it shows a better correlation to the expected SST.

IGERT sponsored Vassar undergraduate Joshua Dorin, a Geology and Mathematics double major, as our 2004 Lamont-Doherty summer intern. His research was overseen by Doherty Senior Research Scientists Yochanan Kushnir and Richard Seager.

Josh analyzed observational rainfall data and optimally interpolated marine observations of sea surface temperature (SST) and sea level pressure (SLP) in computer model simulations with the NCAR CCM3 Atmospheric General Circulation Model. His study, "Examining the Relationship Between the Indian Monsoon and ENSO," found that most droughts in India are a result of the El Nino Southern Oscillation phenomenon.

However, he also observed that when Indian Ocean SSTs reached record highs in 1982-83 and 1997-98, India did not experience severe droughts. His current hypothesis is that when the Indian Ocean is relatively warm, Indian droughts are reduced. Further investigation is needed.

Columbia's IGERT program sponsored University of Chicago's undergraduate Eric Jacobsen as a 2005 summer intern at Lamont-Doherty Earth Observatory. His research was overseen by Doherty Senior Research Scientists Yochanan Kushnir and Richard Seager.

Using observational datasets, Eric's research showed that precipitation variability in the Middle East/Central Asia correlated with both the El Nino Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO), but more heavily on the latter.

Eric's study also focused on assessing the suitability of GOGA/POGA climate models for studying the atmospheric dynamics behind precipitation variability in the Middle East.

Eric plans to continue his studies in geosciences next year.

IGERT sponsored Columbia undergraduate Applied Math major Robin Koytcheff to do summer research in the Columbia Math Department's VIGRE Program 2004.

Robin collaborated with Professor Chris Wiggins of the Applied Mathematics department, and with several other students, on a project relating information theory to graph partitioning. His research focused on theoretically justifying why information-theoretic algorithms, which are normally used to efficiently encode signals, give good partitions when applied to graphs.

In addition to conceptual problem-solving attempts, Robin performed numerical experiments in Matlab in order to better understand the problem. In August he presented the project at Mathfest, a three-day conference for students and faculty. As of early fall, 2004, the research is still in progress, but publishable results are expected soon.

Robin plans to continue his studies in math or applied math as a graduate student next year.

The IGERT program sponsored Columbia University undergraduate Applied Mathematics major Timothy Merlis to do summer research in 2005 with Doherty Associate Research Scientist Samar Khatiwala and Fu Foundation Professor of Applied Mathematics David Keyes.

Tim worked on applying Jacobian-Free Newton-Krylov, a leading numerical method, to finding the steady-state solution of a transient fluid problem, the lid and thermally driven cavity.

In this idealized setting, the study sought efficient and parallelizable techniques to computing equilibrium solutions of steady and periodically forced systems that can be readily transferred to other geophysical problems, such as ocean general circulation models. Initial results from applying the technique to the prototype fluid problem are promising, and work is underway to extend these results to more challenging geophysical problems.

Tim plans to continue pursuing interdisciplinary problems in earth science and applied mathematics as a graduate student.

The IGERT program sponsored Columbia University undergraduate Applied Mathematics major Sergey Voronin to do summer research in 2005 with Fu Foundation Professor of Applied Mathematics Lorenzo Polvani.

Sergey helped prepare a suite of graphical results for visualizing major, mid-winter, stratospheric sudden warming events and a website for presenting these and other results, which may be reached at: http://www.appmath.columbia.edu/ssws/.

Sergey currently continues his studies at Columbia and wants to keep applying computers to applications in Applied Mathematics.

Columbia's IGERT program sponsored Cornell University undergraduate Allison Wing, an Atmospheric Science major, as a 2006 summer intern at Lamont-Doherty Earth Observatory. Her research was overseen by Associate Professor Adam Sobel and Associate Research Scientist Suzana Camargo.

Allison's research examined the relationship between actual and potential intensities of tropical cyclones. Based on the Carnot cycle, the potential intensity (PI) theory finds an upper bound on intensity using only thermodynamic variables. Allison tested this PI theory by analyzing its relation to and correlation with actual tropical cyclone wind speeds in the time period 1950-2005. Previous studies found that wind speeds follow a uniform distribution up until their theoretical maximum (PI), suggesting that a climatic change in PI would affect the intensity distribution of real storms uniformly. Thus, Allison examined if fluctuations in actual wind speeds were indeed governed by fluctuations in PI on interannual time scales. After comparing yearly average time series of wind speed and PI, she found that the two were correlated fairly well for most breakdowns of the data. However, her results indicated that actual wind speed was varying with PI less than the theory predicts, by about a factor of two. Allson will continue her studies in Atmospheric Science at Cornell University as a junior in Fall 2006.