Paleoclimate research often requires interpolation of measurement data on the surface of the Earth to a global regular grid. Usually, variables of paleoclimatic interest have a spatial structure best captured by kriging interpolation. However, to date there are no systematized techniques to perform anisotropic kriging on a sphere.

Our objective is to develop numerical recipes to perform this task, so that Earth System Model outputs can be easily compared with observations, and design a MATLAB graphical user interface that allows the user to make their own decisions in the path to obtaining interpolated fields based on their datasets.

While our primary motivation behind this work is in relation to paleoclimatic applications, we are programming this software package so that it may be used with any global-scale dataset of a scalar variable.

The figures on the left are from Lambert et al. (2015):

Lambert, F., A. Tagliabue, G. Shaffer, F. Lamy, G. Winckler, L. Farias, L. Gallardo, and R.De Pol-Holz (2015), Dust fluxes and iron fertilization in Holocene and Last Glacial Maximum climates, Geophys. Res. Lett., 42, 6014–6023, doi:10.1002/2015GL064250.

I collaborate with colleagues from CICTERRA in Argentina (Diego Gaiero, Gabriela Torre and Renata Coppo) on studies of deposition mechanisms of the Pampean loess during the last glacial-interglacial cycle.

The Pampean loess is the most extensive loess deposit in the Southern Hemisphere, and an excellent record of past deposition fluxes of dust derived from sources in southern South America.

These studies provide an excellent opportunity to constrain past dust fluxes coming off South America. However, it is not straightforward to infer past dust emission fluxes from South American sources as measured deposition fluxes in the Pampean region are sensitive not only to those emission fluxes, but also to short-range (<1500 km) transport mechanisms from sources to sink, and to climatological conditions in the sources and in the Pampean region (as well as along the short transport pathways). We use a combination of dating techniques for the loess sediments, isotopic fingerprinting to infer dust provenance, and magnetic properties of sediments to infer past local precipitation conditions.

I work alongside Victoria Nogués and Diego Montecino Jara, two PhD students from the Universidad Nacional de Córdoba (Argentina) whom I advise, in deciphering past dust activity in South America based on studying the Pampean loess.

Does dust and volcanic ash from southern South American enhance primary productivity of the southern oceans in the present-day climate system? How does the solubility of micronutrient iron in South American dust and ash depend on grain size and mineralogy?

I try to answer these questions by performing satellite clorophyll-a analysis of waters impacted by dust storms in the south Atlantic Ocean, coupled with chemical characterization of iron in dust, dust-emitting topsoils and ash.

It appears that most dust coming out of the continent is low in soluble iron (Simonella et al., in prep.), which can explain the lack of response of phytoplankton in a region with low atmospheric processing (see our case study in Cosentino et al., 2020). Ash has higher iron solubility and we are extending our methodology to characterize events of ash deposition to the oceans during volcanic eruptions.