Powerful imaging tools can reveal processes of biological growth and tranformation with unprecedented detail, but require special, computational techniques to process large quantities of data.

To more fully understand the dynamic process of mineralization, in collaboration with my brother Gregory Green, I employ computational techniques including Markov chain Monte Carlo sampling and Bayesian probability frameworks that integrate large and diverse data sets into comprehensive and usable models.

Generic placeholder image

Transverse section capturing synchrotron-based mineral density profiles of herbivore second molars in a range of developmental states.


Modeling mineralization requires methods that can derive complex biological parameters from the integration of many data sources.

I use python coding and nonlinear optimization routines to both understand the process of mineralization over space and time, and to extract seasonal climatic data from mineralizing teeth.

Generic placeholder image

Density maps derived from nonlinear optimization algorithms showing the likelihood of mineralization parameters for herbivore molars.


Past seasonal hydrology is considered a critical part of the environment that shaped human evolution, and can be reconstructed from isotope values in teeth.

The main goal of my research is to reconstruct seasonal rainfall and resource patterns at sites of hominin occupation by combining isotopic data with computational methods that can infer most likely body and drinking water chemistry from abundant large herbivores.

Generic placeholder image

Optimization routine returns a most likely seasonal oxygen isotopic record from drinking water, using tooth isotope measurements in an herbivore tooth.