Modeling Phytoplankton: Covariance and Variogram Model Specification for Phytoplankton Levels in Lake Michigan

Abstract

Algae and phytoplankton are crucial elements of marine ecosystems and of the global carbon cycle, which engenders widespread interest in better understanding their spatial and temporal variability. In situ fluorometry provides detailed measurements of phytoplankton levels; appropriate statistical models are necessary in order to elicit information about the distribution of phytoplankton biomass from this data. Challenges associated with such a data analysis include covariance model specification for processes in which variation in the vertical and horizontal directions differ greatly. Though the ideas presented here were developed with an eye to understanding phytoplankton dynamics, they may be helpful in developing models for other geophysical and environmental processes measured along vertical and horizontal dimensions.