Mathematical modeling and data analysis play a critical role in the mathematics of Planet Earth. This theme was brought home in a panel discussion “Big Data Meets Big Models,” particularly in the presentation by Anna Michalak.
The general public is generally aware of how models are used for weather prediction, but perhaps a bit less aware of how modeling and the ability to process and analyze large data sets plays a critical role in climate science. However, mathematical models are necessary to understand the many aspects of climate science and underlie our ability to predict future changes. One example lies in the complex interplay of carbon – its transition from sources (like the burning of fossil fuels) through the ocean, air, land and biosphere.
A critical component to understanding the role of carbon dioxide (CO2) in global warming is this global carbon cycle – the transmission of carbon through ocean, atmosphere, and land. Human activities produce several gigatons of carbon per year; some fraction of this is absorbed by plants, oceans, and other mechanisms. Having a better quantitative understanding of the natural carbon sinks is essential for better predictions of the future. This leads to a need for better measurements of the carbon cycle to ensure that we have good data upon which to base our models. And there is also a need for improved monitoring as states may agree to limit carbon emissions.
This has led to a major infrastructure project to gather data from observations. FluxNet, a “network of regional networks,” coordinates regional and global analysis of observations from micro-meteorological tower sites. The flux tower sites use eddy covariance methods to measure the exchanges of carbon dioxide, water vapor, and energy between terrestrial ecosystems and the atmosphere.
The idea is to gather massive amounts of data on CO2 along with data on rainfall and fires to better calibrate carbon transmission. The various sites around the world will gather vast amounts of data. Since we often can’t obtain data on the quantities we want but rather on a related variable, there is a need for improved mathematical models related to these variables, in order to make sense of the massive amounts of data to be collected. Thus mathematical models will be required to fill in the gaps (since data is only located at spatially dispersed sites, for example) – an example of the interplay between mathematical models, massive data sets, and climate science.