Internships

Saumya Malik ’24

Major

Computer Science

Project Title

Can We Predict Primary Production in the Ocean?

Presentation Link

View Saumya's Presentation

My project aimed to quantify the predictability of multiple biogeochemical drivers of the ocean ecosystem. One driver that I focused on is net primary production, which is the amount of biomass or carbon produced by primary producers per unit area and time; this is estimated by subtracting plant respiratory costs from gross primary productivity or total photosynthesis. Quantifying the predictability of these drivers allows stakeholders like fisheries to have a better understanding of forecasting capabilities and thus improve their management. My work involved analyzing the data from a new set of simulation experiments run on the Geophysical Fluid Dynamics Laboratory’s Earth System Model. I wrote code in the program Python to perform calculations of metrics like prognostic potential predictability (PPP) on 300 years’ worth of simulation data. I produced many plots that visualize the PPP of multiple variables in different ways — taking averages over regions and globally, looking at individual grid points over time, and making animations of global maps over time. By the end of the project, I was able to quantify the predictability of many variables and I gained an understanding of the many interesting ways of looking at predictability, which was a fascinating realization for me.