top of page

Your certificate is now private

CertificateBackground.png

Certificate of Completion

ErdosHorizontal.png

THIS ACKNOWLEDGES THAT

HAS COMPLETED THE FALL 2026 DATA SCIENCE BOOT CAMP

Andrea Lomelin

Roman Holowinsky, PhD

November 15, 2026

DIRECTOR

DATE

clear.png

TEAM

Data-Driven Environmental Infrastructure Placement

Andrea Lomelin,Charlsey Tomassetti,Karuna Sangam,Gavin Niendorf

clear.png

Can we use machine learning to identify where environmental intervention would have the greatest impact?

This project will use a real agricultural watershed in coastal California to build a Python-based geospatial decision-support framework. We will combine real-world water-quality, agricultural, hydrologic, terrain, soil, weather, infrastructure, and community-vulnerability datasets to understand where polluted runoff poses the greatest environmental risk.

First, we will engineer spatial features and apply supervised learning methods such as regression and/or classification to estimate pollution pressure in areas without direct monitoring. We will compare models and evaluate how well they generalize.

We will then use those predictions to tackle a second problem: where should nature-based treatment infrastructure be placed when environmental benefit, physical feasibility, and community benefit all matter? We will rank candidate locations and compare environment-first, equity-focused, and balanced deployment scenarios.

The case study involves Vegetated Treatment Systems, nature-based features that capture and treat agricultural runoff before it reaches downstream waterways. The project is connected to an active conservation initiative, so successful results could eventually help inform real planning decisions.

Screen Shot 2022-06-03 at 11.31.35 AM.png
github URL

©2017-2026 by The Erdős Institute.

bottom of page