Intermediate Machine Learning in Python for Environmental Science Problems 2026

The AMS Intermediate Machine Learning in Python for Environmental Science Problems Short Course presents a selection of intermediate-level ML topics for environmental scientists who already have a background in ML fundamentals. The course will be taught by a team of instructors who are presenting course modules related to their ML expertise. Participants will work with real-world, spatio-temporal environmental data and follow along by programming using the Google Colab Python programming environment.

The course will be divided into modules that focus on a single topic, led by one of the instructors. Modules will contain a mix of lecture material and interactive coding. Participants can work together on interactive problem-solving. The planned topics include data processing and handling imbalanced data distributions, architecture selection, uncertain quantification, model evaluation strategies, model interpretability, and generative AI methods.

We assume that the participants have some experience developing ML models in Python. Students are expected to be familiar with TensorFlow, a Python framework for model development. In addition, students should have a basic familiarity with a selection of commonly-used scientific Python modules: numpy, pandas, and matplotlib. For participants who do not have this experience, we highly suggest taking the Short Course on Machine Learning for Weather and Climate offered for free by CIRA. 

106th AMS Annual Meeting
George R. Brown Convention Center
January 25, 2026 at 8:30 AM - 3:45 PM Central Time (Hybrid)

REGISTRATION RATES

AMS Member Early Rates Non-Member Early Rates AMS Student Early Rate AMS Member Late Rate Non-Member Late Rate AMS Student Late Rate
$155 $180 $120 $295 $220 $160
REGISTER NOW


Course Description:

The major goal of this course is to help participants better apply ML for environmental science. With machine learning frameworks in Python and the plethora of tutorials and example code available online, ML has seen widespread adoption across many fields, including environmental science. It has become easy to develop a simple ML model, but there are many pitfalls that can make it very difficult to achieve meaningful ML applications for complex environmental science problems. There is a major gap between the simple toy problems presented in beginner’s ML tutorial and the complex environmental science applications: practitioners can be misled by the commonly recommended evaluation metrics, loss functions, explainability techniques, etc. Here, we will present strategies developing ML models and verifying their performance for complex spatial-temporal environmental applications. Our goal is share our experiences in applying ML to environmental science, highlighting key strategies and pitfalls to guide researchers toward successful applications.

Participants will learn:

  1. Strategies for dealing with highly imbalanced datasets from model development, evaluation, and interpretation.
  2. Strategies and best-practices for selecting model architectures suitable for their problem.
  3. Strategies for uncertainty quantification (UQ), including awareness of common pitfalls that can yield very misleading UQ especially when extrapolating from the distribution seen during training.
  4. Understanding explainable AI methods and how to apply them for environmental science models. Also, understanding their limitations and pitfalls especially for spatio-temporal data with strong autocorrelation (i.e. much of environmental data), as well as strategies for extracting meaningful insights into the models despite those challenges.
  5. An introduction to the generative AI methods behind recent rapid advances in ML, and how they can be applied to environmental science problems.

VIEW AGENDA

If you have questions regarding the course, please contact Evan Krell.

Instructors:

Evan Krell

Naval Research Lab - Marine Meteorology Division

Jhayron Steven Perez Carrasquilla

University of Maryland

Ryan Lagerquist

MyRadar

Piyush Garg

RWE

Fraser King

University of Wisconsin-Madison