Machine Learning for Environmental Prediction

Machine Learning for Environmental Prediction

Machine Learning for Environmental Prediction

Default Expiration: Does not expire

Earners of this badge have completed a four-part, hands-on machine learning module using real Chesapeake Bay Program water quality monitoring data. Working in Google Colab with LLM-assisted Python coding, participants built a complete ML workflow — from raw data cleaning through model training, interpretation, and cross-station prediction. Participants trained both a Random Forest and a Multilayer Perceptron (MLP) neural network to predict bottom dissolved oxygen from surface water quality variables, evaluated model performance using RMSE and R-squared, interpreted model behavior using SHAP values, and applied saved models to an independent monitoring station. The module demonstrates how modern ML tools and AI-assisted coding can accelerate environmental data science research.

About this Credential

Learning Objectives

  1. Apply prompt engineering techniques to generate, debug, and refine Python code for each stage of a machine learning pipeline using an LLM assistant.
  2. Clean and prepare long-term environmental monitoring data for machine learning by handling missing values, duplicates, layer/parameter filtering, date completeness checks, and long-to-wide data transformation.
  3. Train and evaluate a Random Forest regression model using scikit-learn, visualize decision tree structure, and assess performance with RMSE and R-squared.
  4. Train and evaluate a Multilayer Perceptron (MLP) neural network using TensorFlow/Keras, apply feature scaling, configure early stopping, and interpret training and validation loss curves.
  5. Apply SHAP (SHapley Additive exPlanations) to interpret feature contributions to both Random Forest and MLP model predictions, using summary plots, bar charts, and LOWESS regression visualizations.
  6. Apply saved models to an independent monitoring station for cross-station generalization assessment, and produce comparative time series and scatter plots of true vs. predicted dissolved oxygen for both models.
How to add this type of credential to LinkedIn:
  • Copy the specific credential link for your earned record from your My Credentials page.
  • In LinkedIn, choose Add profile section > Licenses & Certifications.
  • Use this badge title as the certification name and “University of Maryland Center for Environmental Science (UMCES)” as the issuing organization.
  • Paste your unique credential link into “Credential URL”.