Lesson 6: AI in Climate Science and Prediction

Quick Reference Guide

For students who finish early or want to explore deeper:

Activity	Time Required	Difficulty	Skills Developed
Citizen Science Data Collection	Ongoing	Beginner	Data collection, scientific method
Build a Simple AI Model	2-3 hours	Intermediate	Coding, AI concepts
Climate Solutions Challenge	3-5 hours	Intermediate	Creativity, problem-solving
Deep Dive Research Project	1-2 weeks	Advanced	Research, writing, analysis
Career Exploration Portfolio	2-4 hours	All levels	Career planning, research

Extension Activity 1: Citizen Science Projects

Overview

Participate in real scientific research by collecting climate data that feeds into AI-powered systems.

Recommended Projects

NASA GLOBE Observer

• What: Record cloud observations, land cover, and habitat data
• How: Free mobile app (iOS/Android)
• Time: 5-10 minutes per observation
• Learning: Your data helps validate satellite observations and improve AI models
• Website: observer.globe.gov
• Assignment: Make 10 observations over 2 weeks, analyze your own data patterns

CoCoRaHS (Community Collaborative Rain, Hail & Snow Network)

• What: Report daily precipitation measurements
• How: Install simple rain gauge ($30) or use existing
• Time: 2 minutes daily
• Learning: Understand data collection protocols, see how your data appears on national maps
• Website: cocorahs.org
• Assignment: Document one month of precipitation, compare to nearby stations

iNaturalist

• What: Document plants, animals, insects in your area
• How: Mobile app with AI-powered species identification
• Time: Variable
• Learning: See how AI assists biodiversity tracking; contribute to phenology studies (timing of biological events affected by climate)
• Website: inaturalist.org
• Assignment: Document 25 species, research how climate change affects one species you observed

eBird (Cornell Lab of Ornithology)

• What: Report bird sightings
• How: Mobile app
• Learning: Track bird migration patterns affected by climate change
• Website: ebird.org

Reflection Questions

1. How does your data contribute to larger AI-powered climate monitoring systems?
2. What challenges did you encounter in collecting accurate data?
3. How might bias in citizen science data affect AI models trained on it?
4. What patterns did you notice in your own observations?

Extension Activity 2: Build a Simple Predictive Model

Overview

Create a basic machine learning model that makes predictions from climate data using beginner-friendly platforms.

Option A: Google's Teachable Machine (No Coding)

Project: Weather Pattern Classifier

Materials Needed:

• Computer with webcam
• Internet access
• Weather images (collect from internet or windows)

Steps:

1. Go to teachablemachine.withgoogle.com
2. Create "Image Project"
3. Create categories: "Sunny," "Cloudy," "Stormy"
4. Collect 50+ images per category (take photos or use online images)
5. Train model (takes 2-5 minutes)
6. Test with new images
7. Export model

Analysis Questions:

• How accurate is your model? What images confuse it?
• How is this similar to how AI analyzes satellite imagery?
• What would you need to make this useful for actual weather prediction?

Option B: Excel/Sheets Trend Prediction (Intermediate)

Project: Predict Next Year's Temperature

Data Needed: Historical temperature data (provided in datasets)

Steps:

1. Plot temperature data over time (line graph)
2. Add trendline: Right-click line → Add Trendline → Linear
3. Check "Display Equation" and "Display R-squared"
4. Use equation to predict future values
5. Compare your prediction to actual data
6. Try different trendline types (linear, polynomial, exponential)

Discussion:

• Which trendline fits best? How do you know?
• How confident are you in predictions 1 year out? 10 years out? 50 years out?
• How does this simple model compare to sophisticated AI climate models?

Option C: Python Programming (Advanced)

Project: Temperature Prediction with Machine Learning

Prerequisites: Basic Python knowledge

Tools: Google Colab (free, no installation needed)

Simplified Code Template Provided (see GitHub repository)

What You'll Learn:

• Load climate data with Pandas
• Create scatter plots with Matplotlib
• Build linear regression model with Scikit-learn
• Make predictions
• Evaluate model accuracy

Resources:

• Tutorial video: "Climate Data Analysis with Python"
• Code template with comments explaining each step
• Dataset: temperature_data.csv (included)

Extension Activity 3: Climate Solutions Entrepreneurship Challenge

Overview

Design an AI-powered solution to a specific climate problem, create a proposal, and pitch it.

Challenge Structure

Phase 1: Problem Selection (30 minutes)

Choose one of these problems or identify your own:

• Reducing food waste in school cafeterias
• Optimizing urban tree planting for maximum cooling
• Predicting and preventing pipeline leaks (methane emissions)
• Helping farmers decide optimal planting times
• Improving electric bus route efficiency
• Identifying buildings wasting energy

Phase 2: Research (1-2 hours)

Investigate:

• What data exists about this problem?
• Who else is working on this?
• What technologies are currently used?
• What gap could AI fill?
• Who would benefit from your solution?

Phase 3: Solution Design (1-2 hours)

Create proposal including:

1. Problem Statement: Clear description of issue (with data)
2. AI Solution: How would AI help? What would it analyze? What decisions would it make?
3. Data Requirements: What data is needed? Where would it come from?
4. Implementation: How would it work in practice?
5. Impact: How much carbon could be saved? Who benefits?
6. Cost Estimate: Rough budget (be realistic)
7. Potential Challenges: What could go wrong?

Phase 4: Create Pitch (30-60 minutes)

• 3-minute presentation
• Visual aid (poster, slide, prototype)
• Clear value proposition
• Call to action

Phase 5: Present (Class Presentations)

• Shark Tank style presentations
• Peer feedback
• Teacher/guest judge evaluation

Evaluation Criteria

• Innovation and creativity
• Feasibility
• Potential impact
• Quality of research
• Presentation effectiveness

Extension: Submit to Competitions

• National STEM Competition
• Climate Change AI Innovation Challenge
• Local business plan competitions

Extension Activity 4: Deep Dive Research Projects

Advanced Research Topics

For students interested in pursuing deeper investigation:

Option 1: Climate AI Ethics Analysis

Research Question: Should AI predictions influence climate policy?

Key Issues to Explore:

• Accountability when AI predictions are wrong
• Access equity (rich vs. poor countries)
• Indigenous knowledge vs. AI data
• Privacy concerns in environmental monitoring
• Decision-making authority: humans vs. algorithms

Deliverable: 5-7 page analytical essay with annotated bibliography (10+ sources)

Option 2: Comparative Model Analysis

Research Question: How do different AI climate models compare?

Analysis Tasks:

• Compare 3+ climate AI models (DeepMind, IBM, NOAA)
• Evaluate prediction accuracy
• Assess computational requirements
• Analyze what each does best
• Recommend optimal use cases

Deliverable: Technical report with data tables, graphs, comparison matrix

Option 3: AI and Climate Justice

Research Question: How can AI address or worsen climate inequity?

Investigation Areas:

• Technology access gaps
• Data representation (Global South vs. North)
• Who benefits from AI climate solutions?
• Case studies of AI enabling or harming marginalized communities
• Policy recommendations

Deliverable: Position paper with policy recommendations

Option 4: Career Path Investigation

Research Question: What skills are needed for AI + Climate careers?

Research Activities:

• Interview 3+ professionals in field
• Analyze job postings
• Map educational pathways
• Identify skill gaps and learning resources
• Create personal development plan

Deliverable: Career exploration portfolio with action plan

Extension Activity 5: Cross-Curricular Connections

Mathematics Integration

Project: Statistical Analysis of Climate Data

Activities:

• Calculate mean, median, mode of temperature datasets
• Determine standard deviation and confidence intervals
• Perform correlation analysis (temperature vs. CO2)
• Create regression models
• Analyze uncertainty in predictions
• Compare different statistical models

Learning Objectives:

• Apply statistical concepts to real data
• Understand uncertainty quantification
• Practice data analysis skills

Writing/English Integration

Project: Climate Communication Campaign

Activities:

• Analyze how climate change is communicated in different media
• Write op-ed about AI's role in climate solutions
• Create infographic explaining complex climate science
• Develop social media campaign communicating climate data
• Compare scientific papers to news articles (analyze translation of technical content)

Learning Objectives:

• Effective science communication
• Audience awareness
• Persuasive writing
• Media literacy

Social Studies Integration

Project: Climate Policy Analysis

Activities:

• Research climate policies in different countries
• Analyze how AI data influences policy decisions
• Investigate climate justice movements
• Compare international climate agreements
• Study economic impacts of climate action/inaction

Learning Objectives:

• Understanding policy-making process
• Global perspectives
• Economic analysis
• Historical context

Computer Science Integration

Project: Build Climate Data Dashboard

Activities:

• Learn HTML/CSS/JavaScript basics
• Create interactive climate data visualizations
• Build website displaying real-time climate data
• Implement data API connections
• Design user-friendly interface

Learning Objectives:

• Web development skills
• Data visualization programming
• API usage
• User experience design

Long-term Projects (Multi-week or Semester-long)

Project 1: Local Climate Assessment

Duration: 6-8 weeks

Objective: Analyze climate change impacts in your local community

Tasks:

1. Research historical climate data for your region
2. Interview long-time residents about observed changes
3. Analyze local temperature and precipitation trends
4. Investigate impacts on local ecosystems, agriculture, infrastructure
5. Document with photos, videos, data visualizations
6. Present findings to city council or community groups
7. Propose AI tools that could help community adapt

Outcomes:

• Detailed local climate report
• Community presentation
• Recommendations for local action
• Potential media coverage
• Service learning credit

Project 2: School Sustainability Audit with AI

Duration: Full semester

Objective: Use AI tools to optimize school's environmental footprint

Tasks:

1. Audit energy use, waste, transportation, water usage
2. Collect baseline data
3. Identify AI tools that could optimize each area
4. Test pilot programs (e.g., optimal thermostat scheduling)
5. Calculate potential cost and carbon savings
6. Present business case to administration

Outcomes:

• School sustainability report
• Implemented pilot programs
• Measurable impact
• Real-world problem-solving experience

Resources for Extended Learning

Online Courses (Free)

Beginner Level:

• "AI for Everyone" by Andrew Ng (Coursera) - 4 weeks
• "Climate Change: The Science" by UBC (edX) - 6 weeks
• "Data Science for Everyone" (DataCamp) - Self-paced

Intermediate Level:

• "Machine Learning Crash Course" by Google - 15 hours
• "Python for Data Science" (Kaggle) - 6 hours
• "Climate Change Science and Negotiations" by MIT - 12 weeks

Advanced Level:

• "Deep Learning Specialization" by Andrew Ng - 3 months
• "Climate Change AI" summer school - Annual program

Books

For Students:

• "The Uninhabitable Earth" by David Wallace-Wells (accessible climate science)
• "Drawdown" by Paul Hawken (solution-focused)
• "Machine Learning for Kids" by Dale Lane (AI basics)

For Advanced Students:

• "Climate Change: What Everyone Needs to Know" by Joseph Romm
• "Artificial Intelligence: A Guide for Thinking Humans" by Melanie Mitchell
• "The Future We Choose" by Christiana Figueres

Competitions and Challenges

For High School Students:

• Climate Change AI Innovation Challenge
• Conrad Challenge (Climate, Energy & Environment category)
• GLOBE Student Research Competition
• NSTA Student Research Competitions
• Google Science Fair

Summer Programs

STEM Summer Programs with Climate/AI Focus:

• NASA SEES Summer Internship
• NOAA Hollings Scholarship Program
• AI4ALL Summer Programs
• Climate Corps
• Various university summer research programs

Assessment for Extension Activities

Extension Activity Evaluation Form

Criteria	Points	Student Self-Assessment	Teacher Assessment
Depth of investigation	/10
Quality of work product	/10
Independence and initiative	/5
Time management	/5
Reflection on learning	/5
Connection to lesson content	/5
Total	/40

Reflection Questions:

1. What did you learn from this extension activity that you wouldn't have learned from the regular lesson?
2. What challenges did you overcome?
3. How does this connect to potential career interests?
4. What would you want to investigate next?

© 2025 Evolve AI Institute LLC. May be reproduced for educational use.

Lesson 6: AI in Climate Science and Prediction