El Nino Prediction Challenge: A Classroom Competition Activity
Turn your students into climate scientists with this engaging prediction challenge that uses real NOAA data and mirrors how researchers actually forecast El Nino events.
Activity Overview
In this multi-day activity, student teams analyze historical Pacific Ocean data to make predictions about El Nino/La Nina conditions. Teams compete to create the most accurate prediction model, learning about oceanic-atmospheric interactions along the way.
How It Works
Students receive historical data from a specific time period—sea surface temperatures, atmospheric pressure readings, trade wind measurements—but the year is hidden. Their task is to analyze the data and predict: Will an El Nino, La Nina, or neutral condition develop in the coming months?
Phase 1: Data Exploration (Day 1)
Teams receive their data package and begin exploratory analysis. They identify patterns in sea surface temperature anomalies, map atmospheric pressure systems, and examine trade wind patterns. The Data in the Classroom interactive tools help students visualize these complex datasets.
Phase 2: Model Development (Day 2)
Teams develop their prediction approach. Some may focus on the Southern Oscillation Index (SOI), others on sea surface temperature patterns in specific regions. Students learn that different indicators have different lead times and reliability levels.
Phase 3: Prediction & Justification (Day 3)
Teams submit their predictions along with written justifications explaining their reasoning. Predictions must include: the expected condition (El Nino/La Nina/Neutral), the expected intensity (weak/moderate/strong), and their confidence level (high/medium/low).
Phase 4: Reveal & Analysis (Day 4-5)
The teacher reveals the actual outcome. Teams analyze why their predictions were accurate or inaccurate. Class discussion focuses on:
- Which indicators were most reliable?
- What data would have improved predictions?
- How do real climate scientists handle uncertainty?
- What are the real-world implications of accurate (or inaccurate) predictions?
Learning Objectives
This activity addresses multiple NGSS standards while developing critical thinking and collaboration skills:
Differentiation Strategies
The activity can be adapted for different skill levels:
For Beginners
Provide more structured data sets, pre-made visualization tools, and explicit guidance on which indicators to examine.
For Advanced
Have students access raw NOAA data, calculate their own anomalies, and develop quantitative prediction criteria.
Assessment Options
The prediction justification serves as a rich formative assessment. For summative assessment, consider having students individually write a reflection on what they learned about climate prediction and uncertainty.
Continue Reading
Tool 1: Google Earth Engine
What it is: A cloud-based platform for analyzing satellite imagery. Satellite images constantly capture Earth's surface, showing changes in vegetation, water, ice, human development, and more.
How it works: Google Earth Engine combines satellite imagery from multiple sources (Landsat, Sentinel, MODIS) with cloud computing. Rather than downloading massive files, you access imagery through a browser and write simple code to analyze it.
For environmental science: You can analyze forest extent, track deforestation, monitor ocean temperatures, identify algal blooms, map agricultural changes, study glacier retreat. Any environmental phenomenon visible from space can be studied.
Grade level: High school (with instructor support) or advanced middle school
Setup complexity: Moderate. Requires creating a Google Account and learning basic coding syntax. But the interface is intuitive and there are many tutorials.
Cost: Free
Why it's great: Real satellite data. Genuine cutting-edge tools that environmental scientists actually use. Students can answer real questions about environmental change. The scope is global.
What to know: Requires some comfort with code. Start with provided examples rather than asking students to write code from scratch. Some processing tasks are computationally intensive and require patience.
Tool 2: Kaggle
What it is: A platform with thousands of datasets and coding challenges. Kaggle hosts open environmental datasets alongside tutorials, code examples, and competitions.
How it works: You browse available datasets, download them, and analyze using Python, R, or other tools. Kaggle provides notebooks where you can code directly in your browser.
For environmental science: Datasets on climate change, species conservation, ocean health, renewable energy, environmental disasters. You can find data about virtually any environmental topic.
Grade level: High school and college-prep
Setup complexity: Low to moderate. Creating an account is simple. Coding requires Python/R experience. But many examples are provided.
Cost: Free (with paid premium tier)
Why it's great: Enormous dataset collection. Community-contributed solutions you can learn from. Competition format motivates some students. No download/upload hassles.
What to know: Community code is sometimes excellent, sometimes poor quality. Teach students to evaluate what they find. Not all datasets are well-documented. Some data exploration is needed to understand structure.
Tool 3: Tableau Public
What it is: A free data visualization and exploration tool. Create interactive dashboards exploring data visually, with built-in AI features suggesting visualizations.
How it works: Upload data, drag variables into the visualization interface, and create interactive charts, maps, and dashboards. Share publicly or keep private.
For environmental science: Visualizing environmental datasets. Creating dashboards showing environmental trends. Telling stories with environmental data. Excellent for presenting findings.
Grade level: Middle school and up (with guidance for younger students)
Setup complexity: Low. Interface is visual rather than code-based. Learning by exploration works well.
Cost: Free (with public data)
Why it's great: Highly visual. Creates impressive-looking dashboards. Encourages exploration. AI-assisted visualization suggestions introduce machine learning in low-stress way.
What to know: The free version requires published data to be public. Visualizations are impressive but might oversimplify complex data. Students should understand what visualizations don't show.
Tool 4: Python with Scikit-learn
What it is: Programming language with machine learning libraries. For serious data science work.
How it works: Write Python code using scikit-learn to build machine learning models. Install libraries, write code, run analyses.
For environmental science: Building predictive models from environmental data. Classification (is this water polluted or clean?). Clustering (group ecosystems by similarity). Anomaly detection (find unusual environmental events).
Grade level: Advanced high school (AP Computer Science, data science courses)
Setup complexity: Moderate to high. Requires installing Python and libraries. Learning Python syntax takes time. But incredibly powerful once learned.
Cost: Free
Why it's great: Real tool that professional data scientists use. Students who learn Python here gain a genuine career skill. Nearly unlimited capability.
What to know: Steeper learning curve. Requires solid programming foundation. Best for students already interested in computer science. Reward for effort is substantial.
Tool 5: NOAA Data Tools
What it is: National Oceanic and Atmospheric Administration provides direct access to environmental data plus tools for visualization and analysis.
How it works: Access NOAA's data portals (Climate Data Online, Earth Systems Research Laboratories, etc.). Download data. Use NOAA's visualization tools or import into analysis platforms.
For environmental science: Direct access to temperature data, precipitation records, ocean measurements, atmospheric data. Data from government agencies, vetted and well-documented.
Grade level: All levels, with adjustments for sophistication
Setup complexity: Low. Data is well-documented. Visualization tools are straightforward. No coding required.
Cost: Free
Why it's great: Authoritative source. Real government data. Covers nearly any environmental topic. Excellent documentation. Local data available (students can analyze their own region's weather).
What to know: Vast amount of data available. Beginners might feel overwhelmed. Help students identify specific datasets to focus on. Some data formats are technical (NetCDF format requires tools to open).
Choosing Tools for Your Classroom
Which tools work for you depends on your students' experience and your learning goals:
For visualization and exploration: Start with Tableau Public or NOAA Data Tools. Low learning curve, high visual impact, builds engagement.
For real data science: Teach Python with scikit-learn. More effort, but genuine skill development.
For authentic environmental analysis: Use Google Earth Engine. Cutting-edge, powerful, genuinely useful.
For large dataset exploration: Kaggle provides everything in one place.
For starting small: NOAA Data Tools or Tableau Public. Easiest entry points.
My recommendation: Start with NOAA data and Tableau Public. They're accessible, authentic, and engaging. Once students are comfortable, introduce more sophisticated tools.
The point isn't learning specific tools—it's developing data literacy and understanding how AI helps us analyze environmental change. Choose tools that support those learning goals.