Story 5: Climate & Storms
Temperature-Storm Correlations
Statistical relationships between rising temperatures and storm activity.
Focus window: 1995–present (last 30 years)
Note: Pearson correlation means two variables tend to move together, but it does not prove one causes the other. A third factor can drive both (like warmer oceans and storm energy), the direction can be reversed, or it can be a coincidence. To show causation you usually need controlled experiments or strong evidence that rules out other explanations.
How Correlation Is Calculated
r = cov(X, Y) / (σₓ · σᵧ)
1. Center each series: (xᵢ − x̄), (yᵢ − ȳ)
2. Multiply pairs and average → covariance
3. Divide by both standard deviations
r = 1 → perfect positive
r = 0 → no linear link
r = −1 → perfect negative
|r| closer to 1 = stronger
Correlation Ladder (0 = none, 1 = strong)
1.0
0.0
Temp ↔ Tornado Count
Strongest relationship
0.852
Temp ↔ Tornado Deaths
Human impact signal
0.699
Temp ↔ Max Hurricane Wind
Peak intensity link
0.515
Temp ↔ Avg Hurricane Wind
Near-zero average
-0.029
📈 What This Means
- • Tornado link is strongest (0.852): Years with higher temps tend to line up with more tornadoes, but the chart alone does not prove cause
- • Hurricane peak wind (0.515): Warmer oceans are associated with stronger peak winds, but other factors also matter
- • Deaths correlate with frequency: More storms can mean more exposure, which can raise fatalities
- • Big idea: Correlation shows a pattern, not a direct cause-and-effect rule
Tornadoes & Temp
1950–2025 (76 years) — overlapping full range
Hurricanes & Temp
1980–2025 (46 years) — satellite era data