SUMMARY - Dealing with Uncertainty: What Science Can—and Can’t—Predict

Climate science offers perhaps the most significant predictions ever attempted—projections of planetary change spanning decades to centuries that could reshape civilization itself. Yet these predictions come hedged with caveats, ranges, and probabilities that sometimes frustrate those seeking definitive answers. Understanding what science can and cannot predict about our climate future is essential for navigating the difficult decisions ahead.

What Climate Models Actually Do

Climate models simulate Earth's climate system mathematically—atmosphere, oceans, land surfaces, ice sheets, and increasingly, ecosystems and human activities. They divide the planet into grid cells and calculate physical processes: how heat moves, how water evaporates and precipitates, how carbon cycles through the system. These calculations, repeated for simulated decades, project how climate might evolve.

Models don't claim to predict specific weather on specific days far in the future. Weather is inherently chaotic beyond a week or two. Climate, however—the statistical summary of weather over time—has more predictable properties. We cannot say whether December 15, 2050 will be rainy, but we can project that winters will likely be warmer and wetter or drier in various regions.

Multiple independent modeling groups worldwide develop their own climate models. That these different approaches produce broadly consistent results increases confidence in the projections. Where models disagree highlights genuine scientific uncertainty rather than fundamental ignorance.

Emissions Scenarios and Human Choice

The largest uncertainty in climate projections isn't physics—it's human behaviour. How much carbon dioxide will we emit? Will economies decarbonize rapidly or continue business as usual? Will population growth continue or stabilize? These choices haven't been made yet; they depend on policy, technology, economics, and culture that cannot be scientifically predicted.

Climate projections therefore explore scenarios rather than making single predictions. Low-emissions scenarios assume aggressive climate action; high-emissions scenarios assume minimal action. The resulting range of warming—roughly 1.5 to 4+ degrees Celsius by 2100—primarily reflects this uncertainty about future human choices, not about climate physics.

This matters for how we interpret projections. The worst scenarios remain avoidable if we choose to avoid them. The best scenarios require deliberate effort. Climate science shows us the consequences of different paths; it cannot tell us which path we will choose.

Tipping Points and Nonlinear Change

Climate systems contain potential tipping points—thresholds beyond which changes become self-reinforcing and irreversible. Ice sheet collapse could accelerate once started. Permafrost thaw releases carbon that causes more warming that causes more thaw. Amazon rainforest dieback could shift regional climate enough to prevent forest recovery.

These tipping points are difficult to predict precisely. We know they exist; we don't know exactly where they lie. Passing a tipping point might trigger rapid change; or change might unfold gradually over centuries. This uncertainty isn't a reason for complacency—if anything, it argues for caution, since we might trigger irreversible changes before recognizing we've done so.

Scientists increasingly study "tipping cascades"—the possibility that triggering one tipping point might trigger others. If Arctic warming accelerates Greenland ice loss, raising sea levels and altering ocean circulation, which affects monsoons and threatens Amazon stability, the cumulative effect could exceed the sum of individual changes.

Regional and Local Prediction

Global average changes are clearer than regional details. We're confident the planet will warm; exactly how rainfall patterns shift in specific regions is harder to project. Mediterranean climates will likely dry; higher latitudes will likely see more precipitation. But precisely where boundaries lie, and how extreme the changes become, involves greater uncertainty.

This matters because people don't experience global averages—they experience local weather. Farmers need to know how their region will change. Cities need to plan for specific flooding or drought risks. The mismatch between what science can confidently predict (global patterns) and what planners need to know (local details) creates frustration.

Downscaling techniques translate global projections to regional scales, but they add additional uncertainty. They may suggest trends—this region will likely face more drought—without providing the precision that planning requires. Decision-making under uncertainty becomes necessary.

Embracing Uncertainty

Uncertainty isn't a bug in climate science; it's an honest acknowledgment of limits. Those who claim certainty about precisely what will happen are overstating what science allows. Those who claim uncertainty means we know nothing are misrepresenting how uncertainty works.

Decision-making under uncertainty is normal. We buy insurance without knowing whether our house will burn. We invest in education without knowing future job markets. We make medical decisions based on probabilities rather than certainties. Climate decisions require similar comfort with uncertainty—acting on the best available information while acknowledging we cannot know everything.

Risk management frameworks help navigate uncertainty. Rather than asking "what will happen?" we can ask "what might happen, and how bad would each outcome be?" Preparing for a range of scenarios, focusing on robust strategies that work across multiple possible futures, and maintaining flexibility to adapt as understanding improves—these approaches don't require perfect prediction.

What Science Actually Tells Us

Despite uncertainties, climate science tells us quite a lot. The planet is warming due to human activities—that's established beyond reasonable doubt. Continued emissions will cause continued warming—the more we emit, the more warming. Warming will intensify many extreme weather events. Sea levels will rise, with higher emissions producing more rise. Ecosystems will face unprecedented stress.

What remains uncertain is magnitude and timing. Will warming reach 2 degrees or 4 degrees? Will ice sheets take centuries to collapse or decades? Will societies adapt successfully or face catastrophic failures? These questions remain open—but they're open questions about degree, not about direction.

Questions for Consideration

How should society make major decisions when climate predictions involve significant uncertainty about magnitude and timing?

What's the appropriate balance between preparing for worst-case scenarios versus planning for more likely outcomes?

How can scientists better communicate uncertainty without undermining public understanding or sense of urgency?

Should precautionary approaches dominate when facing potential tipping points that might trigger irreversible changes?

How can local planners work with regional climate projections that involve substantial uncertainty about specific details?