A caution sign for low water at the Great Salt Lake marina during a drought. (Photo by Rob Martin on Unsplash)
After a record fall, when nearly the entire United States experienced drought—unprecedented in US Drought Monitor history—water managers and planners nationwide are nervously hoping for a wet winter, cautious about prolonged dry spells that leave their communities vulnerable.
This caution is well-founded. Climate models paint an alarming picture of escalating drought risks in many parts of the world as temperatures rise. But as sophisticated as these numerical models are, they may actually be underestimating the true risks from prolonged dry spells in a warming world.
As a climate scientist who studies drought, I’ve spent years analyzing the projections of the global climate models that form the basis of reports from the Intergovernmental Panel on Climate Change. They are indisputably our only and best tools for understanding how human-caused warming will affect the planet’s water cycle, and they clearly show that many regions will likely face more frequent and intense droughts in the coming decades.
However, there’s a crucial aspect of drought risk that many models struggle to capture: natural variability from year-to-year and decade-to-decade. This internal variability in the climate system can amplify or counteract human-driven trends, potentially leading to more extreme outcomes than the models suggest.
The recent California megadrought offers a stark example. Studies suggest climate change made the drought about 40 percent more severe. But natural variability also played a major role in its unprecedented duration. The models didn’t predict a drought of this magnitude and length—it emerged from the combination of the climate change trend and an extended run of naturally dry years.
The models excel at capturing long-term trends driven by rising greenhouse gasses. But they have a harder time simulating the natural multi-year ebbs and flows of wet and dry periods that occurred even before humans started influencing the planet’s climate. In some cases, they may even underestimate the potential for back-to-back dry years that can accumulate into a megadrought.
These model limitations matter because the worst impacts often come not from a single dry year, but from droughts that persist for multiple years, depleting reservoirs and groundwater. By focusing primarily on average conditions, we risk missing the most damaging scenarios.
We see similar limitations when looking at other regions. Take the Colorado River Basin, which supplies water to 40 million people in the Southwest. Models project declining flows as the region warms. Yet, they too may underestimate the potential for extended low-flow periods even more severe than what we’ve seen in recent years. Natural variability could pile on top of the human-caused trend, potentially leading to more rapid depletion of reservoirs like Lake Mead.
Or consider a place like New York City. The watersheds that supply its drinking water have been relatively wet in recent decades. But tree ring records show intense multi-year droughts have occurred in the past. Climate change may increase the odds of such a drought recurring, potentially challenging the water supply for millions—a risk that may not be fully captured in model projections focused on average conditions.
To be clear, climate models are not wrong. They provide crucial insight into how human activities are reshaping the planet’s water cycle. The problem is that by necessity, they simplify countless complex processes operating across a continuum of spatial and temporal scales. Once they are run, their simulations are averaged together to identify common patterns of predicted change. The resulting picture of drought during climate change is analogous to taking thousands of photos of the Statue of Liberty and averaging them together: you’d capture the general shape, but miss a lot of important details.
How to address the hidden risks. First, climate modelers need to look beyond averages. Rather than focusing solely on the “most likely” outcomes from climate models, we should examine the full range of possibilities they generate, including low-probability but high-impact scenarios and especially their evolution through time. We also need to continue to study past climate variations recorded in tree rings and other natural archives to quantify the risks of multi-year and multi-decade droughts.
Second, stakeholders and climate scientists should develop “stress test” scenarios that combine projected climate trends with plausible natural variability. This will help reveal compound risks or tipping points that may be missed by looking at future trends or historical variations independently of each other.
Third, water managers, farmers, and others need to plan for a wider range of future possibilities that factor in both the changes in the means and the paths that the climate system takes to get to those average conditions. Think of two hikers in the desert: one who finishes off a quart of water at the start, trudges for an hour without drinking, then desperately downs three quarts all at once, versus another who sensibly hydrates every fifteen minutes. Even though both hikers consume the same amount over the course of one hour, the first hiker is considerably worse off than the second.
Just as knowing the hikers’ average water consumption (one gallon per hour) tells us nothing about the dangerous pattern of dehydration and overhydration experienced by the first hiker, studies that focus on average conditions can mask the potentially hazardous paths we might take to reach those averages. When considering drought risks from the combined effects of climate change, we need flexible strategies that can adapt to both long-term trends and short-term swings between wet and dry periods. In other words, we need to understand not just how much water we might have, but how its availability might vary over time.
The American West’s recent prolonged drought has offered a preview of the challenges to come as the climate warms. It should serve as a wake-up call that the future may hold more extreme dry spells than our current cohort of models may suggest. By preparing for a wider range of scenarios—including “worst case” droughts beyond what we’ve seen in the historical record—we can reduce our vulnerability to one of climate change’s most pernicious threats.
Finally, we must remember that the climate system’s natural variability hasn’t gone away. It’s still there, interacting in complex and unpredictable ways with human-caused changes. Sometimes it may offer a welcome reprieve from warming trends. But it can also amplify them in dangerous ways. When it comes to drought, we need to read between the lines to see the full picture of risk. Our water security depends on it.
The Bulletin elevates expert voices above the noise. But as an independent, nonprofit media organization, our operations depend on the support of readers like you. Help us continue to deliver quality journalism that holds leaders accountable. Your support of our work at any level is important. In return, we promise our coverage will be understandable, influential, vigilant, solution-oriented, and fair-minded. Together we can make a difference.