"Annual Rainfall" Cannot Predict the Future — Earth's Rainfall Patterns Are Breaking

"Rain Increases, Yet the Land Dries"—How the "Pattern of Rainfall" Determines the Future

In discussions about global warming, we often hear the narrative that "as temperatures rise, the atmosphere holds more water vapor, leading to stronger rainfall." Indeed, the intensification of heavy rain has been observed in many regions and frequently makes the news. However, what truly affects our lives is not just the annual total rainfall.

The real challenge lies in the **"rhythm" of the rain**. The rainy season shortens. The intervals between rains lengthen. And then, after these dry spells, it rains "all at once." This "intermittency" could be the mechanism that simultaneously increases both floods and droughts.

The study introduced by Phys.org, conducted by a research team from the University of Utah and others, examines how rain patterns change on a warming Earth, using the hyperthermal period from about 66 million to 48 million years ago (Paleogene) as a clue. The conclusion is not intuitive.

Contrary to the "familiar notion" that "wet regions become wetter and dry regions become drier" with warming, the research team suggests that "even mid-latitudes could become drier." The key lies not in the total amount of rainfall but in the **temporal distribution of precipitation (when and how frequently it falls)**.

Reading 66 Million Years Ago as a "Future Laboratory"

The Paleogene period, which the study focuses on, includes times when atmospheric CO₂ levels reached 2 to 4 times the current levels, serving as a near "field test" of an extremely warm Earth.
Among the most famous events in the Paleogene is the PETM (Paleocene-Eocene Thermal Maximum) around 56 million years ago. Known as a "hyperthermal" event, the PETM saw a rapid increase in greenhouse gases and a sharp rise in Earth's temperature. NOAA's explanation suggests that during the PETM, global average temperatures may have risen by as much as 5 to 8°C.

Another study estimates that global average temperatures during the PETM were significantly higher than pre-industrial levels (with some range of estimates), making it an important comparative material for considering "how the water cycle behaves when the Earth is much hotter than now."

What Was Investigated: Reconstructing "Rainfall Patterns" from Fossils, Paleosols, and River Sediments

The problem is that there are no rain gauges from tens of millions of years ago. Therefore, the research mobilizes "proxies"—indirect evidence preserved in geological records. The Phys.org article cites examples such as plant fossils (leaf shapes, etc.), paleosol chemistry, and river geomorphology and sediments. Whether rain falls "gently every day" or "increases rapidly after a long dry spell" changes the river's ability to transport and erode rocks, as well as the shape of the riverbed. Thus, geomorphology and sediments can serve as mirrors reflecting the "intensity" and "intermittency" of precipitation.

Additionally, the abstract of the Nature Geoscience paper states that they constructed a multi-proxy approach integrating these sedimentary proxies to constrain the "seasonal to interannual variability (intermittency)" and "rainfall rate (intensity)" of precipitation.

Emerging Conclusions: Poles Get Wetter, Mid-Latitude Interiors Face a Combination of "Dryness + Heavy Rain"

The broad outline presented by the study is as follows.

• Polar Regions Could Become Humid to Monsoonal

• Meanwhile, mid- to low-latitude continental interiors show a trend towards dryness, but with "occasional heavy rains" (heavy rain following prolonged dryness)

What is important here is that dryness cannot be explained solely by a "decrease in total rainfall." The abstract states that aridification is driven by "changes in distribution," such as shortening of the rainy season and lengthening of intervals between rains, independent of average annual precipitation.

The Phys.org article similarly notes that dryness can arise not just because "rain decreases," but due to short rainy seasons and long dry spells.

Moreover, the deviation from the commonly discussed notion that "wet regions become wetter, dry regions become drier (wet-gets-wetter / dry-gets-drier)" is highlighted. The abstract suggests that the wetting of polar regions and the drying of mid-latitudes indicate a departure from that simple response.

Why the Deviation? Nonlinearity and "Thresholds" Invisible in "Averages"

Even more intriguing is the possibility that these hydrological climate shifts began about 3 million years before the PETM and continued for about 7 million years after. This suggests that it is not just the "aftermath of a single event," but that when the Earth system exceeds certain conditions, the behavior of rain can change **nonlinearly (not in direct proportion)**—a possible interpretation.

The Phys.org article also discusses how precipitation behavior can change unexpectedly when the climate exceeds certain "thresholds."

Practical Impact: The Design Philosophy Centered on "Annual Rainfall" Becomes Risky

To translate the message of this study for the general public:
Even in "years when it rains," water shortages can occur.

This is because dams, groundwater, farmland, forests, and urban drainage infrastructure require not just "how much it rains," but also "when it rains" and "when it will rain next."

• Long Dry Spells → Soil Hardens and Becomes Less Permeable

• Subsequent Heavy Rain → Rapid Runoff Increases Floods, Landslides, and Turbid Water

• Short Rainy Seasons → The "Storage Period" for Reservoirs Decreases, Making Management Difficult Even with the Same Total Rainfall

The Phys.org article also clearly states that in the future, the timing and reliability of rain will become more important than the annual average, emphasizing implications for floods, droughts, and water management.

A Striking Point: "Models Might Underestimate Rainfall Irregularity?"

The study also touches on the possibility that current climate models might underestimate "rainfall irregularity" through comparisons with paleoclimates. Paleoclimates have different boundary conditions (such as continental configurations and ice sheets) from today. However, this makes them an ideal teaching material to test "whether models can withstand unknown conditions."
This suggestion strongly highlights the risk of overlooking future predictions by merely viewing them as "maps of averages."

Reading from a Japanese Perspective: The Rainy Season, Typhoons, and Droughts Become "The Same Story"

Japan already has strong seasonality, with a rainy season, typhoons, and winter patterns. Therefore, if "intermittency" progresses, the impact will be twofold.

• Even if the total rainfall during the rainy season remains the same, the number of rainy days decreases and "concentrated rainfall" increases

• Typhoon rains are more likely to exceed flood control capacity On the other hand, in years without typhoons, droughts become more severe

• In agriculture, the timing of sowing and transplanting, the flexibility of water use, and the maintenance of soil moisture become more challenging

• In forests and ecosystems, both drought stress and disturbances from heavy rain simultaneously affect (uprooting, soil erosion, etc.)

In short, instead of considering "heavy rain measures" and "drought measures" separately, it becomes necessary to design them simultaneously as "different faces of the same climate change."

Reactions on Social Media (Note: Example Posts: Reconstructed Typical Discussions, Not Quotes from Actual Posts)

Note: This section reconstructs potential discussion points from the article content into phrases commonly seen on social media (not indicative of actual posts by specific individuals). The original paper is published in Nature Geoscience, with observable access and Altmetric scores, indicating some impact outside the research community.

• Disaster Prevention and Civil Engineering Cluster
  "Even if the annual rainfall is the same, an increase in floods is the most troublesome. We need to review design rainfall and operational rules, including 'seasonal shortening,' or we're doomed."

• Agriculture Cluster
  "It's not about whether it rains, but 'when the next rain is.' If dry intervals lengthen, crops won't survive with the same amount of rain."

• Meteorology and Research Cluster
  "The emergence of discussions centered on intermittency, as opposed to the simple wet-gets-wetter diagram, is significant. You misinterpret if you only look at maps of average values."

• Skeptical and Opposing Views
  "Isn't it unreasonable to apply stories from tens of millions of years ago to the present? (→ The counterargument is that while boundary conditions differ, they serve as 'material for model verification.')"

• General Public Sentiment
  "The feeling that 'recently, when it rains, it pours, and when it doesn't, it doesn't rain at all' is exactly right. It's scary that average rainfall is less reliable than average temperature."

• Easily Shareable Phrases
  "Future water shortages won't be because 'rain decreases,' but because 'the intervals between rains lengthen.'"

(Note: In posts or threads introducing this research, numbers like "18°C" from the article tend to be taken out of context and spread. The PETM involves separate concepts of "rise (how many °C it increased)" and "how much higher the average temperature was compared to pre-industrial times," so it's safer to read with context.)

Conclusion: The Question Shifts from "How Much It Rains" to "How It Rains"

What this study confronts us with is the possibility that the center of future water risk is shifting from gradual changes in average values to the intermittency and intensity of precipitation.
Instead of discussing reassurance or pessimism based solely on increases or decreases in annual rainfall, attention should be paid to the length of the rainy season, intervals between rains, and the concentration of heavy rains—the very "editing of rain."

To prepare for the simultaneous occurrence of floods and droughts, the worst combination.