New Climate Metric Accounting for Humidity Finds Cooling Efficiency Declining Since 1971

For a century, energy planners and utility companies have relied on a simple mathematical yardstick to predict how much electricity the world needs to stay cool. Known as "cooling degree days" (CDD), the metric assumes that every degree of heat requires the exact same amount of energy to offset, regardless of the surrounding atmospheric conditions.[1][2]

But a new study published in the journal Nature Communications reveals that this foundational assumption is fundamentally flawed, masking a quiet but significant strain on global power grids. By ignoring the role of humidity, the traditional CDD metric fails to capture the actual thermodynamic workload required by air conditioners and refrigeration systems.[6]

To correct this blind spot, atmospheric scientists at the University of Hawai'i at Mānoa have developed a new, physics-grounded metric called "effective cooling degree days" (eCDD). This updated framework couples both temperature and moisture to the actual physics of refrigeration cycles, providing a much more accurate picture of climate-driven energy demand.[1][5]

The primary claim advanced by the researchers is that the efficiency of cooling systems is not static; it degrades as atmospheric conditions change. Applying the eCDD metric to 50 years of high-resolution weather data from 1971 to 2020, the research team was able to quantify this historical shift with a high degree of confidence.[3][4]

The evidence shows that cooling efficiency across North America has declined by roughly 2% to 4% per decade since 1971. This decline means that modern air conditioners must consume significantly more electricity today to achieve the same cooling effect they did fifty years ago, even before accounting for the increased frequency of extreme heatwaves.[1][2][5]

The mechanism behind this decline is rooted in basic thermodynamics. Real cooling systems do not simply lower the temperature of the air; they must also wring moisture out of it. As temperatures climb, compressors and heat exchangers become less efficient. When high humidity is added to the mix, the system is forced to expend massive amounts of energy condensing water vapor rather than just cooling the air.[3][4]

This dynamic creates stark regional disparities, leading to the study's second major finding: a climatic "tug-of-war" between temperature and humidity. The data highlights that the old temperature-only metric misjudges cooling demand differently depending on the region, overstating it in some places and severely understating it in others.[1][4]

In humid regions, such as the American South and the Mid-Atlantic, heat and moisture compound each other. The energy burden in these areas is significantly worse than a temperature-only analysis would suggest, as the atmosphere holds more water vapor that cooling systems must actively remove.[3][5]

Conversely, the researchers discovered a counterintuitive phenomenon in arid regions like the desert Southwest. While they initially expected efficiency to plummet due to extreme heat, they found that the atmosphere in these regions is actually becoming drier. Because drier air is genuinely easier to cool, this lack of moisture partially offsets the thermodynamic penalty of higher temperatures.[1][4]

In a few specific desert locations, cooling efficiency has actually held steady or even slightly improved over the last fifty years. The researchers note that this competing influence—temperature pushing efficiency down while decreasing humidity pulls it back up—is completely invisible to traditional forecasting models.[1][3]

To assess future grid vulnerability, the research team applied the eCDD framework to projections from 19 different climate models under a high-emissions scenario. The goal was to map the results onto the U.S. electricity grid and identify where infrastructure is most at risk over the coming decades.[1][2][5]

The projections indicate that the regions facing the steepest future increases in cooling demand are not necessarily the hottest, but rather historically temperate zones like the Northwest, the Great Lakes, and the Mid-Atlantic. In these areas, grid regions are projected to see cooling-related electricity demand more than double by mid-century.[1][2]

While the evidence regarding the physical requirements of cooling is robust, the researchers acknowledge transparent uncertainty regarding human adaptation and infrastructure response. The physics of the refrigeration cycle are well-understood, but the exact strain on future power grids will depend heavily on how quickly utility companies update their forecasting models.[2][4][6]

Furthermore, the widespread adoption of emerging, highly efficient dehumidification technologies—such as liquid desiccant cooling—could alter these demand trajectories, representing an edge case that current models cannot fully predict.[4][6]

Ultimately, getting cooling demand right is not merely an academic exercise. Financial markets trade futures based on these metrics to hedge against unusually hot summers, and grid operators use them to decide where to build new power plants. By transitioning to a physics-based metric that respects the complex relationship between heat and moisture, energy planners can ensure that the grid remains resilient as the climate continues to change.[1][3][4][5]