The motivation for the study stems from a persistent blind spot in climate scenario databases. Future projections produced by coupled climate models under the Shared Socioeconomic Pathways are readily available for some variables, such as air temperature and precipitation, but are not directly published for several key atmospheric quantities. Dew point temperature, wet-bulb temperature, humidity ratio, water vapor partial pressure, and total column precipitable water all fall into this category, even though each plays a decisive role in human heat stress, agriculture, water resources, and the energy balance of the climate system. Without forward projections of these variables, comprehensive assessments of future climate risk remain incomplete, and adaptation planners are left working with half the picture.

To overcome this limitation, the researcher developed an integrated forecasting framework that combines three complementary ingredients: externally available climate scenario data, a statistical time-series forecasting model known as ARIMAX, and physically based thermodynamic equations. The logic is elegant in its simplicity. Where future scenario outputs exist, they are used directly. Where they do not, the ARIMAX model—a variant of the classical autoregressive integrated moving average approach that incorporates exogenous predictor variables—generates statistically robust projections using air temperature as the driving input. Finally, established psychrometric relationships, of the kind used for decades in engineering handbooks, translate the projected temperatures and humidity measures into the remaining indicators, ensuring that every number in the resulting dataset is physically consistent with every other.

Specifically, future projections of dew point temperature and total column precipitable water were obtained using ARIMAX with air temperature as the exogenous variable, while the remaining indicators—wet-bulb temperature, dew point depression, relative humidity, humidity ratio, water vapor partial pressure, air density, and total precipitation—were derived through established thermodynamic relationships. This hybrid design means the framework does not attempt to conjure atmospheric physics from statistics alone. Instead, it anchors the statistical forecast to a well-constrained predictor and lets the known equations of moist air physics do the rest, a strategy that dramatically reduces the risk of producing projections that violate conservation laws or known vapor-pressure behavior.

The historical reconstruction, spanning 1940 to 2024, establishes the baseline against which the future is measured. By 2024, the study records global mean dry-bulb temperature at 15.1 degrees Celsius, dew point temperature at 10.0 degrees Celsius, and wet-bulb temperature at 12.2 degrees Celsius. Dew point depression—the gap between air temperature and dew point, a measure of how much additional moisture the air can hold before saturation—stood at 5.1 degrees Celsius. Total column precipitable water, the depth of liquid water that would result if all vapor in a vertical column of atmosphere were condensed, reached 25.6 kilograms per square meter. Water vapor partial pressure registered 12.3 hectopascals, the humidity ratio 7.8 grams of water vapor per kilogram of dry air, relative humidity 71.4 percent, air density 1.191 kilograms per cubic meter, and total annual precipitation 1.073 meters.

Projected forward under SSP5-8.5, the high-emissions pathway in which fossil fuel use continues largely unabated through the century, these indicators shift substantially by 2100. Dry-bulb temperature climbs to 18.9 degrees Celsius, a rise of 3.8 degrees over the 2024 value. Dew point temperature reaches 13.4 degrees Celsius and wet-bulb temperature 15.6 degrees Celsius, while dew point depression widens slightly to 5.5 degrees Celsius. Total column precipitable water rises to 30.0 kilograms per square meter, water vapor partial pressure increases to 15.5 hectopascals, and the humidity ratio grows to 9.9 grams per kilogram. Relative humidity edges down marginally to 70.5 percent, air density thins to 1.174 kilograms per cubic meter, and total precipitation increases to 1.128 meters per year.

Several of these projected changes carry implications that extend well beyond academic climatology. The rise in wet-bulb temperature is perhaps the most consequential for human societies, because this quantity effectively measures the body’s ability to cool itself through sweating; sustained wet-bulb values approaching the mid-thirties Celsius are incompatible with human survival outdoors, and even far lower values impose severe health and labor-productivity penalties. A global mean wet-bulb temperature of 15.6 degrees Celsius by 2100 implies that regional extremes in already hot, humid parts of the world will push far closer to dangerous thresholds than they do today. Meanwhile, the roughly 17 percent increase in precipitable water is a direct manifestation of the Clausius–Clapeyron relationship, which dictates that the atmosphere holds about seven percent more water vapor for every degree of warming—a feedback that intensifies heavy rainfall events even as some regions grow drier overall.

The projected decline in air density, from 1.191 to 1.174 kilograms per cubic meter, is a subtler but technically significant outcome. Warmer, more humid air is less dense, and this thinning has measurable consequences for wind turbine power output, aircraft lift and engine performance, and the combustion efficiency of engines and turbines. As the atmosphere warms, engineers and energy planners will increasingly need to account for these density effects in siting and design decisions. The slight decrease in relative humidity alongside rising absolute moisture content reflects the fact that warmer air’s capacity to hold vapor grows faster than actual vapor supply in many regions, a distinction that matters for evaporation rates, wildfire risk, and agricultural water demand.

The study’s methodological contribution may prove as influential as its numerical results. By demonstrating that ARIMAX modeling anchored to temperature projections can generate physically plausible, long-horizon estimates of variables missing from standard scenario archives, the framework provides a practical, reproducible template that other researchers can apply at regional scales or with different scenario pathways. The author acknowledges the Climate Change Institute at the University of Maine for supplying the reanalysis data underpinning the historical record. As climate impact assessment, adaptation planning, and long-term environmental decision-making increasingly demand multi-variable, internally consistent futures, tools of this kind bridge the gap between what climate models currently publish and what practitioners actually need—turning the full thermodynamic state of the future atmosphere from an afterthought into a forecastable, actionable dataset.

Subject of Research: Historical analysis and future projections of ten fundamental global climate indicators using statistical forecasting and thermodynamic modeling

Article Title: Historical analysis and future projections of fundamental global climate indicators

Article References: Historical analysis and future projections of fundamental global climate indicators. (n.d.). https://doi.org/10.1007/s00704-026-06543-x

Image Credits: AI Generated

DOI: 10.1007/s00704-026-06543-x

Keywords: climate change, climate indicators, ARIMAX forecasting, dew point temperature, wet-bulb temperature, relative humidity, precipitable water, air density, SSP5-8.5 scenario, thermodynamic modeling, climate projections, global warming