Warm cumulus clouds form when humid air rises, expands and cools to saturation. Further cooling creates a small excess of water vapor pressure over the equilibrium value for liquid water, a quantity called water-vapor supersaturation. That excess determines which aerosol particles can activate into droplets, but it is difficult to measure directly in a turbulent cloud. Airborne estimates of relative humidity commonly inherit uncertainties of about 5–10% from temperature and water-vapor measurements, whereas cumulus supersaturation is commonly only about 0.1–1%. The new method instead diagnoses the relevant vapor-state component from the thermodynamic transition between saturated vapor and a mixture of vapor and liquid water.

The study treated that short transition, extending roughly 10 m near cloud base, as locally quasi-isobaric. The expected pressure change was about 1 hPa, or approximately 0.1% of ambient pressure, allowing the researchers to represent the process as adiabatic and reversible without applying an extended moist-adiabatic ascent. At each sampled cloud base, the measured liquid water content was converted into a mixing ratio and used with temperature, pressure and cloud heat capacity to estimate the latent heat released by condensation. This energy was then divided according to the first law of thermodynamics into a change in internal energy and work associated with expansion of saturated water vapor.

That division matters because vapor expansion work is not available for condensation. In a simplified calculation assigning all condensation energy to droplet growth, more liquid water would form than the fuller energy budget permits. The researchers therefore defined a total cloud supersaturation and separated it into a liquid-phase component associated with internal-energy change and a vapor-phase component associated with expansion work. The latter is the diagnosed water-vapor supersaturation used to relate droplet number concentration to activation conditions. An iterative calculation ensured that condensate amount, heat capacity, saturation mixing ratio and temperature remained mutually consistent.

Testing relied on airborne observations gathered during the 2014 ACRIDICON-CHUVA campaign using HALO, the High Altitude and Long Range Research Aircraft. The samples included growing convective cumuli over polluted and forested or deforested parts of the Amazon and over cleaner marine air near the Atlantic. Cloud properties were measured at 1 Hz with the Cloud Combination Probe–Cloud Droplet Probe, while meteorological temperature and pressure came from the BAHAMAS system. The analysis selected developing cloud passes with more than 20 droplets per cubic centimeter, excluded passes containing raindrops, and focused on the initial activation stage near cloud base.

Atmospheric conditions from polluted environments were associated with stronger buoyant forcing, greater cloud liquid water, higher diagnosed supersaturation and updraft speeds reaching about 4 m s−1. Cleaner, cloud-condensation-nuclei-limited conditions generally had weaker updrafts, no higher than about 2 m s−1, along with lower liquid water and supersaturation. The maximum diagnosed water-vapor supersaturation was 0.27% in the clean marine flight segment. It reached 0.34% and 0.44% in two flight legs over forested regions and 1.15% in a Central Amazon segment affected by urban and biomass-burning emissions. The fitted droplet spectra broadly reproduced the variation expected between cleaner and aerosol-rich environments.

The central validation compared two independent quantities: droplet number concentration at cloud base, paired with the newly derived supersaturation, and cloud-condensation-nuclei activation spectra measured below cloud base. Two aircraft counters were used, one cycling through supersaturations of approximately 0.2–0.55% and another measuring near 0.55%. Their supersaturation was calibrated from the activation of particles with known properties, generally ammonium sulfate, through Köhler theory. The comparison covered four Central Amazon flight segments. Most cloud-base and sub-cloud values agreed within combined uncertainty, although the largest departures occurred in heterogeneous air masses affected by pollution and in cloud passes more strongly influenced by entrainment and mixing with dry air.

Measurement and model limitations constrain the precision of the diagnosis. The cloud probe’s droplet number concentration and liquid water content each carried about 10% uncertainty in the main analysis, while the derived water-vapor supersaturation had a relative uncertainty of approximately 12%. A conservative scenario assuming 30% uncertainty in liquid water content produced supersaturation uncertainties of about 36%. The air sampled below and within cloud base was not always identical: the CCN measurements were roughly 700 m below cloud base on two flights and about 1,400 m below it on another. Consequently, the comparison was a closure between nearby air masses rather than a strict tracking of the same parcels.

The framework is also limited by its assumptions of thermodynamic reversibility, negligible entrainment and sufficiently accurate airborne measurements, none of which fully describe the turbulence and environmental heterogeneity of real cumulus bases. Its strongest result is therefore not that every cloud can be measured without error, but that an independent thermodynamic calculation can recover a physically plausible activation spectrum from aircraft observations. The method may be particularly useful where direct CCN-counter measurements are unavailable. The authors also note that simplified parcel calculations that omit vapor-expansion work can overassign available energy to condensation, potentially overestimating condensate production, latent heating, buoyancy and updraft speed.

Subject of Research: Thermodynamic diagnosis of water-vapor supersaturation at warm cumulus cloud bases

Article Title: A new method to derive water vapor supersaturation at cumulus cloud base from in situ measurements

Article References: Braga, R. C., Wendisch, M., Albrecht, R., & Harrison, D. (2026). A new method to derive water vapor supersaturation at cumulus cloud base from in situ measurements. Atmospheric Chemistry and Physics, 26(20), 14769-14791. https://doi.org/10.5194/acp-26-14769-2026

Image Credits: AI Generated

DOI: 10.5194/acp-26-14769-2026

Keywords: cumulus clouds, water-vapor supersaturation, aircraft measurements, Amazon Basin, cloud condensation nuclei, cloud microphysics, energy budget, cloud droplet activation, atmospheric thermodynamics, method, derive, water