- 1Max Planck Institute for Biogeochemistry, Biogeochemical Integration, Jena, Germany (spaulus@bgc-jena.mpg.de)
- 2European Commission, Joint Research Centre (JRC), Ispra, Italy
- 3Department Hydrosystemmodellierung, Helmholtz Centre for Environmental Research—UFZ, Leipzig, Germany
- 4Institute of Geoscience, University of Jena, Jena, Germany
- 5Faculty of Environment and Natural Resources, University of Freiburg, Freiburg, Germany
- 6Fundacion Centro de Estudios Ambientales del Mediterráneo (CEAM), Paterna, Valencia, Spain
- 7Faculty of Geo-information Science and Earth Observation (ITC), University of Twente, Enschede, The Netherlands
In this contribution, we aim at assessing the detectability of atmospheric water vapor uptake by dry soils at a variety of spatial scales and methodologies, from the ecosystem scale via eddy covariance, through larger scales via earth system models, and gridded products.
Water vapor fluxes in the soil and at the soil-atmosphere interface are driven by vapor concentration gradients. Until today, it is mostly assumed that the soil pore air is roughly at 100% relative humidity (RH), resulting in vapor fluxes that are almost always towards the atmosphere. However, the vapor state in soil pore air is linked to the soil water (matric) potential. As the water potential becomes more negative, the equilibrium RH within the soil decreases substantially. Under these conditions, the soil behaves like a ‘thirsty material’: when the atmospheric vapor pressure exceeds that of the soil pores, vapor is adsorbed onto the solid soil particle surfaces, and the net vapor flux is directed towards the soil.
Using subdaily measurement data from a globally distributed network of eddy covariance stations, we show an emergent functional relationship between volumetric water content (VWC), RH, and latent heat (λE) flux direction at the ecosystem scale. Vapor fluxes towards the soil under dry conditions can be explained by the soil's sorptive forces inducing very low water potentials. Based on eddy covariance data, we find that soil vapor adsorption most frequently occurred in arid and semi-arid regions, particularly in ecosystems with sparse vegetation such as savannas and dry shrublands. On average, soil vapor adsorption occurs for 4 ± 1.1 hours per night, and may last up to 7 hours and on more than 150 nights per year in some drylands.
Furthermore, we demonstrate that the relationship between VWC, RH, and the vapor flux direction is evident in a wide range of in situ measurements in drylands, including lysimeter and humidity profile data. However, this relationship is absent in site-level runs of gridded observation-based data products and land surface models.
We demonstrate for the first time that the effect of adsorptive forces can be detected at the ecosystem scale, several meters above the ground. Our findings at the operating scale of flux towers can be used to evaluate and improve model representation of land-atmosphere exchange in dry conditions. Additionally, the results highlight the influence of sorptive forces on sub-daily soil-atmosphere interactions, particularly in sparsely vegetated drylands.
How to cite: Paulus, S. J., Migliavacca, M., Hildebrandt, A., Orth, R., Lee, S.-C., Carrara, A., Reichstein, M., Zeng, Y., and Nelson, J. A.: The underestimated thirst: detectability of atmospheric water vapor uptake in ecosystem measurements and global models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7376, https://doi.org/10.5194/egusphere-egu26-7376, 2026.