The overlooked non-rainfall water input sibling of fog and dew: Daily water vapor adsorption on a !Nara hummock in the Namib Sand Sea

[Display omitted] •Melon-bearing !Nara thrive in the Namib Sand Sea.•Groundwater resources may be supplemented with fog.•The role of dew and water vapor adsorption is virtually unknown.•Water vapor adsorption in the soil on the !Nara hummock occurred every night.•In the soil, water vapor adsorption contributed more water than fog or dew. The Namib Sand Sea is a hyper-arid coastal fog desert, with limited vegetation scattered across the extensive sand dunes. Minimal and unpredictable rainfall, along with proximity to the ocean, make non-rainfall water inputs (NRWIs), i.e. fog, dew and water vapor adsorption, an important part of the ecohydrology. We studied water accumulation across an elevated hummock formed around a !Nara (Acanthosicyos horridus) thicket in the Namib dunes during February 2020. Continuous measurements of temperature and relative humidity (RH) gradients between the soil and the air were complemented with three nights of intensive NRWI observations using 14 micro-lysimeters. While the overall objective was to study total NRWI accumulation, micro-lysimeter placement was designed to address additional questions about how NRWI accumulation was affected by hummock micro-topography and the position of the !Nara thicket. Water vapor adsorption was the primary form of NRWI, occurring nightly and accounting for at least half of the total water accumulation, even on nights with fog. This was confirmed by the measured water vapor pressure gradients between the air and the soil, which indicated nightly downward vapor fluxes. Actual vapor pressures in the air were similar on nights with and without fog, indicating fog was mostly determined by temperature. Soil RH was used as a proxy for soil water content and was found to be strongly correlated with air RH. Average nighttime air RH was correlated with total nighttime water accumulation with an R2 of 0.69–0.78. Average nighttime air RH was correlated with total nighttime water accumulation with a coefficient of determination (R2) of 0.69–0.78. Even though measurements were limited to the soil, comparison between micro-lysimeters placed on the hummock inside and outside the !Nara thicket indicated consistent, but not significant, lower water contents inside the thicket, suggesting interception of water vapor by the spines and stems of the !Nara. We conclude that water vapor adsorption, though not always appreciated as such, is a large contributor to total NRWI. The fact that water vapor adsorption was th