Measurements and estimates of leaf wetness over agricultural grassland for dry deposition modeling of trace gases

Leaf wetness is an important and frequent phenomenon for the surface¿atmosphere exchange of some atmospheric trace gases that are well soluble in water, such as ammonia (NH3 and SO2), as well as for plant disease epidemiology. This study shows a comparison of two different techniques to measure leaf wetness; namely a painted flat-plate grid sensor and a system of four clip sensors. Although both techniques gave comparable results, the flat-plate grid sensor was favored, because of its stable signal and its ease of use. In this technique, the measurement height turned out to be of great importance for the leaf wetness duration (LWD); the flat-plate sensor at 1.0 m systematically underestimated LWD, while the flat-plate sensor at 0.1 m better represented the actual LWD. To obtain a representative signal, leaf wetness should be measured close to the surface. Using the available leaf wetness measurements, a comparison was made between three physical and four empirical leaf wetness models. Without any optimization, the physical model that calculates the potential condensation at the leaf surface gave the best results. However, after optimizing the humidity thresholds in the empirical leaf wetness models, the optimized model based on the difference between the actual and saturated specific humidity at the surface gave best results. For practical applications in atmospheric transport models, like for the calculation of dry deposition of well-soluble gases, the relative humidity (RH) threshold model might be easiest to implement. This study showed that different thresholds should be used for different vegetation types. In this study, an optimized RH threshold of 71% was derived for agricultural grassland