On dew and micrometeorology in an arid coastal ecosystem

This study investigated intriguing aspects of dew within a sandy arid ecosystem situated in the NW Negev desert, Israel. The goal was to quantify dew formation and evaporation processes through sensor design, field measurements and modelling. To do this, two new sensors were developed. The first sensor is an automated microlysimeter for measurements of water content changes within the upper soil layer. This sensor uses a sensitive inexpensive Load Cell to provide daily records of dew input and evaporation. Dew amounts reached almost 0.4 mm day-1 but were found to depend strongly on the soil type. The second sensor developed was a remote optical surface wetness sensor that was designed and tested in the laboratory and in the Negev desert. The novelty of the sensor is that it measures surface wetness directly at the interface of the atmosphere, it is non-destructive and allows for repetitive measurements on the same sample, and it can operate outdoors without being affected by direct sunlight. The sensor was found to be capable of measuring leaf water content and near soil surface water content. Data obtained for an Anabasis articulata bush showed that leaves change their leaf water content during the evening. In general, errors in flux measurements can lead to difficulties in closing the energy balance at the surface, especially if turbulent flux or soil heat flux measurements are underestimated. Soil heat flux was assessed during the field experiments using a new approach. By placing the soil heat flux sensor near the surface, depth corrections are no longer needed. It is shown that the large soil heat fluxes in hot desert regions are very important in energy balance studies. The soil heat flux measurements led to a very good closure of the surface energy balance and highlighted the importance for careful soil heat flux measurements.An analysis of dew formation conditions revealed that the actual dew is greater than the calculated potential dew. The dew formation process seems to be enhanced by the extreme dryness of the soil. The dry soil pores reduce vapour pressure and increase the downward vapour flux. This is demonstrated by using a Penman model and incorporating the vapour pressure reduction of the soil surface. The dew amount that is available within a soil is limited by the soil physical properties. The term FAD “Free Available Dew” is introduced to define the amount of water available for biological activity. The FAD would be higher for a biological crust with cyanobacteria than for vascular plant roots. The FAD for cyanobacteria is around 20% of total dew. To improve our understanding of the distribution and flow of dew water within a sandy soil covered with a biological crust, a multilayer soil model was constructed. The model was tested with the microlysimeter data and gravimetric soil moisture profile measurements. Diurnal soil moisture variation was found to be confined to the top 15 mm. The FAD is available in the upper half of a 4 mm thick biological crust only and limited to < 25% of the day. The possibility of harvesting dew from an artificial flat surface within the interdune was investigated by constructing a dew collector and analyzing the measurements. This resulted in a detailed dew model to simulate dew yield in general. Dew yields of 0.30 mm day-1 are possible in the NW Negev desert at a location 40 km from the coast at the end of the dry season. The artificial flat surface tested here provided 0.10 mm day-1. Modelled yields were found to be larger near the coastline. Open path gas analyzers and ultrasonic anemometers have been used to measure water vapour fluxes. Dew can affect the performance of open path gas analyzers and thus affect data quality in an arid environment. A field experiment showed that additional sensor heating can avoid dew problems and yet greatly improve the quality of flux measurements. A detailed energy balance model of the gas analyzer window housings is presented so that existing datasets can be filtered for dew contamination events on the windows. Additional heating does influence the measured concentration but the effect is very minor.