| dc.description.abstract | In the present study a commercial CFD code was used in order to investigate the dispersion of a pesticide inside an arch type tunnel greenhouse with continuous side vents. The greenhouse was cultivated with a tomato crop planted in double rows, which, during the period of measurements had a height of 1.5 m. In parallel, measurements were carried out in order to experimentally determine the decay rate of pesticide concentration. Air samples were continuously taken at seven points inside the greenhouse (six air pumps and one gas analyser). In the 3D numerical model calculations were done for several wind directions and wind speeds, using the experimental values as boundary conditions. The final solution for every case of wind direction and wind velocity was obtained, firstly by a converge solution under steady - state condition; and secondly by an unsteady one, where at the time which equals to zero the air volume in the experimental greenhouse was considered to contain a mixture consisting of air and the used pesticide. The simulation results that were in the same order of magnitude with the experimental values only during the first 5-6 minutes of unsteady solution and only qualitatively good agreement for the rest time, while the decay process was simulated. Even if different wind characteristics, as boundary conditions, were introduced in the CFD model every minute, the agreement remained in a qualitative level, showing that the concentration of the pesticide inside the greenhouse is a function not only of the greenhouse ventilation rate but also of the pesticide evaporation-volatilisation rate from the greenhouse crop, soil and cover surfaces. As the simulation model almost always underestimates the concentration levels inside the greenhouse, more investigation has to be carried out in order to obtain better agreement between measured and estimated values of pesticide concentration inside the greenhouse. | en |