The growth of the world's population and the consequent food shortage, requires the expansion of agriculture into arid zones which constitute about 60 % of the earth's land area and are characterized by high levels of solar radiation and shortage of freshwater. The objective of the seawater greenhouse for arid lands was to develop and demonstrate a cost effective means of producing both crops and pure water in hot, arid coastal regions. The project exploited both the high solar radiation and prevailing wind to drive most of the energy exchange processes in the greenhouse. In addition to a crop grown inside the greenhouse, a shade tent provided shelter for nursery plants and an outdoor planting scheme was maintained with the supply of freshwater as well as protected by the structure of the greenhouse itself. The greenhouse is cooled with an evaporative cooling pad (Celdek) through which an air flow is promoted by the prevailing wind. The water vapour transpired by the plants combines with the cooled and humidified ventilation air stream to generate a high relative humidity in the exhaust air. A second evaporation pad (Celdek) is used to further humidify the exhaust air which passes through a condenser cooled by seawater to produce fresh water. The wind also promotes an air flow through a roof cavity in the greenhouse where more seawater is evaporated and the humid air passes through the condenser. A computer program was modified to describe the greenhouse, which at first was triangular in plan, and models were written to describe the evaporative cooling systems, the selectively absorbing roof, the humidification of the air in the roof and the seawater condenser. These models were incorporated into the existing greenhouse model. Analytical techniques were used to create the hourly values of solar radiation, air temperature, humidity, cloud cover and wind speed required by the simulation model. This simulation model was used to predict system performance and determine the sensitivity of the greenhouse environment and fresh water output to air and water flowrates and the climatic conditions. The sensitivity analysis showed that adding the second (rear pad) evaporative cooling pad increased the fresh water production compared to using only the front evaporative cooling pad. The use of cooling water at the wet bulb temperature gave a lower condenser output than using surface seawater. The results showed that an effective desalination system would consist of an evaporating cooling pad coupled directly to the seawater condenser. Data were recorded in the greenhouse in Tenerife in order to determine the heat and mass transfer coefficients of the second Celdek; these established the program of the system and provided data for validating the model. The experiments performed on a prototype Celdek heat exchanger in the laboratory were used to design the Celdek condenser. The validity of the seawater greenhouse simulation model was checked using experimental data recorded on the prototype seawater greenhouse in Tenerife, in December 1994 and June 1995. The model was revised as a result of the validation exercise. The model was then used to assess the performance of the different systems of the seawater greenhouse. Using meteorological data for the Tenerife island (Canaries Islands, Spain) and with a cooling water temperature of 10°C, the simulations showed the complete system could have an annual net water output of 350 kg m-2 greenhouse, and the conditions created inside the greenhouse would be suitable for protected crop production. The simulations showed that the major sources of vapour were both evaporative pads (90 %) followed by the roof cavity (9.5 %) and plant transpiration (0.5 %) in December, while in June the two pads contributed 80 %, the plant transpiration 13% and the roof cavity 7 % of the total water evaporated. The model was suitable for designing and establishing the performance of the Greenhouse and fresh water condensing system for other locations given appropriate meteorological data.