Evaluation of the VegSyst model with muskmelon to simulate crop growth, nitrogen uptake and evapotranspiration

Like many intensive vegetable production systems, the greenhouse-based system on the south-eastern (SE) Mediterranean coast of Spain is associated with considerable NO3- contamination of groundwater. Drip irrigation and sophisticated fertigation systems provide the technical capacity for precise nutrient and irrigation management of soil-grown crops which would reduce NO3- leaching loss. The VegSyst crop simulation model was developed to simulate daily crop biomass production, N uptake and crop evapotranspiration (ETc). VegSyst is driven by thermal time and consequently is adaptable to different planting dates, different greenhouse cooling practices and differences in greenhouse design. It will be subsequently incorporated into a practical on-farm decision support system to enable growers to more effectively use the advanced technical capacity of this horticultural system for optimal N and irrigation management. VegSyst was calibrated and validated for muskmelon grown in Mediterranean plastic greenhouse in SE Spain using data of four melon crops, two grown in 2005 and two in 2006 using two management strategies of water and N management in each year. VegSyst very accurately simulated crop biomass production and accurately simulated crop N uptake over time. Model performance in simulating dry matter production (DMP) over time was better using a double radiation use efficiency (RUE) approach (5.0 and 3.2 g MJ(-1) PAR for vegetative and reproductive growth phases) compared to a single RUE approach (4.3 g MJ(-1) PAR). The simulation of ETc over time, was very accurate in the two 2006 muskmelon crops and somewhat less so in the two 2005 crops. The error in the simulated final values, expressed as a percentage of final measured values was -1 to 6% for DMP, 2-11% for crop N uptake, and -11 to 6% for ETc. VegSyst provided effective simulation of DMP, N uptake and ETc for crops with different planting dates. This model can be readily adapted to other crops. (C) 2011 Elsevier B.V. All rights reserved.