The objective of this study was to estimate the maximum evapotranspiration (ETm) of tomato inside greenhouse as a function of evaporation measured with "Piche" evaporimeter exposed to solar radiation (Epi) with and without easily measured meteorological and phenological variables during the 2005 spring growing season. Daily ETm (measured with drainage lysimeters), Epi, air temperature, air vapor pressure deficit (D), leaf area index (LAI), plant height (PH), and number of leaves (LN) were measured. Models were fitted to estimate ETm using data collected on even days of the experimental period. In the multiple regression analysis only models with parameters significant at 5% by t test and coefficient of determination (R²) greater than 0.85 were selected. Models were evaluated with independent data collected on odd days of the period. Statistics used for model evaluation were regression analyses between observed and estimated data, root mean square error (RMSE), and accuracy (r), agreement (d) and confidence (c) indices. Results showed that it is possible to estimate ETm of tomato crop with acceptable accuracy (R² between 0.87 e 0.92) to perform irrigation inside plastic greenhouse using models that take into account Epi and at least one plant growth parameter; models with PH should be preferred. Including D measured at 9 h in the models with Epi and LN or PH or LAI improved ETm estimates.
irrigation; Piche evaporimeter; Lycopersicon esculentum Mill; greenhouse crops
