ABSTRACT
Sugar-cane is cropped in tropical and subtropical areas of the world, where precipitation behaviour is seasonally distributed. Hence, irrigation supply is a largely recognized necessity for obtaining higher yields, particularly in the first sugar-cane development stages (Reynoso, 1862; Thompson, 1977; Fonseca, 1984; Wiedenfeld, 2000). In many sugarcane-cropping places of the world, irrigation is expensive and water scarce. Therefore, farmers must find a profitable relationship between irrigation supply and sugar-cane yields. Presently available crop simulation models offer agricultural decision-makers an excellent tool for selecting the water management solution that would achieve the most reliable yields under variable soil or weather conditions (Burke et al., 1999; Singh et al., 1999). Nevertheless, modelling predictions are sometimes inaccurate (Landau et al., 1998; Richter, 1999), which
diminishes model confidence for decision-making and points out the need for model validations under wide ranks of modelling inputs.
