this study was conducted in greenhouse conditions at the dser/cca/ufpb, located in areia, pb, brazil. the main objective was to evaluate two methods of determination of available soil water, a direct method (dm) and a laboratory method (lm), and the effect of three levels of available water, 10, 25 and 40% as indicative of irrigation management in cotton crop (gossypium hirsutum l. r. latifolium hutch). a completely randomized block was utilized in a 2 x 3 factorial scheme, with three replications. the variance analysis showed that the values of the available soil water using the direct method were significantly (p < 0.01) higher than the laboratory method. the management of irrigation in the cotton crop using 40% of available water by dm, showed better results of evapotranspiration, leaf area and above ground biomass. interaction of methods versus levels of available water for all variables studied was also found.