Seven classic and recently proposed methods used for the estimation of total arterial compliance have been evaluated for their accuracy and applicability in different physiological conditions. The pressure and flow data are taken from a computer model that provides realistic simulations of the nonlinear-distributed systemic arterial tree. Besides the great flexibility in simulating different physiological or pathological cases, the major advantage of the computer model is that it allows precise knowledge of the pressure-dependent total arterial compliance, which is the variable of interest. The results show that the methods based on the two-element windkessel (WK) model are more accurate than those based on the three-element WK model. The classic exponential decay and the diastolic area method yield essentially similar results, and their compliance estimates are accurate within 10% except at high heart rates. The later part of diastole, i.e., from the time that the systolic pressure wave has reached all peripheral beds, gives the best results. The newly proposed two-area and pulse pressure methods, both based on the two-element WK model, are accurate (errors in general < 10%) and can be applied to other locations in the arterial tree where the decay time and area method cannot. Methods based on the three-element WK model consistently overestimate total arterial compliance (> or = 25%). The errors in the methods based on the three-element WK model arise from the fact that the input impedance in that model deviates significantly from the true input impedance at low frequencies. The strong dependence of compliance on pressure (elastic nonlinearity) does not invalidate the compliance estimates.(ABSTRACT TRUNCATED AT 250 WORDS)