Aim of study: Cross-sectional area is one of the most important forest inventory variable since it is highly correlated with growth and yield at both tree and stand levels. In this research, we evaluated the bias, precision and accuracy of three measurements such as cross-sectional area: the girth, the arithmetic mean diameter, and the geometric mean diameter normally used to estimate the cross-sectional area in practical forestry. Area of study: Measurements were taken in a poplar plantation (Populus x euramericana (Dode) Guinier cv. Luisa Avanzo) located in Huesca, Spain. Material and Methods: A total of 5,408 cross-sectional areas from 48 poplar trees were measured with and image based software. To test the differences between real and estimated cross-sectional area based on the three measurements of study, a multilevel mixed-effect model was used. Main Results: All three measurements overestimated the cross-sectional area by (0.47%-2.37%) and were found to be biased. Estimations based on arithmetic or geometric mean diameter of the maximum and minimum axes were more accurate than those using tree girth. Research highlights: There was a strong correlation between estimation errors and departures from a circumference in the cross section i.e. estimation errors were larger in elliptical cross-sections than in those closer to a circumference. In order to avoid overestimation of growth and yield derived from cross-sectional area estimates, we recommend using the geometric mean diameter trying to measure the largest and the smallest diameters of the section, especially on trees that are clearly elliptical.