The assessment of the current annual increment of forest standing volume (CAI) is a fundamental tool to support forest management and planning. A suitable approach to such an end is to rely on growth and yield models. However, this kind of models are often not available for many countries and/or locations and/or species. Furthermore, they may become obsolete due to potential changes in the environmental and silvicultural conditions. Such shortcomings may be distinctively detrimental in the context of forest inventories. Several methods have been proposed to inventory CAI on one single occasion, i.e., when repeated measurements of standing volume are not available. A wellknown family of methods, still largely exploited in Alpine and Eastern European countries, derives from the estimation of the percentage current annual increment of forest standing volume by the current annual increments of stem diameter and tree height (Δh). In this study an experimental comparison of Δh assessment by three different approaches is presented with reference to a properly designed case study: (i) Δh is measured on felled trees; (ii) Δh is estimated by dynamic height curve (i.e., diameter-height-age model); (iii) Δh is estimated by conventional height curve (i.e., diameter-height model). Under the examined experimental conditions (a pure forest of silver fir on highly fertile soils in southern Italy, aged around 60 years), both simplified approaches (ii) and (iii) have proven to underestimate height increments, with a larger underestimation by the approach based on the conventional height curve. However, the consequent error in the estimation of percentage current annual increment of forest standing volume has proved to be quite limited (4% for the approach based on the dynamic height curve and around 9% for the approach based on the conventional height curve). Hence, such simplified approaches may be rather safely considered for estimating percentage current annual increment of forest standing volume when neither Δh is directly detectable on standing trees nor sample trees can be felled, nor an appropriate model to predict Δh is available. The Δh estimation on the conventional height curve should turn out to be even more suitable in the case of uneven-aged stands, where the position of the height curve remains stationary over time. ; L'articolo è disponibile sul sito dell'editore www.sisef.it