A key process in forest management planning is the estimation of tree volume and, more specifically, merchantable volume. The ability to predict cumulative stem volume to any upper stem diameter on standing trees or stands is essential for forest inventories and management of forest resources. In the 1980s, the Hellenic Public Power Corporation (HPPC) started the rehabilitation of lignite post-mining areas in Greece by planting mainly black locust (Robinia pseudoacacia, L.). Today, these plantations occupy an area of about 2570 ha, but the stem volume has not yet been estimated. Therefore, we aimed at estimating the over- and under-bark stem volume using taper function models for 30 destructively sampled trees. Of the nineteen calibrated fixed-effects models, Kozak’s (2004) performed best for both the over-bark and under-bark datasets, followed by Lee’s (2003) and Muhairwe’s (1999) equations. The addition of random effects on a single-stem basis for two coefficients of Kozak’s (2004) equation improved the model fit significantly. Two locally calibrated models were cross validated with calibration coefficients from Poland and the United States. Integrated taper functions were found to slightly outperform three tested volume equations for volume predictions of single stems over and under bark. A validation data set for the taper model verified that those models can be used to predict diameter at a specific point along the stem, merchantable volume, and total stem volume of black locust plantations. Finally, it can be said that taper models permit flexible volume estimation for the population average as well for specific trees.