When Inventories Lag Behind Forests, Updating Is Inevitable

National forest inventory (NFI) data are often collected over a 5-year or 10-year period, meaning some are already outdated by the time the complete results are available. This study assesses changes in growing stock volume (GSV, m³/ha) using hybrid estimation combined with Sentinel-2 metrics. It focuses on constructing a model for estimating gain in GSV using NFI plot data for two points in time (t_1 and t_2) with remotely sensed data for both t_1 and t_2 for a bitemporal approach, and remotely sensed data only for t_2 for a unitemporal approach. A machine-learning approach based on the random forests (RF) algorithm was used to predict GSV change. The original data for t_2 and additional data for a time (t_3) were then used to evaluate the accuracy of the change prediction at the plot level, after which the predicted changes were applied to update the plot-level GSV to predict plot-level GSV at t_3, which was then validated against the observed plot-level GSV at t_3. Changes were assessed with the Mean Average Annual Volume Change (MAAVC) method representing the average annual change in GSV over a given period. The results indicate that at plot level, the bitemporal model produced GSV change estimates with low accuracy R² = 0.26, RMSE = 4.06 m³/ha and MAE = 3.26 m³/ha, while the unitemporal model, achieved R² = 0.40, RMSE = 3.64 m³/ha, and MAE = 2.65 m³/ha when predicting GSV change. Using the estimated change to project into t_3 the MAAVC based on field data yielded an R² = 0.91, RMSE = 45.11 m³/ha, while the RS unitemporal yielded R² = 0.73, RMSE = 83.79 m³/ha, and the bitemporal yielded an R² = 0.72, RMSE = 83.61 m³/ha. Model performance stability were evaluated using a Monte Carlo simulation approach with a novel stopping criterion. A linear mixed effect model showed a significant difference between methods and post-hoc pairwise comparisons were then applied to determine which groups differ significantly. Conclusively, MAAVC and spatiotemporal RS methods provide a robust framework for projecting GSV using NFI and Sentinel-2 data.