Data on forests (e.g., height, diameter at breast height and volume) are increasingly being collected using remote sensing methods, leading to improved forest inventories. At present, a popular method for data collection is the use of unmanned aerial vehicles (UAVs) equipped with LiDAR sensors, which allow for a more detailed assessment of structural parameters in both space and time, facilitating the practical application of more complex forest management systems. Therefore, the main objective of this study was to measure the structural elements of stands (e.g., volume, basal area, tree count, height, diameter at breast height, crown width, and area) from LiDAR images and determine the accuracy of the obtained results. The research was conducted in a 5500 ha area containing some of the most valuable forests in Croatia, namely, lowland oak forests. The results of this study show that there are no statistically significant differences between the diameters and heights measured in the field and those from LiDAR images and, consequently, no difference in the calculated volume. This study also concluded that the use of unmanned aerial vehicles with various sensors can significantly reduce fieldwork while achieving the same accuracy in the results, thus leading to substantial savings in both time and money.

Data on forests (height, diameter at breast height, volume, etc.) are increasingly being collected using remote sensing methods, which improve forest inventories. The latest popular method of data collection is through unmanned aerial vehicles (UAVs) equipped with LiDAR sensors, which allow for a more detailed assessment of structural parameters in space and time, facilitating the practical application of more complex forest management systems. Therefore, the main objective of this study was to measure the structural elements of the stand (volume, basal area, tree count, height, diameter at breast height, crown width and area, etc.) from LiDAR images and determine the accuracy of the results obtained. The research was conducted in the area of the most valuable forests in Croatia – the lowland oak forests, covering an area of 5500 ha. The results of the study showed that there is no statistically significant difference between the diameters and heights measured in the field and those from LiDAR images, and consequently, no difference in the calculated volume. The study also concluded that the use of unmanned aerial vehicles with various sensors can significantly reduce fieldwork while achieving the same accuracy in the results, thus leading to substantial savings in both time and money.

Negativne posljedice propadanja šuma nastoje se umanjiti provođenjem gospodarskih mjera i praćenjem zdravstvenog stanja stabala, odnosno procjenom oštećenosti krošanja. Utvrđivanje zdravstvenog stanja šuma provodi se, osim terestričkim načinom, i met...

One of the priorities in sustainable forest management is monitoring the health status of trees and stands. From the aspect of remote sensing (RS), the best way of doing this is by interpreting color infrared (CIR) aerial photographs; however, this raises the issue of sample size. For this reason, to apply this method in practice, it is indispensable to determine an appropriate sample size to ensure sufficient reliability of the health status assessment of trees in CIR aerial photographs. This research was conducted in lowland forests of pedunculate oak in Croatia. To determine damage in the photographs of the main tree species, a systematic sample with varying dot grid densities—100 × 100 m, 200 × 200 m, 300 × 300 m, 500 × 500 m, 1000 × 1000 m—was used with combinations of different numbers of interpreted trees per sample. Damage indicators were also calculated based on tree distributions obtained by interpreting four trees, two trees and one tree in different sample sizes. The results of the testing showed that there were no statistically significant differences between different sample densities and numbers of interpreted trees in relation to mean damage assessment. Regardless of the fact that there were no statistically significant differences during damage assessment, it was found that by lowering sample densities, starting with 200 × 200 m, the number of trees and the number of sample points per particular sub-compartment significantly decreased, and so did the desired accuracy. Consequently, the participation (distribution) of particular species and damage degrees in the sample were lost, which significantly affected the overall tree health assessment. In contrast, grid densities of 100 × 100 m with one interpreted tree at the raster point proved to be the optimal sample size. This confirms the fact found in earlier research, that is, that the selected sample should have several spatially well-distributed points with a smaller number of trees in the point, and samples with larger numbers of trees in a smaller number of points should be avoided.

Hybrid radiative transfer model–machine learning (RTM–ML) approaches can incorporate physical constraints into vegetation-trait retrieval, but their advantage over empirical learning remains uncertain for very-high-resolution, limited-band observations of structurally heterogeneous forest canopies. We compared six empirical machine-learning regression algorithms with a hybrid PROSAIL–Gaussian Process Regression (GPR) approach for retrieving derived canopy chlorophyll content (CCC) from four-band UAV multispectral imagery of pedunculate oak (Quercus robur L.) stands under contrasting oak lace bug (Corythucha arcuata) pressure. CCC was derived from field-derived leaf chlorophyll content and UAV-estimated leaf area index. Data were collected across three sampling periods from 2021 to 2023, and models were evaluated using tree-grouped three-fold cross-validation to maintain independence between training and held-out trees. Empirical performance varied among algorithms, sites, and sampling periods. At the infested site, GPR provided the strongest performance in May (R² = 0.61, RMSE = 27.04 µg cm⁻², NRMSE = 10.87%), whereas Random Forest performed best in September. At the reference site, Random Forest, least-squares linear regression, and kernel ridge regression provided the strongest results depending on sampling period and performance metric, with R² reaching 0.60. According to the primary cross-validated predictive metrics (R², RMSE, and NRMSE), the tested PROSAIL–GPR approach generally showed lower predictive performance than the empirical models under the same tree-grouped evaluation framework. Empirical GPR maps resolved substantial within-stand spatial variation in derived CCC. Overall, incorporating PROSAIL-simulated information did not improve predictive performance under the tested four-band, crown-scale configuration. Rather than identifying a single cause, the results indicate that the benefit of physically constrained retrieval depends on compatibility among model representation and parameterization, canopy structure, spectral information, and available training data.