In the last decade, small unmanned aerial vehicles (UAVs/drones) have become increasingly popular in the airborne observation of large areas for many purposes, such as the monitoring of agricultural areas, the tracking of wild animals in their natural habitats, and the counting of livestock. Coupled with deep learning, they allow for automatic image processing and recognition. The aim of this work was to detect and count the deer population in northwestern Serbia from such images using deep neural networks, a tedious process that otherwise requires a lot of time and effort. In this paper, we present and compare the performance of several state-of-the-art network architectures, trained on a manually annotated set of images, and use it to predict the presence of objects in the rest of the dataset. We implemented three versions of the You Only Look Once (YOLO) architecture and a Single Shot Multibox Detector (SSD) to detect deer in a dense forest environment and measured their performance based on mean average precision (mAP), precision, recall, and F1 score. Moreover, we also evaluated the models based on their real-time performance. The results showed that the selected models were able to detect deer with a mean average precision of up to 70.45% and a confidence score of up to a 99%. The highest precision was achieved by the fourth version of YOLO with 86%, as well as the highest recall value of 75%. Its compressed version achieved slightly lower results, with 83% mAP in its best case, but it demonstrated four times better real-time performance. The counting function was applied on the best-performing models, providing us with the exact distribution of deer over all images. Yolov4 obtained an error of 8.3% in counting, while Yolov4-tiny mistook 12 deer, which accounted for an error of 7.1%.

The selection of tree species can affect the success of afforestation in the rehabilitation of degraded forest sites and forest restoration. In general, black locust (Robinia pseudoacacia L.) and black pine (Pinus nigra Arnold.) represent the most commonly used species in the afforestation of soils that have been degraded by erosion. As far as the extent of the ameliorative effects of black locust and black pine are concerned, it was found that they may play an important role in the selection of species for the afforestation of extremely degraded sites. This study is aimed at determining the potential of black locust and black pine to affect several soil properties, erosion control and C stock, thus creating favourable site conditions for the restoration of previous forest vegetation. This research was conducted in the Grdelica Gorge in south east Serbia, where eight sample plots with an average size of 0.47 ha were established 60 years ago on terrain afforested with black locust and black pine. In each sample plot, we measured the diameter at breast height of all black locust and black pine trees, and the height of 10 black locust and 10 black pine trees in each diameter class. In addition, samples of mineral soil (from depths of 0-5, 5-10 and 10-20 cm) were taken at 4 randomly selected soil profiles in each sample plot, and 8 samples of litter (30 × 30 cm) were also collected. Additionally, laboratory analyses of the physical and chemical properties of the soil and litter were performed in 2 replicates. The obtained results showed that: (1) at the 0-5 cm depth, there was no statistically significant difference in the reaction of the soil solution, although a significant difference in the reaction of the soil solution between the soils under the two species was observed at soil depths greater than 5 cm; (2) there was a significantly higher N content under black locust in the 0-5 cm soil layer; (3) the reduction of soil loss under black locust is statistically significant in all observation periods; (4) black pine is more efficient in C storage. Our results demonstrate that black locust has the potential to improve soil properties and reduce soil loss caused by erosion, while its favourable impact does not decrease over time, making it more suitable for afforestation on degraded land in the examined area.

The growth characteristics of silver fir are of high importance for selection forest management, and for the current aims laid out in Serbia?s forest management focused on increasing the share of silver firs in Serbia?s growing stock. With the objective of increasing the understanding of the growth characteristics of silver fir, the growth of two silver fir trees felled during forest site production research on Mt. Goc, located in Central Serbia, have been analyzed. Both trees showed significant differences in their growth dynamics over long periods as results of micro-site and micro-stand effects (primarily ambient light regime). The common growth characteristic of the two trees, a 450-year-old tree as the main study object (labeled Tree A) and a 270-year-old Tree B is a long stagnation stage. For Tree A the latent phase, with small interruptions, lasted 410 years; one phase lasted 330 years in continuity, which is the longest period of silver fir stagnation recorded in Europe. Tree B showed a long-lasting stagnation stage that lasted 170 years. The long stagnation stage of Tree A, characterized by an average diameter increment of 1.4 mm/year (average growth ring width of 0.7 mm) and an average height increment of 0.08 m/year, shows the extraordinary silver fir capacity for physiological survival in complete shade. This study adds to the existing knowledge of the shade tolerance of the silver fir. Therefore, the silver fir belongs to the group of extremely shade-tolerant tree species. This characteristic makes silver fir an irreplaceable tree species in the selection forest structure. It offers a wide range of silvicultural flexibility in the management of these forests, and is applicable to silver fir selection Serbia?s forests.

Populus nigra L. is one of the rarest and most endangered tree species in Western and Central Europe. Its genetic diversity is of great importance in enabling a native riparian population to survive and reproduce under changing environmental conditions. The aim of this research was assessment of P. nigra viability in one of the best preserved riparian ecosystems in Europe, Special Nature Reserve "Gornje Podunavlje" (Upper Danube), Serbia. Additionally, the analysis of the genetic diversity was made to support the effective conservation in the future. During our study, we have mapped 931 P. nigra trees, which were used for the assessment of present native population. Furthermore, we used 14 microsatellite markers to assess the genetic structure of this this population. Viability assessment showed considerable occurrence of P. nigra in the research area, even though the results show fragmentation. P. nigra occurs mostly individually or in small groups of trees, and has a non-sustainable age structure due to insufficient or lacking regeneration. Despite the limited size of the studied population, the apparent overall genetic diversity was high (He = 0.759) and comparable to other known native populations of P. nigra along the Danube basin. However, the results also confirmed existence of recent bottleneck effect. Significantly positive and quite high F is value (0.147) was noted, which may be ascribed to the "Wahlund effect" because of the population substructure that was revealed by the STRUC-TURE analysis (K=2). Although results say that coverage of native stands is not so promising, most of selected trees within our research assessed showed good viability with potential for natural reproduction However, the problem is that suitable areas for natural seedling establishment are scarce and with that gene flow is probably limited. The fragmentation of the area must be reduced and isolated stands must be interlinked as there is need to create larger non-fragmented areas.

The oak forests in the area of the Special Nature Reserve ?Obedska bara? are extremely valuable and unique at the national and European level, which is why they are recognized as the fundamental value of the reserve. The local name of the oldest complex of oak forests in the special nature reserve is ?Debela gora rainforest? and it has been placed in the first degree of protection regime. Given that it is included in the first level protection regime, the priority goal of forest management is the preservation of biological diversity, which implies the absence of any management treatments. As a result of this approach, there is a loss of oaks and other species with higher demands for light, therefore other species such as hornbeam appear and take over, which is subject of analysis in this paper. Given that the replacement of species is not the goal of forest management in areas under the first degree of protection, this paper proposes forest management protective measures aimed at the restoration and revitalization of areas that could be implemented with the consent of the competent institutions.