IT technologies and remote sensing

Digital terrain models (DTMs) present important data source for different applications in
environmental disciplines including forestry. At regional level, DTMs are commonly created
using airborne digital photogrammetry or airborne laser scanning (ALS) technology. This
study aims to evaluate the vertical accuracy of DTMs of different spatial resolutions derived
from high-density ALS data and existing photogrammetric (PHM) data in the dense lowland
even-aged pedunculate oak forests located in the Pokupsko basin in Central Croatia. As expected,
the assessment of DTMs’ vertical accuracy using 22 ground checkpoints shows higher
accuracy for ALS-derived than for PHM-derived DTMs. Concerning the different resolutions
of ALS-derived (0.5 m, 1 m, 2 m, 5 m) and PHM-derived DTMs (0.5 m, 1 m, 2 m, 5 m,
8 m) compared in this research, the ALS-derived DTM with the finest resolution of 0.5 m
shows the highest accuracy. The root mean square error (RMSE) and mean error (ME) values
for ALS-derived DTMs range from 0.14 m to 0.15 m and from 0.09 to 0.12 m, respectively,
and the values decrease with decreasing spatial resolution. For the PHM-derived DTMs, the
RMSE and ME values are almost identical regardless of resolution and they are 0.35 m and
0.17 m, respectively. The findings suggest that the 8 m spatial resolution is optimal for a
given photogrammetric data, and no finer than 8 m spatial resolution is required. This research
also reveals that the national digital photogrammetric data in the study area contain certain
errors (outliers) specific to the terrain type, which could considerably affect the DTM accuracy.
Thus, preliminary evaluation of photogrammetric data should be done to eliminate possible
outliers prior to the DTM generation in lowland forests with flat terrain. In the absence
of ALS data, the finding in this research could be of interests to countries, which still rely on
similar photogrammetric data for DTM generation.

Salvage logging is performed to remove the fallen and damaged trees after a natural disturbance,
e.g., fire or windstorm. From an economic point of view, it is desirable to remove the
most valuable merchantable timber, but usually, the process depends mainly on topography
and distance to forest roads. The objective of this study was to evaluate the suitability of the
Black-Bridge satellite imagery for the spatial distribution of salvage cutting in southern Poland
after the severe windstorm in July 2015. In particular, this study aimed to determine which
factors influence the spatial distribution of salvage cutting. The area of windthrow and the
distribution of salvage cutting (July–August 2015 and August 2015–May 2016) were delineated
using Black-Bridge satellite imagery. The distribution of the polygons (representing
windthrow and salvage cutting) was verified with maps of aspect, elevation and slope, derived
from the Digital Terrain Model and the distance to forest roads, obtained from the Digital
Forest Map. The analysis included statistical modelling of the relationships between the process
of salvage cutting and selected geographical and spatial features. It was found that the higher
the elevation and the steeper the slope, the lower the probability of salvage cutting. Exposure
was also found to be a relevant factor (however, it was difficult to interpret) as opposed to the
distance to forest roads.

Since wood represents an important proportion of the delivered cost, it is important to embrace
and implement correct measurement procedures and technologies that provide better wood
volume estimates of logs on trucks. Poor measurements not only impact the revenue obtained
by haulage contractors and forest companies but also might affect their contractual business
relationship. Although laser scanning has become a mature and more affordable technology in
the forestry domain, it remains expensive to adopt and implement in real-life operating
conditions. In this study, multi-view Structure from Motion (SfM) photogrammetry and
commercial 3D image processing software were tested as an innovative and alternative method
for automated volumetric measurement of truckloads. The images were collected with a small
UAV, which was flown around logging trucks transporting Eucalyptus nitens pulplogs.
Photogrammetric commercial software was used to process the images and generate 3D models
of each truckload. The levels of accuracy obtained with multi-view SfM photogrammetry and
3D reconstruction obtained in this study were comparable to those reported in previous studies
with laser scanning systems for truckloads with similar logs and species. The deviations between
the actual and predicted solid volume of logs on trucks ranged between –3.2% and 3.5%, with
an average deviation of –0.05%. In absolute terms, the average deviation was only 0.5 m3 or
1.7%. Although several aspects must be addressed for the operational implementation of SfM
photogrammetry, the results of this study demonstrate the great potential for this method to be
used as a cost-effective tool to aid in the determination of the solid volume of logs on trucks.

The Airborne Laser Scanning (ALS) technology has been implemented in operational forest
inventories in a number of countries. At the same time, as a cost-effective alternative to ALS,
Digital Aerial Photogrammetry (PHM), based on aerial images, has been widely used for the
past 10 years. Recently, PHM based on Unmanned Aerial Vehicle (UAV) has attracted great
attention as well. Compared to ALS, PHM is unable to penetrate the forest canopy and, ultimately,
to derive an accurate Digital Terrain Model (DTM), which is necessary to normalize
point clouds or Digital Surface Models (DSMs). Many countries worldwide, including Croatia,
still rely on PHM, as they do not have complete DTM coverage by ALS (DTMALS). The
aim of this study is to investigate if the official Croatian DTM generated from PHM (DTMPHM)
can be used for data normalization of UAV-based Digital Surface Model (DSMUAV) and estimating
plot-level mean tree height (HL) in lowland pedunculate oak forests. For that purpose,
HL estimated from DSMUAV normalized with DTMPHM and with DTMALS were generated and
compared as well as validated against field measurements. Additionally, elevation errors in
DTMPHM were detected and eliminated, and the improvement by using corrected DTMPHM
(DTMPHMc) was evaluated. Small, almost negligible variations in the results of the leave-oneout
cross-validation were observed between HL estimated using proposed methods. Compared
to field data, the relative root mean square error (RMSE%) values of HL estimated from DSMUAV
normalized with DTMALS, DTMPHM, and DTMPHMc were 5.10%, 5.14%, and 5.16%, respectively.
The results revealed that in the absence of DTMALS, the existing official Croatian DTM
could be readily used in remote sensing based forest inventory of lowland forest areas. It can
be noted that DTMPHMc did not improve the accuracy of HL estimates because the gross errors
mainly occurred outside of the study plots. However, since the existence of the gross errors in
Croatian DTMPHM has been confirmed by several studies, it is recommended to detect and
eliminate them prior to using the DTMPHM in forest inventory.

The emergence of hand-held Personal Laser Scanning (H-PLS) systems in recent years resulted in initial research on the possibility of its application in forest inventory, primarily for the estimation of the main tree attributes (e.g. tree detection, stem position, DBH, tree height, etc.). Research knowledge acquired so far can help to direct further research and eventually include H-PLS into operational forest inventory in the future. The main aims of this review are:

Þ   to present the current state of the art for H-PLS systems

Þ   briefly describe the fundamental concept and methods for H-PLS application in forest inventory

Þ   provide an overview of the results of previous studies

Þ   emphasize pros and cons for H-PLS application in forest inventory in relation to conventional field measurements and other similar laser scanning systems

Þ   highlight the main issues that should be covered by further H-PLS-based forest inventory studies.

Proximal sensing technologies are becoming widely used across a range of applications in environmental sciences. One of these applications is in the measurement of the ground surface in describing soil displacement impacts from wheeled and tracked machinery in the forest. Within a period of 2–3 years, the use photogrammetry, LiDAR, ultrasound and time-of-flight imaging based methods have been demonstrated in both experimental and operational settings. This review provides insight into the aims, sampling design, data capture and processing, and outcomes of papers dealing specifically with proximal sensing of soil displacement resulting from timber harvesting. The work reviewed includes examples of sensors mounted on tripods and rigs, on personal platforms including handheld and backpack mounted, on mobile platforms constituted by forwarders and skidders, as well as on unmanned aerial vehicles (UAVs). The review further highlights and discusses the benefits, challenges, and some of the shortcomings of the various technologies and their application as interpreted by the authors.

The majority of the work reviewed reflects pioneering approaches and innovative applications of the technologies. The studies have been carried out almost simultaneously, building on little or no common experience, and the evolution of standardized methods is not yet fully apparent. Some of the issues that will likely need to be addressed in developing this field are (i) the tendency toward generating apparently excessively high resolution micro-topography models without demonstrating the need for or contribution of such resolutions on accuracy, (ii) the inadequacy of conventional manual measurements in verifying the accuracy of these methods at such high resolutions, and (iii) the lack of a common protocol for planning, carrying out, and reporting this type of study.

Cable yarders are often the preferred harvesting system when extracting trees on steep terrain. While the practice of cable logging is well established, productivity is dependent on many stand and terrain variables. Being able to continuously monitor a cable yarder operation would provide the opportunity not only to manage and improve the system, but also to study the effect on operations in different conditions.

This paper presents the results of an automated monitoring system that was developed and tested on a series of cable yarder operations. The system is based on the installation of a Geographical Navigation Satellite System (GNSS) onto the carriage, coupled with a data-logging unit and a data analysis program. The analysis program includes a set of algorithms able to transform the raw carriage movement data into detailed timing elements. Outputs include basic aspects such average extraction distance, average inhaul and outhaul carriage speed, but is also able to distinguish number of cycles, cycle time, as well as break the cycles into its distinct elements of outhaul, hook, inhaul and unhook.

The system was tested in eight locations; four in thinning operations in Italy and four clear-cut operations in New Zealand, using three different rigging configuration of motorized slack-pulling, motorized grapple and North Bend. At all locations, a manual time and motion study was completed for comparison to the data produced by the newly developed automated system. Results showed that the system was able to identify 98% of the 369 cycles measured. The 8 cycles not detected were directly attributed to the loss of GNSS signal at two Italian sites with tree cover. For the remaining 361 cycles, the difference in gross cycle time was less than 1% and the overall accuracy for the separate elements of the cycle was less than 3% when considered at the rigging system level. The study showed that the data analyses system developed can readily convert GNSS data of the carriage movement into information useful for monitoring and studying cable yarding operations.

The objective of this study was to construct a BeiDou navigation satellite system (BDS)/global positioning system (GPS)-integrated positioning algorithm that meets the accuracy requirement of forest surveys and to analyze its accuracy to provide theoretical and technical support for accurate positioning and navigation in forests. The Quercus variabilis broad-leaved forest in Jiufeng National Forest Park and the Sabina Coniferous forest in Dongsheng Bajia forest farm were selected as the study area. A Sanding T-23 multi-frequency three-constellation receiver and a u-blox NEO-M8T multi-constellation receiving module were used for continuous observation under the forest canopy. Compared with T-23, the u-blox NEO-M8T is much lighter and more flexible in the forest. The BDS/GPS-integrated positioning algorithm for forests was constructed by temporally and spatially unifying the satellite systems and using a reasonable observed value weighting method. Additionally, the algorithm is also written into the RTKLIB software to calculate the three-dimensional (3D) coordinates of the forest observation point in the World Geodetic System 1984 (WGS-84) coordinate system. Finally, the results were compared with the positioning results obtained using GPS alone. The experimental results indicated that, compared with GPS positioning, there were 13–27 visible satellites available for the BDS/GPS-integrated positioning algorithm for forests, far more than the satellites available for the GPS positioning algorithm alone. The Position Dilution of Precision (PDOP) values for the BDS/GPS-integrated positioning ranged from 0.5 to 1.9, lower than those for GPS positioning. The signal noise ratio (SNR) of the BDS/GPS-integrated satellite signals and GPS satellite signals were both in the range of 10–50 dB-Hz. However, because there were more visible satellites for the BDS/GPS-integrated positioning, the signals from the BDS/GPS-integrated satellites were stronger and had a more stable SNR than those from the GPS satellites alone. The results obtained using the BDS/GPS-integrated positioning algorithm for forests had significantly higher theoretical and actual accuracies in the X, Y and Z directions than those obtained using the GPS positioning algorithm. This suggests that the BDS/GPS-integrated positioning algorithm can obtain more accurate positioning results for complex forest environments.

High temporal resolution of synthetic aperture radar (SAR) imagery (e.g., Sentinel-1 (S1) imagery) creates new possibilities for monitoring green vegetation in urban areas and generating land-cover classification (LCC) maps. This research evaluates how different pre-processing steps of SAR imagery affect classification accuracy. Machine learning (ML) methods were applied in three different study areas: random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGB). Since the presence of the speckle noise in radar imagery is inevitable, different adaptive filters were examined. Using the backscattering values of the S1 imagery, the SVM classifier achieved a mean overall accuracy (OA) of 63.14%, and a Kappa coefficient (Kappa) of 0.50. Using the SVM classifier with a Lee filter with a window size of 5×5 (Lee5) for speckle reduction, mean values of 73.86% and 0.64 for OA and Kappa were achieved, respectively. An additional increase in the LCC was obtained with texture features calculated from a grey-level co-occurrence matrix (GLCM). The highest classification accuracy obtained for the extracted GLCM texture features using the SVM classifier, and Lee5 filter was 78.32% and 0.69 for the mean OA and Kappa values, respectively. This study improved LCC with an evaluation of various radiometric and texture features and confirmed the ability to apply an SVM classifier. For the supervised classification, the SVM method outperformed the RF and XGB methods, although the highest computational time was needed for the SVM, whereas XGB performed the fastest. These results suggest pre-processing steps of the SAR imagery for green infrastructure mapping in urban areas. Future research should address the use of multitemporal SAR data along with the pre-processing steps and ML algorithms described in this research.

Many dendrometric parameters have been estimated by light detection and ranging (LiDAR) technology over the last two decades. Handheld mobile laser scanning (HMLS), in particular, has come into prominence as a cost-effective data collection method for forest inventories. However, most pilot studies were performed in domesticated landscapes, where the environmental settings were far from those presented by (near)natural forest ecosystems. Besides, these studies consisted of numerous data processing steps, which were challenging when employed by manual means. Here we present an automated approach for deriving key inventory data using the HMLS method in natural forest areas. To this end, many algorithms (e.g., cylinder/circle/ellipse fitting) and machine learning models (e.g., random forest classifier) were used in the data processing stage for estimation of the tree diameter at breast height (DBH) and the number of trees. The estimates were then compared against the reference data obtained by field measurements from six forest sample plots. The results showed that correlations between the estimated and reference DBHs were very strong at the plot level (r=0.83–0.99, p<0.05). The average RMSE for tree DBHs was 1.8 cm at the forest landscape level. As for tree detection, 92.5% of 292 trunks were correctly classified on point cloud data. In general, estimation accuracy was sufficient for operational forest inventory needs. However, they could markedly decrease in »hard plots« located at rocky terrains with dense undergrowth and irregular trunks. We concluded that area-based forest inventories might hugely benefit from the HMLS method, particularly in »easy plots«. By improving the algorithmic performances, the accuracy levels can be further increased by future research.

When investigating the forwarding process within the timber supply chain, insufficient data often inhibits long-term studies or make real-time optimisation of the logistics process difficult. Information sources to compensate for this lack of data either depend on other processing steps or they need additional, costly hardware, such as conventional crane scales. An innovative weight-detection concept using information provided by a commonly available hydraulic pressure sensor may make the introduction of a low-cost weight information system possible. In this system, load weight is estimated by an artificial neural network (ANN) based on machine data such as the hydraulic pressure of the inner boom cylinder and the grapple position.

In our study, this type of crane scale was set up and tested under real working conditions, implemented as a cloud application. The weight scale ANN algorithm was therefore modified for robustness and executed on data collected with a commonly available telematics module. To evaluate the system, also with regard to larger sample sizes, both direct weight-reference measurements and additional volume-reference measurements were made. For the second, locally valid weight-volume conversion factors for mainly Norway spruce (Picea abies, 906 kg m-3, standard error of means (SEM) of 13.6 kg m-3), including mean density change over the observation time (–0.16% per day), were determined and used as supportive weight-to-volume conversion factor.

Although the accuracy of the weight scale was lower than in previous laboratory tests, the system showed acceptable error behaviour for different observation purposes. The twice-observed SEM of 1.5% for the single loading movements (n=95, root-mean-square error (RMSE) of 15.3% for direct weight reference; n=440, RMSE=33.2% for volume reference) enables long-term observations considering the average value, but the high RMSE reveals problems with regard to the single value information.

The full forwarder load accuracy, as unit of interest, was observed with an RMSE of 10.6% (n=41), considering a calculated weight-volume conversion as reference value. An SEM of 5.1% for already five observations with direct weight reference provides a good starting point for work-progress observation support.

An accurate tree diameter (DBH) measurement is a significant component of forest inventory. This study assessed the reliability of remote dendrometers to measure tree DBH. We compared direct caliper measurements (reference measurements) to the remote measurements collected from a laser caliper and a smartphone at 0.5 m, 1 m, and 1.5 m distances from each tree within three forest types (pine, oak, and poplar forests). In general, all remote dendrometers underestimated the mean diameter compared to direct caliper measurements, regardless of forest types and distances. We observed that the mean deviation of direct caliper measurement and smartphone measurement at 1.5 m within a pine forest and oak forest were the lowest (0.3 cm and 0.36 cm, respectively). The deviations between direct caliper measurements and smartphone measurements at a 0.5 m distance, across forest types, were noticeably larger compared to others. An ANOVA test was used to determine whether significant deviations existed between caliper measurements and remote measurements at a specific distance, and among three different forest types. We rejected the null hypothesis, which suggested that there were no statistically significant differences (p<0.05) between tree DBH measurements obtained from the direct caliper measurements and indirect measurements (smartphone and laser caliper) captured at a distance. Then, a post-hoc test was performed to examine which set of estimated deviations was different from the reference data. The results suggested that indirect tree DBH measurements using the smartphone app at 1 m and 1.5 m in certain forest types (pine and oak) were not significantly different from direct tree DBH measurements. Also, our test results mostly indicated no significant difference within each forest, except for measurements using the smartphone app at 0.5 m across all forest types when the smartphone measurements were compared to laser caliper measurements. Although forest characteristics and measurement distance may play an important role in remote tree DBH measurement accuracy, the smartphone app may be used as a practical alternative to direct measurement in measuring the DBH of a tree, which may be a positive development for forestry due to the increased use of smartphones and the availability of a free measure app.

In the last couple of years, there have been a great number of articles that cover and emphasize the advantages and possibilities that UAS (Unmanned Air System) offers in forest ecosystem research. In the available research, alongside UAS, the importance of developing sensors that are designed to be used with UAV (Unamnned Air Vehicle), a flight programming software and UAS collected data processing software have been pointed out. With the widespread use of high-precision sensors and accompanying software in forestry, it is possible to obtain accurate data in a short time that replaces long-term manpower in the field with equal or in some cases, such as windthrow calculation or wildlife counting, greater accuracy. The former practice of manual imagery processing is being partly replaced with automated approaches. The paper analyses studies that deal with some form of application of UAS in forestry, e.g. forest inventory, forest operations, ecological monitoring, forest pests and forest fires, and wildlife monitoring. In the forest inventory, a large number of studies deal with the possibilities of applying UAS in mapping vegetation and individual trees, morphological research of individual parts of trees, surface analysis, etc. The use of remote and proximal sensing technologies in forest engineering has mainly been focused on defining surface roughness and topology, road geometry, planning and maintenance, ground-based and cable-based harvesting and soil characteristics and displacement. Wildfire monitoring already relies heavily on the use of UAS and thermal cameras in operations, and it is similar to the mapping of windthrow or directions of the spread of certain insects important for forestry. In wildlife research, numerous studies deal with abundance research of individual terrestrial birds and mammals using UAS thermal imagery. With some drawbacks such as wildlife disturbance or limited UAV range, common to most of the processed studies are positive attitudes regarding the application of UAS in forestry sensing and monitoring, which is slowly becoming a common operative practice, with the scientists’ focus being on developing automated approaches in UAS imagery processing. Reducing the error by improving the technological characteristics of the sensors will in the long run reduce the number of people required to collect data important for forestry, reduce risks and in some cases increase accuracy.

In Slovenia, torrent areas and forest roads are being regulated and built mostly in steep, erosion-prone areas. In addition to the geometry of extrapolated works, calculating bulk factors is key for estimating haulage masses. We have determined bulk factors for compact carbonate rock, mixed soil, and carbonate deposits. Each construction site was recorded with an unmanned aerial vehicle (UAV) before the excavation and after every 4±2 m3 of excavated material. The average point cloud density was 9535 points/m2. We processed the point clouds from each construction site and determined the difference in volume between the volume of excavated area and the volume of deposited material. The average bulk factor for compact carbonate rock is 1.42, 1.20 for mixed soil and 1.15 for carbonate deposits (calculated for fully loaded eight-wheeled truck). The calculated bulk factors for soils and carbonate deposits match with the already established values, while the factor for compact rock is 20% lower than the factor currently in use by the Slovenian forest engineers.

This study intended to develop a road network planning for timber harvesting in tropical forests in Peru using georeferenced and field data and Geographic Information System (GIS). The Digital Elevation Model (DEM) Alos Palsar 12.5 m was used. The DEM was processed to generate the hydrography and terrain slope maps. A weighted raster was created using overlapping weights of the slope raster and the hydrography map. We created a least cost path raster by using the weighted raster origin and destination points. We used a Triangular Irregular Network (TIN) to validate the Least Cost Path. In addition, histograms of the trajectory of each path with altitude and slope values ​​ were generated. We observed that the forest road planning using GIS provided better definition (economically and environmentally) of road networks in our forest site than those traditionally defined using conventional mapping techniques.

This article presents the performance of a vegetation fuel type (FT) classification based on conditional rules according to the Prometheus system, including an analysis of the effect of cell size and scan density on mapping vertical structural types, exemplified as FT, using exclusively LiDAR data. Since the Prometheus system does not specify any criterion for the minimum extension where those methodologies can be applied, we searched for the optimal classification cell size by gridding the study area at 20 and 40 m cell sizes. We also included a study of the effects of varying the scan density from 2 to 0.5 pulses·m-2. To validate the classification method, we used a stratified random sampling without replacement of 15 cells per FT and made an independent visual assessment of FTs. The best results in terms of precision were obtained for the combination of 0.5 pulses·m-2 and 20 m-resolution dataset, with an overall accuracy of 84.13%. It was also showed that an increase in scan density would not improve the global accuracy of the classification, but it would be desirable for a better detection of the shrub stratum.

Uncontrolled spread of eastern red cedar invades the United States Great Plains prairie ecosystems and lowers biodiversity across native grasslands. The eastern red cedar (ERC) infestations cause significant challenges for ranchers and landowners, including the high costs of removing mature red cedars, reduced livestock forage feed, and reduced revenue from hunting leases. Therefore, a fleet of autonomous ground vehicles (AGV) is proposed to address the ERC infestation. However, detecting the target tree or trunk in a rangeland environment is critical in automating an ERC cutting operation. A tree trunk detection method was developed in this study for ERC trees trained in natural rangeland environments using a deep learning-based YOLOv5 model. An action camera acquired RGB images in a natural rangeland environment. A transfer learning method was adopted, and the YOLOv5 was trained to detect the varying size of the ERC tree trunk. A trained model precision, recall, and average precision were 87.8%, 84.3%, and 88.9%. The model accurately predicted the varying tree trunk sizes and differentiated between trunk and branches. This study demonstrated the potential for using pretrained deep learning models for tree trunk detection with RGB images. The developed machine vision system could be effectively integrated with a fleet of AGVs for ERC cutting. The proposed ERC tree trunk detection models would serve as a fundamental element for the AGV fleet, which would assist in effective rangeland management to maintain the ecological balance of grassland systems.

To further develop forest production, higher automation of forest operations is required. Such endeavour promotes research on unmanned forest machines. Designing unmanned forest machines that exercise forwarding requires an understanding of positioning and angle estimations of logs after cutting and delimbing have been conducted, as support for subsequent crane loading work. This study aims to improve the automation of the forwarding operation and presents a system to realize real-time automatic detection, positioning, and angle estimation of harvested logs implemented on an existing unmanned forest machine experimental platform from the AORO (Arctic Off-Road Robotics) Lab. This system uses ROS as the underlying software architecture and a Zed2 camera and NVIDIA JETSON AGX XAVIER as the imaging sensor and computing platform, respectively, utilizing the YOLOv3 algorithm for real-time object detection. Moreover, the study combines the processing of depth data and depth to spatial transform to realize the calculation of the relative location of the target log related to the camera. On this basis, the angle estimation of the target log is further realized by image processing and color analysis. Finally, the absolute position and log angles are determined by the spatial coordinate transformation of the relative position data. This system was tested and validated using a pre-trained log detector for birch with a mean average precision (mAP) of 80.51%. Log positioning mean error did not exceed 0.27 m and the angle estimation mean error was less than 3 degrees during the tests. This log pose estimation method could encompass one important part of automated forwarding operations.

Introduction: Cable yarding is a technology that enables efficient and sustainable use of timber resources in mountainous areas. Carriages as an integral component of cable yarding systems have undergone significant development in recent decades. In addition to mechanical and functional developments, carriages are increasingly used as carrier platforms for various sensors. The goal of this study was to assess the accuracy of individual standing tree and stand variable estimates obtained by a mobile laser scanning system mounted on a cable yarder carriage.

Methods: Eight cable corridors were scanned across two forest stands. Four different scan variants were conducted, differing in the movement speed of the carriage and the direction of movement during scanning. An algorithm for tree detection, diameter and height estimation was applied to the 3D datasets and evaluated against manual tree measurements.

Results: The analysis of the 3D scans showed that the individual tree parameters strongly depend on the scan variant and the distance of each individual tree to the skyline. This was due to changing 3D point densities and occlusion effects. It turned out that scan variant 1, in which the scan was performed during slow carriage movement downwards and back upwards again, was advantageous. At a distance of 10 m, which is half of the recommended corridor spacing of 20 m for whole tree cable yarding, 95.44% of the trees in stand 1 and 92.16% of the trees in stand 2 could be detected automatically. The corresponding root mean sqare errors of the diameter at breast height estimatimations were 1.59 cm and 2.23 cm, respectively. The root mean square errors of the height measurements were 2.94 m and 4.63 m.

Conclusions: The results of this study can help to further advance the digitization of cable yarding and timber flow from the standing tree to the sawmill. However, this requires further development steps in cable yarder, carriage, and laserscanner technology. Furthermore, there is also a need for more efficient software routines to take the next steps towards precision forestry.

Harvested timber requires efficient and accurate measurements for timber trade. Recently, the amount of timber that is harvested for industrial purposes is growing and methods for roundwood measurements are under constant development. Some of these solutions are with certification confirming its accuracy, some of them are proposed for general use without certification. The aim of this paper was to select the best and most useful electronic solution for timber volume masurement and calculation, and to recommend the most effective and accurate solution for future timber trade. Three photo-optical systems were tested: LogStackPro, iFovea and Timbeter. Each system was used to measure 71 stacks of pine and oak roundwood, which amounted to 3481.15 cubic meter stacked. Timber volumes obtained from the manual measurements were used as reference. Volumes obtained from the photo-optical systems were larger in comparison with the volume from manual measurements, by 3.37, 8.07 and 9.08%, in LogStackPro, iFovea and Timbeter, respectively. It was concluded from the tested systems that, currently, the most recommended solution for timber measurement will be LogStack Pro, which also presented, in most cases, the smallest deviations from the volume obtained in manual measurements.

Advances in sensor technology and computing performance has brought us into an era of digital forestry where a forest environment can be digitally replicated. At the same time, an increasing interest in the use of unmanned vehicles and other autonomous mobile systems (AMSs) in forest mapping and operations has emerged. However, a forest is an unstructured and rather complex environment for AMSs to operate in, and usually some kind of a priori information of traversability is required. The aim of this study was to assess forest traversability for AMSs using high-density airborne laser scanning (ALS) point clouds. It was assumed that such point clouds acquired from a helicopter flying at a low altitude can be used to characterise vegetation obstacles affecting forest traversability. A voxel-based vegetation occupancy analysis was carried out with the aim to detect open space to define traversable three-dimensional space. The experimental setup included seven sample plots (32×32 m) representing diverse boreal forest structures. Terrestrial laser scanning (TLS) was used for obtaining reference for vegetation occupancy. Comparison between ALS and TLS revealed an overall accuracy of 0.85–0.94 with a recall of 0.78–0.91 and a precision of 0.62–0.74 for ALS-based voxel classification for vegetation occupancy depending on forest structure. This implies that up to 91% of voxels assigned a classification »occupied« based on the TLS could be correctly classified using the ALS, while up to 74% of voxels assigned a classification »occupied« using the ALS were occupied based on the TLS. Density of low vegetation accounted for 83% of the variation in accuracy and precision. The feasibility of vegetation occupancy information to be used by an AMS for navigation was also demonstrated. It was assumed that the ALS data convey as sufficient information of AMS path planning as does the TLS data. The experiments showed that out of 1393 randomly generated paths based on empty space detected by the ALS, 72% were considered feasible when validated with the TLS data. The success rate in path planning varied from 0.54 to 0.92 between the sample plots and was seemingly affected by vegetation density that accounted for 53% of variation in success rate. Altogether, the demonstrated possibility to predefine forest traversability using remote sensing will support the use of AMSs in forestry.

Forestry equipment simulators offer opportunities for new operators to become familiar with operating logging machines as well as a promising solution to the high costs of training forestry equipment operators. Current literature lacks a synthesis on how best to train forestry equipment operators using simulators. The goal of this review was to identify effective ways to incorporate forestry equipment simulators into an equipment operator training curriculum. We analyzed a total of 14 independent studies in which construction and forestry equipment operators were trained on simulators and engaged in discussions with nine professionals in the field of heavy equipment operator training. In this review, traditional machine training and simulator training practices are introduced. Then, four key aspects of skill acquisition for forestry equipment operators are identified. Information collected from peer-reviewed literature and discussions with industry experts are used to consider how each aspect of skill acquisition is addressed in both traditional training using real machines and simulator-based training. Drawing on these sources, benefits and drawbacks of traditional machine training and simulator-based training for forestry equipment operators are synthesized and discussed. Finally, a model for an integrated and adaptive training curriculum that incorporates principles and technologies from both traditional machine training and simulator training is presented.

Differentiating areas of insect damage in forests from areas of healthy vegetation and predicting the future spread of damage increase are an important part of forest health monitoring. Thanks to the wide coverage and temporal observation advantage of remote sensing data, predicting the future direction of insect damage spread can enable accurate and uninterrupted management and operational control to minimize damage. However, due to the large amount of remotely sensed data, it is difficult to process the data and to identify damage distinctions. Therefore, this paper proposes a spatio-temporal Autoregressive Integrated Moving-Average (ARIMA) prediction model based on the Machine Learning technique for processing big data by monitoring oak lace bug (Corythucha arcuata (Heteroptera: Tingidae)) damage with remote sensing data. The advantage of this model is the automatic selection of optimal parameters to provide better forecasting with univariate time series. Thus, multiple spatio-temporal warning levels are distinguished according to the damage growth trend in the series, and the network is constructed with improved time series to better predict future insect damage spread. In the proposed model, the historical Red (R) – Green (G) – Blue (B) bands of the Sentinel-2 (GSD 10 m) satellite were tested as a dataset for the oak lace bug damage in the oak forest situated in the campus of Düzce University, Turkey. The dataset, which contained 38 images for each of the RGB bands, was modeled using the open source R programming language for the peak damage period in 2021. As a result of the test, significant correlations were found between the synthetic and true images (True and synthetic band 2: r=0.960, p<0.001; True and synthetic band 3: r=0.945, p<0.001; True and synthetic band 4: r=0.962, p<0.001). Then, the 48-month time series bands were modeled, and the band estimates were made to predict the August 2023 spread. Finally, a synthetic composite image was created for future prediction using the predicted bands. The tests showed that the model had a good performance in insect damage monitoring. With open access Sentinel-2 images, the proposed model achieved the highest prediction accuracy with a rate of 96%, and had a small prediction error.

With recent technological development, photo-optical measurement systems in mobile devices have been increasingly used for automatic wood volume estimation because of their ease of use and efficiency. This study aimed to evaluate the use of photo-optical mobile apps for measuring solid wood volume of the stack in Turkish forestry practices. For this study, 21 log stacks were measured using the traditional technique and two photo-optical mobile apps – iFovea Pro and Timbeter. A strong correlation was found between the traditionally measured solid wood volume of the stack and the volume estimated using both photo-optical apps, the number of logs in the stack, and the mean diameter of the stack. The estimated number of trees from the two apps and manual measurement were not statistically different. However, statistical differences were observed between all three measurement approaches for the mean diameter of the stack. Also, statistical test results indicated mixed results for estimated solid wood volume in the stack. In addition, the study tested whether both apps correctly measure the diameter of the logs in the stack. Thus, manually measured diameter of the randomly selected 50 trees within 21 stacks was compared to the log diameters measured automatically using both mobile apps. The results indicated no statistical difference between the three measurement approaches. The study results are promising for using photo-optical mobile apps in Turkish forestry in terms of transition to digital forestry. However, there are still opportunities to improve the capabilities of the method through further analysis of estimating stack volume using the image from both sides of the logs considering different quality and diameter classes with bark conditions.

Tracking roadside wood inventories is necessary for wood procurement. However, this operation is increasingly problematic due to the costs associated with reaching remote sites, labour shortage, and the methods providing limited information on the characteristics of the logs required for transport planning. To overcome these problems, an automated procedure has been developed in Arcmap to characterize individual wood pile inventories at roadside by using GPS points of forwarders, harvester production files, and the road network shape files. An inventory of the logs in the harvest area followed by a wood pile inventory at the roadside were made to evaluate how the procedure could trace logs from machine operating trail to predicted wood pile locations. The study was done at six harvest blocks in the Saguenay-Lac-Saint-Jean region in the province of Quebec, Canada. The procedure was not able to differentiate individual wood piles and aggregated several piles into predicted unloading areas. Results indicated a similarity index of 72% to manually inventoried wood piles. The similarity index could be explained by the low percentage of inventoried unpredicted wood pile lengths (3%) and a high percentage of overpredicted wood pile lengths (27%). The positive allocation rate could not be assessed at the level of the individual piles. On the other hand, the procedure properly allocated 96% of the logs to unloading areas. With the level of precision obtained, the developed procedure could be beneficial for managing the transportation of wood at the level of the road segment since it provides all the dendrometric data of the logs available for transport without requiring human intervention in the forest.

Tracking roadside wood inventories is necessary for wood procurement. However, this operation is increasingly problematic due to the costs associated with reaching remote sites, labour shortage, and the methods providing limited information on the characteristics of the logs required for transport planning. To overcome these problems, an automated procedure has been developed in Arcmap to characterize individual wood pile inventories at roadside by using GPS points of forwarders, harvester production files, and the road network shape files. An inventory of the logs in the harvest area followed by a wood pile inventory at the roadside were made to evaluate how the procedure could trace logs from machine operating trail to predicted wood pile locations. The study was done at six harvest blocks in the Saguenay-Lac-Saint-Jean region in the province of Quebec, Canada. The procedure was not able to differentiate individual wood piles and aggregated several piles into predicted unloading areas. Results indicated a similarity index of 72% to manually inventoried wood piles. The similarity index could be explained by the low percentage of inventoried unpredicted wood pile lengths (3%) and a high percentage of overpredicted wood pile lengths (27%). The positive allocation rate could not be assessed at the level of the individual piles. On the other hand, the procedure properly allocated 96% of the logs to unloading areas. With the level of precision obtained, the developed procedure could be beneficial for managing the transportation of wood at the level of the road segment since it provides all the dendrometric data of the logs available for transport without requiring human intervention in the forest.