High spatial resolution imagery provided by unmanned aerial vehicles (UAVs) can yield accurate and efficient estimation of tree dimensions and canopy structural variables at the local scale. We flew a low-cost, lightweight UAV over an experimental Pinus pinea L. plantation (290 trees distributed over 16 ha with different fertirrigation treatments) to determine the tree positions and to estimate individual tree height (h), diameter (d), biomass (wa), as well as changes in these variables between 2015 and 2017. We used Structure from Motion (SfM) and 3D point cloud filtering techniques to generate the canopy height model and object-based image analysis to delineate individual tree crowns (ITC). ITC results were validated using accurate field measurements over a subsample of 50 trees. Comparison between SfM-derived and field-measured h yielded an R 2 value of 0.96. Regressions using SfM-derived variables as explanatory variables described 79% and 86-87% of the variability in d and wa, respectively. The height and biomass growth estimates across the entire study area for the period 2015-2017 were 0.45 m ± 0.12 m and 198.7 ± 93.9 kg, respectively. Significant differences (t-test) in height and biomass were observed at the end of the study period. The findings indicate that the proposed method could be used to derive individual-tree variables and to detect spatio-temporal changes, highlighting the potential role of UAV-derived imagery as a forest management tool.
Aim of study: The study aims to analyse the potential use of lowcost unmanned aerial vehicle (UAV) imagery for the estimation of Pinus pinea L. variables at the individual tree level (position, tree height and crown diameter).Area of study: This study was conducted under the PINEA project focused on 16 ha of umbrella pine afforestation (Portugal) subjected to different treatments.Material and methods: The workflow involved: a) image acquisition with consumer-grade cameras on board an UAV; b) orthomosaic and digital surface model (DSM) generation using structure-from-motion (SfM) image reconstruction; and c) automatic individual tree segmentation by using a mixed pixel-and region-based algorithm.Main results: The results of individual tree segmentation (position, height and crown diameter) were validated using field measurements from 3 inventory plots in the study area. All the trees of the plots were correctly detected. The RMSE values for the predicted heights and crown widths were 0.45 m and 0.63 m, respectively.Research highlights: The results demonstrate that tree variables can be automatically extracted from high resolution imagery. We highlight the use of UAV as a fast, reliable and cost-effective technique for small scale applications.
Crown fire initiation and spread are key elements in gauging fire behaviour potential in conifer forests. Crown fire initiation and spread models implemented in widely used fire behaviour simulation systems such as FARSITE and FlamMap require accurate spatially explicit estimation of canopy fuel complex characteristics. In the present study, we evaluated the potential use of very low-density airborne LiDAR (light detection and ranging) data (0.5 first returns m–2) – which is freely available for most of the Spanish territory – to estimate canopy fuel characteristics in Pinus radiata D. Don stands in north-western Spain. Regression analysis indicated strong relationships (R2=0.82–0.98) between LiDAR-derived metrics and field-based fuel estimates for stand height, canopy fuel load, and average and effective canopy base height Average and effective canopy bulk density (R2=0.59–0.70) were estimated indirectly from a set of previously modelled forest variables. The LiDAR-based models developed can be used to elaborate geo-referenced raster files to describe fuel characteristics. These files can be generated periodically, whenever new freely available airborne LiDAR data are released by the Spanish National Plan of Aerial Orthophotography, and can be used as inputs in fire behaviour simulation systems.
The fuel complex variables canopy bulk density and canopy base height are often used to predict crown fire initiation and spread. Direct measurement of these variables is impractical, and they are usually estimated indirectly by modelling. Recent advances in predicting crown fire behaviour require accurate estimates of the complete vertical distribution of canopy fuels. The objectives of the present study were to model the vertical profile of available canopy fuel in pine stands by using data from the Spanish national forest inventory plus low-density airborne laser scanning (ALS) metrics. In a first step, the vertical distribution of the canopy fuel load was modelled using the Weibull probability density function. In a second step, two different systems of models were fitted to estimate the canopy variables defining the vertical distributions; the first system related these variables to stand variables obtained in a field inventory, and the second system related the canopy variables to airborne laser scanning metrics. The models of each system were fitted simultaneously to compensate the effects of the inherent cross-model correlation between the canopy variables. Heteroscedasticity was also analyzed, but no correction in the fitting process was necessary. The estimated canopy fuel load profiles from field variables explained 84% and 86% of the variation in canopy fuel load for maritime pine and radiata pine respectively; whereas the estimated canopy fuel load profiles from ALS metrics explained 52% and 49% of the variation for the same species. The proposed models can be used to assess the effectiveness of different forest management alternatives for reducing crown fire hazard.
a b s t r a c tLight detection and ranging (LiDAR) has become an important tool in forestry. LiDAR-derived models are mostly developed by means of multiple linear regression (MLR) after stepwise selection of predictors. An increasing interest in machine learning and evolutionary computation has recently arisen to improve regression use in LiDAR data processing. Although evolutionary machine learning has already proven to be suitable for regression, evolutionary computation may also be applied to improve parametric models such as MLR. This paper provides a hybrid approach based on joint use of MLR and a novel genetic algorithm for the estimation of the main forest stand variables. We show a comparison between our genetic approach and other common methods of selecting predictors. The results obtained from several LiDAR datasets with different pulse densities in two areas of the Iberian Peninsula indicate that genetic algorithms perform better than the other methods statistically. Preliminary studies suggest that a lack of parametric conditions in field data and possible misuse of parametric tests may be the main reasons for the better performance of the genetic algorithm. This research confirms the findings of previous studies that outline the importance of evolutionary computation in the context of LiDAR analisys of forest data, especially when the size of fieldwork datatasets is reduced.
scite is a Brooklyn-based startup that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.