Search

Photogrammetric point clouds obtained with unmanned aircraft systems (UAS) have emerged as an alternative source of remotely sensed data for small area forest management inventories (FMI). Nonetheless, it is often overlooked that small area FMI require considerable field data in addition to UAS data, to support the modelling of forest attributes. In this study, we propose a method whereby tree volumes by species are predicted with photogrammetric UAS data and Sentinel-2 images, using models fitted with airborne laser scanning data. The study area is in a managed boreal forest area in Eastern Finland. First, we predicted total volume with UAS point cloud metrics using a prior regression model fitted in another area with ALS data. Tree species proportions were then predicted by nearest neighbor (-NN) imputation based on bi-seasonal Sentinel-2 images without measuring new field plot data. Species-specific volumes were then obtained by multiplying the total volume by species proportions. The relative root mean square error (RMSE) values for total and species-specific volume predictions at the validation plot level (30 m à 30 m) were 9.0%, and 33.4â62.6%, respectively. Our approach appears promising for species-specific small area FMI in Finland and in comparable forest conditions in which suitable field plots are available.kk

This study examines the alternatives to include crown base height (CBH) predictions in operational forest inventories based on airborne laser scanning (ALS) data. We studied 265 field sample plots in a strongly pine-dominated area in northeastern Finland. The CBH prediction alternatives used area-based metrics of sparse ALS data to produce this attribute by means of: 1) Tree-level imputation based on the -nearest neighbor (-nn) method and full field-measured tree lists including CBH observations as reference data; 2) Tree-level mixed-effects model (LME) prediction based on tree diameter (DBH) and height and ALS metrics as predictors of the models; 3) Plot-level prediction based on analyzing the computational geometry and topology of the ALS point clouds; and 4) Plot-level regression analysis using average CBH observations of the plots for model fitting. The results showed that all of the methods predicted CBH with an accuracy of 1â1.5 m. The plot-level regression model was the most accurate alternative, although alternatives producing tree-level information may be more interesting for inventories aiming at forest management planning. For this purpose, -nn approach is promising and it only requires that field measurements of CBH is added to the tree lists used as reference data. Alternatively, the LME-approach produced good results especially in the case of dominant trees.kkk

This study examined the suitability of airborne laser scanner (ALS) data collected under leaf-off conditions in a forest inventory, in which deciduous and coniferous trees need to be separated. All analyses were carried out with leaf-on and leaf-off ALS data collected from the same study area. Additionally, aerial photographs were utilized in the Nearest Neighbor (NN) imputations. An area-based approach was used in this study. Regression estimates of plot volume were more accurate in the case of leaf-off than leaf-on data. In addition, regression models were more accurate in coniferous plots than in deciduous plots. The results of applying leaf-on models with leaf-off data, and vice versa, indicate that leaf-on and leaf-off data should not be combined since this causes serious bias. The total volume and volume by coniferous and deciduous trees was estimated by the NN imputation. In terms of total volume, leaf-off data provided more accurate estimates than leaf-on data. In addition, leaf-off data discriminated between coniferous and deciduous trees, even without the use of aerial photographs. Accurate results were also obtained when leaf-off ALS data were used to classify sample plots into deciduous and coniferous dominated plots. The results indicate that the area-based method and ALS data collected under leaf-off conditions are suitable for forest inventory in which deciduous and coniferous trees need to be distinguished.

Tree species identification constitutes a bottleneck in remote sensing-based forest inventory. In passive images the differentiating features overlap and bidirectional reflectance hampers analysis. Airborne LiDAR provides radiometric and geometric information. We examined the single-trees-level response of two LiDAR sensors in over 13 000 forest trees in southern Finland. We focused on the commercially important species. Our aims were to 1) explore the relevant LiDAR features and study their dependencies on stand and tree variables, 2) examine two sensors and their fusion, 3) quantify the gain from intensity normalizations, 4) examine the importance of the size of the training set, and 5) determine the effects of stand age and site fertility. A set of 570 semiurban broad-leaved trees and exotic conifers was analyzed to 6) examine the LiDAR signal in the economically less important species. An accuracy of 88 90% was achieved in the classification of Scots pine, Norway spruce, and birch, using intensity variables. Spruce and birch showed the highest levels of confusion. Downsizing the training set from 30% to 2.5% of all trees had only a marginal effect on the performance of classifiers. The intensity features were dependent on the absolute and relative sizes of trees, especially for birch. The results suggest that leaf size, orientation, and foliage density affect the intensity, which is thus not affected by reflectance only. Some of the ecologically important species in Finland may be separable, since they gave rise to high intensity values. Comparison of the sensors implies that performance of the intensity data for species classification varies between sensors for reasons that remained uncertain. Both range and gain receiver normalization improved species classification. Weighting of the intensity values improved the fusion of two LiDAR datasets.

Airborne laser scanning (ALS) data have become the most accurate remote sensing technology for forest inventories. When planning new inventories the costs of fieldwork could be reduced if datasets of old inventory areas are effectively reused in the new area. The aim of this study was to apply mixed estimation using a combination of existing and new field datasets in area-based approach. Additionally, combining datasets with mixed estimation was compared with constructing new local models with smaller datasets. The two forest study areas were in Juuka and Matalansalo, which are located about 120 km apart in eastern Finland. ALS-based regression models were constructed using datasets of Matalansalo (472 reference plots) and Juuka (10â212 reference plots). Models were developed for the basal area median tree diameter and height, mean tree height, stem number, basal area and volume. The work was based on a simulation approach which involved five methods for approximating the regression coefficients. The first method merged the datasets using ordinary least squares (OLS) regression models, whereas the second and third methods combined datasets using mixed estimation on different weighting principles, and the final two estimated local models with predetermined and new independent variables. The results indicate that mixed estimation can improve the accuracy of derived stand variables compared with basic OLS models. Additionally, a sample of 40â50 plots was enough to build local models for basal area and volume and produce at least the equal accuracy of results than any other methods in this study.

The development of airborne laser scanning (ALS) during last ten years has provided new possibilities for accurate description of the living tree stock. The forest inventory applications of ALS data include both tree and area-based plot level approaches. The main goal of such applications has usually been to estimate accurate information on timber quantities. Prediction of timber quality has not been focused to the same extent. Thus, in this study we consider here the prediction of both basic tree attributes (tree diameter, height and volume) and characteristics describing tree quality more closely (crown height, height of the lowest dead branch and sawlog proportion of tree volume) by means of high resolution ALS data. The tree species considered is Scots pine (Pinus sylvestris), and the field data originate from 14 sample plots located in the Koli National Park in North Karelia, eastern Finland. The material comprises 133 trees, and size and quality variables of these trees were modeled using a large number of potential independent variables calculated from the ALS data. These variables included both individual tree recognition and area-based characteristics. Models for the dependent tree characteristics to be considered were then constructed using either the non-parametric k-MSN method or a parametric set of models constructed simultaneously by the Seemingly Unrelated Regression (SUR) approach. The results indicate that the k-MSN method can provide more accurate tree-level estimates than SUR models. The k-MSN estimates were in fact highly accurate in general, the RMSE being less than 10% except in the case of tree volume and height of the lowest dead branch.

The low-density airborne laser scanning (ALS) data based estimation methods have been shown to produce accurate estimates of mean forest characteristics and diameter distributions, according to several studies. The used estimation methods have been based on the laser canopy height distribution approach, where various laser pulse height distribution -derived predictors are related to the stand characteristics of interest. This approach requires very delicate selection methods for selecting the suitable predictor variables. In this study, we introduce a new nearest neighbor search method that requires no complicated selection algorithm for choosing the predictor variables and can be utilized in multipurpose situations. The proposed search method is based on Minkowski distances between the distributions extracted from low density ALS data and aerial photographs. Apart from the introduction of a new search method, the aims of this study were: 1) to produce accurate species-specific diameter distributions and 2) to estimate factual saw log recovery, using the estimated height-diameter distributions and a stem data bank. The results indicate that the proposed method is suitable for producing species-specific diameter distributions and volumes at the stand level. However, it is proposed, that the utilization of more extensive and locally emphasized reference data and auxiliary variables could yield more accurate saw log recoveries.

The harvesting of energy wood from young stands is increasing as the demand for renewable wood fuel is growing. Energy wood consists of stems, tree tops, branches and needles, depending on the size of the trees and the logging method used. The current forest inventory and planning systems used in private forests in Finland do not produce estimates of energy wood components. In stands typical for energy wood harvesting, a large share of energy wood consists of trees smaller than the minimum size for pulpwood. In this study, energy wood was included into the calculation system of compartment inventory, and a procedure for simulating the thinning treatments in young stands was developed. The results for six inventory alternatives and prediction of energy wood were compared with the use of inventory material from 37 young stands that have plenty of energy wood. The measurement of additional stand characteristics and the use of a calibration estimation method was tested, as well as the use of plot-level inventory data instead of stand level data. The results showed that the measurement of the number of trees per hectare, in addition to stand basal area and mean diameter, improved the energy wood estimates. The additional minimum and maximum diameters improved the precision of the estimates, but did not affect bias. The removal estimates were more precise when plot-level data was used, rather than stand-level data. The removal estimates were higher with plot-level data. The results suggest that, in heterogeneous young stands, plot by plot prediction would give more accurate removal estimates than the calculation of a corresponding prediction at the stand-level.

Accurate field measurement of the forest canopy cover is too laborious to be used in extensive forest inventories. A possible alternative to the separate canopy cover measurements is to utilize the correlations between the percent canopy cover and easier-to-measure forest variables, especially the basal area. A fairly new analysis technique, the beta regression, is specially designed for modelling percentages. As an extension to the generalized linear models, the beta regression takes into account the distribution of the model residuals, and uses a logistic link function to ensure logical predictions. In this study, the beta regression method was found to perform well in conifer dominated study area located in central Finland. The same model shape, with basal area, tree height and an additional predictor (Scots pine: site fertility, Norway spruce: percentage of hardwoods) as independent variables, produced good results for both pine and spruce dominated sites. The models had reasonably high pseudo R-squared values (pine: 0.91, spruce: 0.87) and low standard errors (pine: 6.3%, spruce: 5.9%) for the fitting data, and also performed well in a cross validation test. The models were also tested on separate test plots located in a different geographical area, where the prediction errors were slightly larger (pine: 8.8%, spruce: 7.4%). In pine plots, the model fit was further improved by introducing additional predictors such as stand age and density. This improved also the performance of the models in the cross validation test, but weakened the results for the external data set. Our results indicated that the beta regression method offers a noteworthy alternative to separate canopy cover measurements, especially if time is limited and the models can be applied in the same region where the modelling data were collected.

In percentile method, percentiles of the diameter distribution are predicted with a system of models. The continuous empirical diameter distribution function is then obtained by interpolating between the predicted values of percentiles. In Finland, the distribution is typically modelled as a basal-area weighted distribution, which is transformed to a traditional density function for applications. In earlier studies it has been noted that when calculated from the basal-area weighted diameter distribution, the density function is decreasing in most stands, especially for Norway spruce. This behaviour is not supported by the data. In this paper, we investigate the reasons for the unsatisfactory performance and present possible solutions for the problem. Besides the predicted percentiles, the problems are due to implicit assumptions of diameter distribution in the system. The effect of these assumptions can be somewhat lessened with simple ad-hoc methods, like increasing new percentiles to the system. This approach does not, however, utilize all the available information in the estimation, namely the analytical relationships between basal area, stem number and diameter. Accounting for these, gives further possibilities for improving the results. The results show, however, that in order to achieve further improvements, it would be recommendable to make the implicit assumptions more realistic. Furthermore, height variation within stands seems to have an important contribution to the uncertainty of some forest characteristics, especially in the case of sawnwood volume.

This study deals with the prediction of the basal area diameter distribution of a stand without using a complete sample of diameters from the target stand. Traditionally, this problem has been solved by either the parameter recovery method or the parameter prediction method. This study uses the parameter prediction method and the percentile based diameter distribution with a recent development that makes it possible to improve these predictions by using sample order statistics. A sample order statistic is a tree whose diameter and rank at the plot are known, and is referred to in this paper as a quantile tree. This study tested 13 different strategies for selection of the quantile trees from among the trees of horizontal point sample plots, and compared them with respect to RMSE and the bias of four criterion variables in a dataset of 512 stands. The sample minimum was found to be the most promising alternative with respect to RMSE, even though it introduced a rather large amount of bias in the criterion variables. Other good and less biased alternatives are the second and third smallest trees and the tree closest to the plot centre. The use of minimum is recommended for practical inventories because its rank is probably easiest to determine correctly in the field.

Abundance and species number of the tree and shrub vegetation in different canopy layers were analysed according to site quality class and drainage succession phase on permanent sample plots on spruce mires (n = 268) and pine mires (n = 628) in the Finnish National Forest Inventory in 1995. The abundances based on the crown coverage were compared with the abundances based on the parallel basal area of the tree stand. The canopy coverages and species number for peatland forests were also compared with those for mineral soil forests on the permanent sample plots (n = 1725) in 1995. In general, effective temperature sum correlated positively, although not very strongly, with the coverages and species number in most of the canopy layers, as well as with the mean range of the diameter distribution. The effects of both site quality class and drainage phase were stronger on pine mires than on spruce mires, most probably due to the longer fertility gradient and large potential free growing space in the former group. On pine mires, drainage increased the abundances and species number in the different canopy layers, as well as the structural inequality of the tree stands. On spruce mires, the increase was principally allocated to the abundances of the dominant and intermediate tree layers. The correlations between the total crown coverage of the tree layers and stand basal area were r = 0.45 for spruce mires and r = 0.70 for pine mires. Compared to mineral soil forests, in addition to having a higher abundance of Betula pubescens, the dominant layer was not as pronounced in peatland forests. On spruce mires, the coverage of the shrub layer on mesotrophic and meso-oligotrophic sites was higher than that in mineral soil forests. The average species number in different canopy layers did not differ significantly between spruce mires and mineral soil forests in corresponding site quality classes. On pine mires, the species number was generally lower (except for the mesotrophic sites) than that in corresponding mineral soil forests.

Several models and/or several variable combinations could be used to predict the diameter distribution of a stand. Typically, a fixed model and a fixed variable combination is used in all conditions. The calibration procedure, however, makes it possible to choose the measurement combination from among many possibilities, although the model used is fixed. In this study, the usefulness of utilizing additional stand characteristics for calibrating the predicted diameter distribution is examined. Nine measurement strategies were tested in predicting the total stand volume, sawlog volume and pulpwood volume. The observed errors of these variables under each strategy were modeled as a function of basal area, basal area median diameter and number of stems. The models were estimated in three steps. First, an Ordinary Least Squares (OLS) model was fitted to the observed errors. Then, a variance function was estimated using the OLS residuals. Finally, a weighted Seemingly Unrelated Regression (SUR) analysis was used to model the observed errors, using the estimated variance functions as weights. The estimated models can be used to anticipate the precision and accuracy of predicted volume characteristics for each stand with different variable combinations and, consequently, to choose the best measurement combination in different stands.

The purpose of this study was to examine the use of non-parametric methods in estimating tree level growth models. In non-parametric methods the growth of a tree is predicted as a weighted average of the values of neighbouring observations. The selection of the nearest neighbours is based on the differences between tree and stand level characteristics of the target tree and the neighbours. The data for the models were collected from the areas owned by Kuusamo Common Forest in Northeast Finland. The whole data consisted of 4051 tally trees and 1308 Scots pines (Pinus sylvestris L.) and 367 Norway spruces (Picea abies Karst.). Models for 5-year diameter growth and bark thickness at the end of the growing period were constructed with two different non-parametric methods: the k-nearest neighbour regression and k-Most Similar Neighbour method. Diameter at breast height, tree height, mean age of the stand and basal area of the trees larger than the subject tree were found to predict the diameter growth most accurately. The non-parametric methods were compared to traditional regression growth models and were found to be quite competitive and reliable growth estimators.

The aim of the study was to develop a yield prediction model using the non-parametric Most Similar Neighbour (MSN) reference method. The model is constructed on stand level but it contains information also on tree level. A 10-year projection period was used for the analysis of stand growth. First, the canonical correlation matrix was calculated for the whole study material using stand volumes at the beginning and at the end of the growth period as independent variables and stand characteristics as dependent variable. Secondly, similar neighbour estimates were searched from the data categories reclassified according to thinnings. Due to this, it was possible to search for growth and yield series which is as accurate as possible both at the beginning and at the end of the growth period. The reliability of the MSN volume predictions was compared to the volumes predicted with the simultaneous yield model. The MSN approach was observed to be more reliable volume predictor than the traditional stand level yield prediction model both in thinned and unthinned stands.

Information about diameter distribution is used for predicting stand total volume, timber volume and stand growth for forest management planning. Often, the diameter distribution is obtained by predicting the parameters of some probability density function, using means and sums of tree characters as predictors. However, the results have not always been satisfactory: the predicted distributions practically always have a similar shape. Also, multimodal distributions cannot be obtained. However, diameter distribution can also be predicted using distribution-free methods. In the percentile method, the diameters at certain percentiles of the distribution are predicted with models. The empirical diameter distribution function is then obtained by interpolating between the predicted diameters. In this paper, models for diameters at 12 percentiles of stand basal area are presented for Scots pine, Norway spruce and birch species. Two sets of models are estimated: a set with and one without number of stems as a predictor. Including the number of stems as a predictor improved the volume and saw timber volume estimates for all species, but the improvements were especially high for number of stems estimates obtained from the predicted distribution. The use of number of stems as predictor in models is based on the possibility of including this characteristic to measured stand variables.

Diameter distribution is used in most forest management planning packages for predicting stand volume, timber volume and stand growth. The prediction of diameter distribution can be based on parametric distribution functions, distribution-free parametric prediction methods or purely non-parametric methods. In the first case, the distribution is obtained by predicting the parameters of some probability density function. In a distribution-free percentile method, the diameters at certain percentiles of the distribution are predicted with models. In non-parametric methods, the predicted distribution is a linear combination of similar measured stands. In this study, the percentile based diameter distribution is compared to the results obtained with the Weibull method in four independent data sets. In the case of Scots pine, the other methods are also compared to k-nearest neighbour method. The comparison was made with respect to the accuracy of predicted stand volume, saw timber volume and number of stems. The predicted percentile and Weibull distributions were calibrated using number of stems measured from the stand. The information of minimum and maximum diameters were also used, for re-scaling the percentile based distribution or for parameter recovery of Weibull parameters. The accuracy of the predicted stand characteristics were also compared for calibrated distributions. The most reliable results were obtained using the percentile method with the model set including number of stems as a predictor. Calibration improved the results in most cases. However, using the minimum and maximum diameters for parameter recovery proved to be inefficient.

The purpose of this study was to compare the Weibull distributions estimated for the entire growing stock of a stand and separately for Scots pine ( L.) and Norway spruce ( (L.) H. Karst.) in describing the basal area diameter distributions in mixed stands. The material for this study was obtained by measuring 553 stands located in eastern Finland. The parameters of the Weibull distribution were estimated using the method of maximum likelihood. The models for these parameters were derived using regression analysis. Also, some parameter models from previous studies were compared with the measured distribution. The obtained distributions were compared using the diameter sums of the entire growing stock, diameter sums by tree species and of the sawtimber part of the growing stock. The results showed that far more accurate results were obtained when the distributions were formed using parameter models separately for the different tree species than when using parameter models for the entire growing stock. This was already true when considering the entire growing stock of the stand and especially when the results were examined by tree species. When the models for the entire growing stock were applied by tree species in relation to basal areas, the results obtained were overestimates for Norway spruce and underestimates for Scots pine. The models from earlier studies, where parameter models were estimated separately for tree species from the National Forest Inventory data, showed good fits also in regard to the data of this study.Pinus sylvestrisPicea abies

Comparisons were made between artificially and naturally regenerated stands in the south-eastern part of North Karelia, Finland, and naturally regenerated stands in the western parts of the Republic of Karelia, Russian Federation. The effect of soil fertility and silvicultural operations on the stand structure was also investigated. The results of the study show clearly that when forests are artificially regenerated the stand structure includes less variation when compared with the stands naturally regenerated. Differences between the regeneration methods are clearer the more fertile the forest site is. Within the regeneration method there is also a clear trend in stand structure, with the variation decreasing the poorer the site. The effect of silvicultural operations, i.e. the cleaning of the sapling stand, has disappeared by the time of first thinning, although it appears to have a permanent effect on the dynamics of the tree species within a stand. The variation of the stand structure can be regarded as an essential factor for the potential biodiversity of the stand also at its young vegetation succession stage. This capacity for maintaining the forest biodiversity, developed at the young vegetation succession stage, becomes increasingly important in subsequent vegetation succession stages. Natural regeneration provides improved possibilities for the operations preserving forest biodiversity, as it generates more dense stands with a wider variation in stand structure, compared to artificial regeneration.

Correlation functions of the mean volume, land use class and soil class were estimated using the data of the Finnish National Forest Inventory. Estimated functions were used for approximating the standard error of e.g. the mean volume of a cluster of plots. Standard error estimates can be used for comparing different inventory designs. The PDF includes an abstract in Finnish.

The paper continues an earlier study by Kilkki and Päivinen concerning the use of the Weibull function in modelling the diameter distribution. The data consists of spruces ( (L.) H. Karst.) measured on angle count sample points of the National Forest Inventory of Finland. First, maximum likelihood estimation method was used to derive the Weibull parameters. Then, regression models to predict the values of these parameters with stand characteristics were calculated. Several methods to describe the Weibull function by a tree sample were tested. It is more efficient to sample the trees at equal frequency intervals than at equal diameter intervals. It also pays to take separate samples for pulpwood and saw timber.Picea abies

Multispectral light detection and ranging (LiDAR) instruments, such as Optech Titan, record intensities at multiple wavelengths and these intensities can be used for tree species prediction in the same way as multispectral image data. In this paper, our main objective was to compare the accuracy of tree species prediction in a boreal forest area using multispectral LiDAR, the 1064-nm wavelength channel (“unispectral LiDAR”), and unispectral LiDAR with auxiliary aerial image data. We also evaluated the effect of the widely used intensity range correction method. We classified the main tree species of field plots using linear discriminant analysis (LDA) and predicted the species-specific volume proportions (%) for Scots pine (Pinus sylvestris), Norway spruce (Picea abies), and broadleaved trees using the $k$ -nearest neighbor imputation. The effect of intensity correction on prediction errors for the dominant tree species was evaluated using optimal parameters derived from: 1) minimal intensity difference between flight lines; 2) parameters suggested by theory; and 3) uncorrected data. Although the range correction increased the classification accuracy slightly, it was observed to be ambiguous, and not consistent with theory for canopy echoes. Regardless, the intensity values were useful for the prediction of dominant tree species and species’ volume proportions. The results for the dominant tree species classification using multispectral LiDAR [overall accuracy (OA) 88.2%, kappa 0.79] were comparable to the use of unispectral LiDAR and aerial images (OA 89.1%, kappa 0.81). We conclude that the multispectral LiDAR may become a useful tool in operational species-specific forest inventories. [ABSTRACT FROM AUTHOR]

While the analysis of airborne laser scanning (ALS) data often provides reliable estimates for certain forest stand attributes-such as total volume or basal area-there is still room for improvement, especially in estimating species-specific attributes. Moreover, while the information on the estimate uncertainty would be useful in various economic and environmental analyses on forests, a computationally feasible framework for uncertainty quantifying in ALS is still missing. In this paper, the species-specific stand attribute estimation and uncertainty quantification (UQ) is approached using Gaussian process regression (GPR), which is a nonlinear and nonparametric machine learning method. Multiple species-specific stand attributes are estimated simultaneously: tree height, stem diameter, stem number, basal area, and stem volume. The cross-validation results show that GPR yields on average an improvement of 4.6% in estimate root mean square error over a state-of-the-art $k$ -nearest neighbors (kNNs) implementation, negligible bias and well performing UQ (credible intervals), while being computationally fast. The performance advantage over kNN and the feasibility of credible intervals persists even when smaller training sets are used. [ABSTRACT FROM AUTHOR]

In this paper, we propose an approach to quantify the plot-level uncertainty in species-specific growing stock volume estimated from airborne laser scanning data and aerial imagery. This is accomplished by adopting the framework of Bayesian inference in the area-based estimation of stock volume. The results show that the proposed approach performs well in quantifying the estimate uncertainty and produces optimal interval estimates for species-specific volumes when sufficient training data are available. Also the point estimate accuracy is competitive with current state-of-the-art methods. Furthermore, we demonstrate how the quantified uncertainties of the stand attributes can be utilized to determine the uncertainty in classification done using the estimated stand attributes. [ABSTRACT FROM AUTHOR]

We describe a novel method to improve the correspondence between field and airborne laser scanning (ALS) measurements in an area-based approach (ABA) forest inventory framework. An established practice in forest inventory is that trees with boles falling within a fixed border field measurement plot are considered “in” trees; yet their crowns may extend beyond the plot border. Likewise, a tree bole may fall outside of a plot, but its crown may extend into a plot. Typical ABA approaches do not recognize these discrepancies between the ALS data extracted for a given plot and the corresponding field measurements. In the proposed solution, enhanced ABA (EABA), predicted tree positions, and crown shapes are used to adjust plot and grid cell boundaries and how ALS metrics are computed. The idea is to append crowns of “in” trees to a plot and cut down “out” trees, then EABA continues in the traditional fashion as ABA. The EABA method requires higher density ALS data than ABA because improvement is obtained by means of detecting individual trees. When compared to typical ABA, the proposed EABA method decreased the error rate (RMSE) of stem volume prediction from 23.16% to 19.11% with $127\;{{\rm m}^2}$ plots and from 19.08% to 16.95% with $254\;{{\rm m}^2}$ plots. The greatest improvements were obtained for plots with the largest residuals. [ABSTRACT FROM PUBLISHER]

In this paper, we consider a computational method for detecting trees on the basis of airborne laser scanning (ALS) data. In the approach, locations, heights, and crown shapes of trees are tracked automatically by fitting multiple 3-D crown height models to ALS data of a field plot. The estimates are computed with an iterative reconstruction method based on Bayesian inversion paradigm. The formulation allows for utilizing prior information on tree shapes in the estimation. Here, the prior models are written based on field measurement data and allometric models for tree shapes. The feasibility of the approach is tested with ALS and field data from a managed boreal forest. The algorithm found 70.2% of the trees in the area, which is a clear improvement compared to a usual 2.5D crown delineation approach (53.1% of the trees detected). [ABSTRACT FROM AUTHOR]

Parametric and nonparametric modeling methods have been widely used for the estimation of forest attributes from airborne laser-scanning data and aerial photographs. However, the methods adopted suffered from complex remote-sensed data structures involving high dimensions, nonlinear relationships, different statistical distributions, and outliers. In this context, artificial neural networks (ANNs) are of interest as they have many clear benefits over conventional modeling methods and could then enhance the accuracy of current forest-inventory methods. This paper examines the ability of common ANN modeling techniques for the prediction of species-specific forest attributes, as exemplified here with the prediction stem volumes (cubic meters per hectare) at the field plot and forest stand levels. Three modeling methods were evaluated, namely, the multilayer perceptron (MLP), support vector regression (SVR), and self-organizing map, and intercompared with the corresponding nonparametric k most similar neighbor method using cross-validated statistical performance indexes. To decrease the number of model-input variables, a multiobjective input-selection method based on genetic algorithm is adopted. The numerical results obtained in the study suggest that ANNs are appropriate and accurate methods for the assessment of species-specific forest attributes, which can be used as alternatives to multivariate linear regression and nonparametric nearest neighbor models. Among the ANN models, SVR and MLP provide the best choices for prediction purposes as they yielded high prediction accuracies for species-specific tree volumes throughout. [ABSTRACT FROM AUTHOR]

Abstract: Diameter distribution is essential for calculating stem volume and timber assortments of forest stands. A new method was proposed in this study to improve the estimation of stem volume and timber assortments, by means of combining the Area-based approach (ABA) and individual tree detection (ITD), the two main approaches to deriving forest attributes from airborne laser scanning (ALS) data. Two methods, replacement, and histogram matching were employed to calibrate ABA-derived diameter distributions with ITD-derived diameter estimates at plot level. The results showed that more accurate estimates were obtained when calibrations were applied. In view of the highest accuracy between ABA and ITD, calibrated diameter distributions decreased its relative RMSE of the estimated entire growing stock, saw log and pulpwood fractions by 2.81%, 3.05% and 7.73% points at best, respectively. Calibration improved pulpwood fraction significantly, which contributed to the negligible bias of the estimated entire growing stock. [Copyright &y& Elsevier]

Errors in individual tree detection and delineation affect diameter distribution predictions based on crown attributes extracted from the detected trees. We develop a methodology for circumventing these problems. The method is based on matching cumulative distribution functions of field measured tree diameter distributions and crown radii distributions extracted from airborne laser scanning data through individual tree detection presented by Vauhkonen and Mehtätalo (2015). In this study, empirical distribution functions and a monotonic, nonlinear model curve are introduced. Tree crown radius distribution produced by individual tree detection is corrected by a method taking into account that all trees cannot be detected. The evaluation is based on the ability of the developed model sequence to predict quadratic mean diameter and total basal area. The studied data consists of 36 field plots in a typical boreal managed forest area in eastern Finland. The suggested enhancements to the model sequence produce improved results in most of the test cases. Most notably, in leave-one-out cross-validation experiments the modified models improve RMSE of basal area 13% in the full data and RMSE of quadratic mean diameter and basal area 69% and 11%, respectively, in pure pine plots. Better modeling of the crown radius distribution and improved matching between crown radii and stem diameters add the operational premises of the full distribution matching. [ABSTRACT FROM AUTHOR]