Large-scale Forest Inventories Research Articles

Monitoring spatial tree diversity indices using relascope sampling: Estimators, interactions and limitations

With ongoing climate change the monitoring of tree diversity has become very important for avoiding or at least decelerating the loss of biodiversity and for maintaining forest ecosystem resilience. As part of such monitoring, spatial indices of tree diversity which are calculated for individual trees often serve as surrogates for more direct measures of biodiversity. Mainly for its efficiency and ease of application, relascope sampling is a widespread method applied in forest inventory of many countries and thus has often been suggested as a data source for the monitoring of tree diversity. Since the interaction between sampling design and spatial diversity indices is not always clear to data analysts, we reviewed existing estimators and experimentally examined a new one, conducted extensive sampling simulations using different indices and estimators and additionally analysed the data from a large-scale forest inventory in Austria. We found that both forest structure and index algorithm greatly influence the sampling error. The largest source of sampling error was the index variance and contrary to our expectation not so much the bias due to spatial effects. For diversity indices related to distances, it has turned out to be best to apply estimators that include spatial edge correction methods. For all other indices an estimator performed better that included information on both the sample trees and their nearest neighbours, as it much reduced overall index variance. However, if possible the plus-sampling edge correction method should be applied.

Linking structure and species richness to support forest biodiversity monitoring at large scales

Key messageAuthors have analyzed the possible correlation between measurements/indicators of forest structure and species richness of many taxonomic or functional groups over three regions of Germany. Results show the potential to use structural attributes as a surrogate for species richness of most of the analyzed taxonomic and functional groups. This information can be transferred to large-scale forest inventories to support biodiversity monitoring.ContextWe are currently facing a dramatic loss in biodiversity worldwide and this initiated many monitoring programs aiming at documenting further trends. However, monitoring species diversity directly is very resource demanding, in particular in highly diverse forest ecosystems.AimsWe investigated whether variables applied in an index of stand structural diversity, which was developed based on forest attributes assessed in the German National Forest Inventory, can be calibrated against richness of forest-dwelling species within a wide range of taxonomic and functional groups.MethodsWe used information on forest structure and species richness that has been comprehensively assessed on 150 forest plots of the German biodiversity exploratories project, comprising a large range of management intensities in three regions. We tested, whether the forest structure index calculated for these forest plots well correlate with the number of species across 29 taxonomic and functional groups, assuming that the structural attributes applied in the index represent their habitat requirements.ResultsThe strength of correlations between the structural variables applied in the index and number of species within taxonomic or functional groups was highly variable. For some groups such as Aves, Formicidae or vascular plants, structural variables had a high explanatory power for species richness across forest types. Species richness in other taxonomic and functional groups (e.g., soil and root-associated fungi) was not explained by individual structural attributes of the index. Results indicate that some taxonomic and functional groups depend on a high structural diversity, whereas others seem to be insensitive to it or even prefer structurally poor stands.ConclusionTherefore, combinations of forest stands with different degrees of structural diversity most likely optimize taxonomic diversity at the landscape level. Our results can support biodiversity monitoring through quantification of forest structure in large-scale forest inventories. Changes in structural variables over inventory periods can indicate changes in habitat quality for individual taxonomic groups and thus points towards national forest inventories being an effective tool to detect unintended effects of changes in forest management on biodiversity.

Beech Bark Disease in an Unmanaged Temperate Forest: Patterns, Predictors, and Impacts on Ecosystem Function

Beech Bark Disease (BBD) is a devastating threat to American beech (Fagus grandifolia), spreading through eastern mixed deciduous forests of North America at unprecedented rates. Understanding how and why some beech trees escape severe BBD effects is important; however, the biotic and abiotic factors that underpin the progression of BBD within unmanaged forests at local scales are not well explored. We surveyed 651 individual beech trees ≥ 10 cm diameter at breast height (DBH) for BBD, in a 13.5-ha unmanaged forest dynamics plot in Ontario, Canada, where >46,000 trees have been identified to species, mapped, and DBH measured at ∼5-year intervals. For each beech tree, BBD severity was ranked on a 5-point severity index, which was then evaluated as a function of tree characteristics including DBH and relative growth rate (RGR). Most beech trees were at either the insect or fungal stage of BBD, with only 22% of beech trees being free of symptoms. Ordinal logistic regression analysis indicated both DBH and RGR were significant predictors of BBD severity. These models, along with both randomization and Moran’s Eigenvector Maps (MEM) analyses, indicated that DBH and RGR and their spatial variation accounted for ∼44.6% of BBD severity in trees. Our MEMs also indicated ∼4.2% of variation in BBD severity was associated with unmeasured spatial variables, which may reflect either the spread of BBD through our study site, or the influence of abiotic variables on BBD severity. At our site, BBD is responsible for at least ∼6.0 Mg C ha–1, or ∼6.5% of the average 92.5 Mg of aboveground biomass C ha–1, transitioning from the live to dead biomass pool. Our study enhances the understanding of the factors predicting the severity of a major forest pathogen in North American temperate forests, assists the integration of BBD severity risk into forest C budget models, and provides insight into how large-scale forest inventories can inform screening for pest or pathogen resistance in trees.

The Dynamics of Secondary Mangrove Forests in Bintuni Bay, West Papua after Harvested on the First 30-Year Rotation Cycle

Mangrove forests in Bintuni Bay, West Papua, Indonesia, have been managed for timber extraction since 1988 to produce wood chips using a 30-year rotation cycle. The first rotation cycle was completed, resulting in secondary mangrove forests with various stand ages (1–30 years). A large-scale forest inventory was conducted for all harvested blocks to recognize actual standing stock for the 2021–2030 management plan. A total of 434 quadrat plots (20 m x 20 m) covering 17.36 ha each were used to observe forest standing stock. The results present the dynamics of secondary mangrove forests after the initial rotation cycle. At the end of the first cycle (30-year-old stand), secondary mangrove forests provided a volume of 290.12 m3 ha-1 and potential extractable biomass of 203.03 ton ha-1 with mean diameter, basal area, and stem density of 16.91 cm, 29.18 m2 ha-1, and 1,370 stem ha-1, respectively. Annual increment of volume, biomass, and diameter were estimated to be 9.67 m3 ha-1 year-1, 6.77 ton ha-1 year-1, and 0.56 cm year-1, consecutively. The dynamics of secondary mangrove forests could provide significant information in the context of the management plan and implementation of the silviculture system to ensure the sustainability of utilization in managed mangrove forests. Keywords: Forest regeneration, Silviculture, Standing stock, Sustainable management, Timber utilization

Exploring recruitment dynamics of sugar maple and yellow birch saplings into merchantable stems following harvesting in the Acadian Forest Region of New Brunswick, Canada

The recruitment of saplings in forest stands into merchantable stems is a very complex process, thus making it challenging to understand and predict. The dynamics of recruitment in the Acadian Forest Region of New Brunswick are not well known or well documented. Our objective was to garner information from existing large-scale routine forest inventories regarding the different dynamics behind the recruitment from the sapling layer into the commercial tree size layer. We focused on the density and occurrence of sugar maple (Acer saccharum Marsh.) and yellow birch (Betula alleghaniensis Britt.) following harvesting, by looking at many factors on a wide range of spatial and temporal scales using models. The results suggest that the variation in density and probability of occurrence is best explained by the intensity of silvicultural treatment, by the merchantable stem density in each plot, and by the proportion of merchantable basal area of each group of species. The number of recruits of sugar maple and yellow birch stems tend be higher when time since last treatment increases, when mid to low levels of silvicultural treatment intensity were implemented, and within plots having intermediate levels of merchantable stem density. Lastly, our modeling efforts suggest that the probability of occurrence and density of recruitment of both species tend to increase while its share of merchantable basal area increases.

A Review of Tree Species Classification Based on Airborne LiDAR Data and Applied Classifiers

Remote sensing techniques, developed over the past four decades, have enabled large-scale forest inventory. Light Detection and Ranging (LiDAR), as an active remote sensing technology, allows for the acquisition of three-dimensional point clouds of scanned areas, as well as a range of features allowing for increased performance of object extraction and classification approaches. As many publications have shown, multiple LiDAR-derived metrics, with the assistance of classification algorithms, contribute to the high accuracy of tree species discrimination based on data obtained by laser scanning. The aim of this article is to review studies in the species classification literature which used data collected by Airborne Laser Scanning. We analyzed these studies to figure out the most efficient group of LiDAR-derived features in species discrimination. We also identified the most powerful classification algorithm, which maximizes the advantages of the derived metrics to increase species discrimination performance. We conclude that features extracted from full-waveform data lead to the highest overall accuracy. Radiometric features with height information are also promising, generating high species classification accuracies. Using random forest and support vector machine as classifiers gave the best species discrimination results in the reviewed publications.

The R Package forestinventory: Design-Based Global and Small Area Estimations for Multiphase Forest Inventories

Forest inventories provide reliable evidence-based information to assess the state and development of forests over time. They typically consist of a random sample of plot locations in the forest that are assessed individually by field crews. Due to the high costs of these terrestrial campaigns, remote sensing information available in high quantity and low costs is frequently incorporated in the estimation process in order to reduce inventory costs or improve estimation precision. With respect to this objective, the application of multiphase forest inventory methods (e.g., double- and triple-sampling regression estimators) has proved to be efficient. While these methods have been successfully applied in practice, the availability of open-source software has been rare if not non-existent. The R package forestinventory provides a comprehensive set of global and small area regression estimators for multiphase forest inventories under simple and cluster sampling. The implemented methods have been demonstrated in various scientific studies ranging from small to large scale forest inventories, and can be used for post-stratification, regression and regression within strata. This article gives an extensive review of the mathematical theory of this family of design-based estimators, puts them into a common framework of forest inventory scenarios and demonstrates their application in the R environment.

A trunk-based SLAM backend for smartphones with online SLAM in large-scale forest inventories

Reliable forest resource information is needed to assess the forest development status and design management plans for forest maintenance and conservation. The forest field sample inventory is a vital forest resource inventory method. Thus, forest inventory reliability depends on tree attribute estimation accuracy and the quantity and quality of field samples. Simultaneous localization and mapping (SLAM)-based mobile laser scanners (MLSs) are convenient inventory tools due to their mobility and global navigation satellite system (GNSS) signal independence. However, such scanners may be heavy, expensive and unable to verify results on-site. With the improved SLAM algorithm and chip computing capabilities, a smartphone can deploy an online SLAM system, which allows the smartphone to perform the relative positioning in forests without GNSS signals. Previous research studies have demonstrated this simple, portable, and economical device for estimating the tree position and diameter at breast height (DBH) through tree-by-tree measurements in real time. However, the device might face a challenge in large-scale forest inventories because the image-feature-based backend may not work well in forests that are not well constructed for traditional SLAM systems. In this paper, an online trunk-based backend was designed to accurately estimate tree position and correct pose drift in large-scale forest inventories in real time. Specifically, a trunk-based loop closure detection algorithm was designed for detecting whether an earlier observed tree is re-observed to provide nodes and constraints for tree position graph optimization; this algorithm uses the provided nodes and constraints to build and optimize the tree position graph and then correct the current pose based on the optimized globally consistent tree position graph. This new backend was integrated with the previous work as an executable program that can be deployed on a smartphone with an online RGB-D SLAM system. The method was tested in 5 field sample plots (32 × 32 m), and the reference tree positions were collected using terrestrial laser scanning (TLS) through multi-scan mode. The distance mean between the estimated and reference tree positions was 0.133 m when using our new backend, and it was 0.759 m when estimated with the RTAB-Map. The tree position estimates were unbiased and had root mean square errors (RMSEs) of less than 0.09 m in the x-axis, y-axis and z-axis directions when using the trunk-based backend. However, the estimates had biases of −0.125 m, −0.261 m and 0.262 m and RMSEs of more than 0.30 m in the x-axis, y-axis and z-axis directions without the new backend. The results showed that the designed trunk-based backend allows a smartphone with an online SLAM system to function as an accurate and efficient tool for large-scale forest inventories. However, the method was tested only in 32 × 32 m square field sample plots. More tests must be performed in larger plots in the future, although enough loop-closure constraints can theoretically guarantee the accuracy of the tree position graph and current pose.

Predicting and mapping site index in operational forest inventories using bitemporal airborne laser scanner data

Forest productivity reflects the wood production capacity of a given site and provides crucial information for forest management planning. The most widely accepted measure of forest productivity is site index (SI), defined as the average height of dominant trees at a given index age. In forest management inventories, SI is commonly interpreted manually from aerial images. While the use of airborne laser scanner (ALS) data has revolutionized operational practices for estimating many forest attributes relevant to forest management planning, practices for determining SI remain unchanged. The main objective of this study was to demonstrate a practical method for predicting and mapping SI in repeated ALS-based forest inventories. We used data acquired as part of three operational large-scale forest inventories in southeastern Norway. First, we identified areas in which forest growth had remained undisturbed since the initial inventory. We then regressed field predictions of SI against bitemporal ALS canopy metrics and used the regression models to predict SI for forest areas classified as undisturbed. The result was SI maps constructed with a spatial resolution of 15.81 m. User accuracies of class predictions of undisturbed forest in the three districts were 92%, 95% and 89%. Plot-level validation revealed root mean squared errors of SI predictions ranging from 1.72 to 2.84 m for Norway spruce, and 1.35 to 1.73 m for Scots pine. The method presented here can be used to map SI over large areas of forest automatically, depicting forest productivity at a much finer spatial resolution than what is common in operational inventories.

Tree species dynamics in Swiss forests as affected by site, stand and management: A retrospective analysis

Knowledge regarding tree species dynamics is essential to understand forest responses to the environment, and to evaluate management options in adapting forest ecosystems to future climates. As maintaining tree species diversity and promoting structural stand heterogeneity are among the strategic elements in adapting forest management to climate change, the monitoring tree diversity is an ongoing challenge. Large-scale forest inventories have been proposed as a suitable basis for forest diversity analysis on large spatial and temporal scales.We used Swiss forest inventory data (NFI) to analyse temporal changes in tree species richness on small plots from 1983 to 2006. For two size groups of trees (‘small’ trees with dbh from 12 to 35 cm from plots with 200 m2 area, and ‘large’ trees with dbh ≥ 36 cm from plots with 500 m2 area), we identified the number and the tree species appearing (‘gains’) or disappearing (‘losses’) from each plot during the study period, and related these changes to site, stand and management characteristics.We found that species richness change was size-dependent and varied largely due to regional differences in the past land-use history of the Swiss forests. ‘Gains’ of ‘small’ trees were higher in stands with diverse vertical structure, with less competitive pressure as well as in warm environments, whereas ‘gains’ of ‘large’ trees were mostly related to climate and were highest in warm and moderately moist habitats. ‘Losses’ in both tree-size groups were mainly promoted by management. Our analysis suggests high vulnerability of Picea abies and high competitiveness of Fagus sylvatica, and underlines the potential of Abies alba in forming future Swiss forests.Despite of the silvicultural paradigm to create more species rich forests, most silvicultural interventions decreased small-scale species richness. This calls for further studies on the effect of management on tree species diversity.

Effects of stand features on aboveground biomass and biomass conversion and expansion factors based on a Pinus sylvestris L. chronosequence in Western Poland

Although Scots pine (Pinus sylvestris L.) is one of the most economically important European timber trees, there is still insufficient data about biomass variability and its relationships with stand features. Therefore, we aimed: (1) to develop biomass models for different aboveground biomass components at tree and stand levels, as well as biomass conversion and expansion factors (BCEFs), (2) to assess the relationships between stand parameters and aboveground biomass and BCEFs and (3) to compare stand biomass obtained using BCEFs with models developed based on stand parameters (age, basal area, stand volume and mean height). Using a chronosequence (3–117 years old) of 120 plots within even-aged pure Scots pine stands and 791 sample trees, we prepared tree- and stand-level allometric equations and BCEFs for aboveground biomass determination. Using stand age, density, stand volume and mean height, we prepared a set of models for biomass and BCEFs. Our study indicated that stand biomass increased with increasing height, volume and age and with decreasing stand density during stand development. Stand-level models provided better accuracy than BCEF-based models. The best predictors of biomass were stand volume and mean height. We also confirmed highly dynamic increases in stand biomass and decreases in BCEFs in the youngest phase of stand growth and relative stabilization in later stages of Scots pine stand development. The models obtained may be used in large-scale forest biomass inventories and increase our knowledge of carbon sequestration in forest biomass.

Potential of remote sensing-based forest attribute models for harmonising large-scale forest inventories on regional level: a case study in Southwest Germany

Image-based 3D information can provide metrics for forest attribute modelling that are robust within the study region. This enables transferability of predictive models to other data sets of the same region without loss of accuracy. As a result, aerial images can potentially provide auxiliary data for supporting different large-scale forest inventories within a geographic region. Due to their high spatial coverage and acquisition frequencies, aerial imagery can provide auxiliary data to support large-scale forest inventories. If the methods are applicable to different data sets and forest inventory protocols, they will also facilitate harmonisation of forest inventory data. This study aims to investigate the level of transferability of such image-based methods. This is crucial for their applicability across different large-scale forest inventories. The investigation focusses on one geographic region. Three blocks of aerial images were used to generate models of forest canopies and extract 3D metrics. These were utilised for building timber volume models separately for each image block. The models were applied to the respective other blocks and the achieved accuracy of timber volume prediction was assessed. Additionally, 3D metric changes between blocks were also assessed. Some metrics were found more robust than others. Transferring models based on robust metrics achieved RMSE% between 38 and 45%, which is similar to the model accuracy. This indicates transferability of models within the study region without loss of accuracy, and there is potential for further improvement of model accuracy. Therefore, forest attribute models based on remote sensing have potential to support harmonisation of large-scale forest inventories within the study region.

Current status of habitat monitoring in the European Union according to Article 17 of the Habitats Directive, with an emphasis on habitat structure and functions and on Germany

Since the beginning of the 1990s, monitoring of habitats has been a widespread tool to record and assess changes in habitat quality, for example due to land use change. Thus, Article 11 of the Habitats Directive (92/43/EEC) requires, inter alia, monitoring of the conservation status of habitat types listed in Annex I of the Habitats Directive, carried out by the Member States of the European Union (EU). This monitoring provides the foundation for the National Reports on the measures implemented and their effectiveness (Art. 17 Habitats Directive), which Member States have to submit to the European Commission every six years. Based on these requirements, Member States have developed different monitoring programmes or have adapted previously existing monitoring schemes to include relevant aspects of the Habitats Directive. The parameter ‘structure and functions’ is a key parameter for the assessment of the conservation status of habitat types as it provides information on the quality of the habitats. A standardised questionnaire was developed and sent to the competent authorities of Member States to compare and analyse the assessment methods of the quality of habitat types. Responses were received from 13 of the 28 Member States, while it was possible to include another Member State in the analysis by evaluating appropriate literature. The analysis revealed very different approaches and progress amongst the Member States in the development and implementation of monitoring programmes tailored to the reporting obligations of Article 17 of the Habitats Directive. Some Member States established a special standardised monitoring programme for Article 11 of the Habitats Directive, while others used data from already existing programmes (e.g. habitat mapping, large-scale forest inventories, landscape monitoring). Most Member States responding to the questionnaire use monitoring based on samples but the data collection, sample sizes and level of statistical certainty differ considerably. The same applies to the aggregation of data and the methods for the assessment of the parameter ‘structure and functions’. In contrast to the assessment of conservation status as part of the reporting obligations according to Article 17 of the Habitats Directive, no standardised EU guidelines exist for monitoring. The present study discusses differences in the monitoring programmes and evaluates them with regard to the objectives of comparable assessments of conservation status of habitat types in the National Reports of Member States or at a biogeographical level.

Evaluating the Condition of Selectively Logged Production Forests in Myanmar: An Analysis Using Large-scale Forest Inventory Data for Yedashe Township

Evaluating the Condition of Selectively Logged Production Forests in Myanmar: An Analysis Using Large-scale Forest Inventory Data for Yedashe Township

Quantifying forest structural diversity based on large-scale inventory data: a new approach to support biodiversity monitoring

BackgroundThe importance of structurally diverse forests for the conservation of biodiversity and provision of a wide range of ecosystem services has been widely recognised. However, tools to quantify structural diversity of forests in an objective and quantitative way across many forest types and sites are still needed, for example to support biodiversity monitoring. The existing approaches to quantify forest structural diversity are based on small geographical regions or single forest types, typically using only small data sets.ResultsHere we developed an index of structural diversity based on National Forest Inventory (NFI) data of Baden-Württemberg, Germany, a state with 1.3 million ha of diverse forest types in different ownerships. Based on a literature review, 11 aspects of structural diversity were identified a priori as crucially important to describe structural diversity. An initial comprehensive list of 52 variables derived from National Forest Inventory (NFI) data related to structural diversity was reduced by applying five selection criteria to arrive at one variable for each aspect of structural diversity. These variables comprise 1) quadratic mean diameter at breast height (DBH), 2) standard deviation of DBH, 3) standard deviation of stand height, 4) number of decay classes, 5) bark-diversity index, 6) trees with DBH ≥ 40 cm, 7) diversity of flowering and fructification, 8) average mean diameter of downed deadwood, 9) mean DBH of standing deadwood, 10) tree species richness and 11) tree species richness in the regeneration layer. These variables were combined into a simple, additive index to quantify the level of structural diversity, which assumes values between 0 and 1. We applied this index in an exemplary way to broad forest categories and ownerships to assess its feasibility to analyse structural diversity in large-scale forest inventories.ConclusionsThe forest structure index presented here can be derived in a similar way from standard inventory variables for most other large-scale forest inventories to provide important information about biodiversity relevant forest conditions and thus provide an evidence-base for forest management and planning as well as reporting.

Delineation of individual deciduous trees in plantations with low-density LiDAR data

ABSTRACTDelineation of individual deciduous trees with Light Detection and Ranging (LiDAR) data has long been sought for accurate forest inventory in temperate forests. Previous attempts mainly focused on high-density LiDAR data to obtain reliable delineation results, which may have limited applications due to the high cost and low availability of such data. Here, the feasibility of individual deciduous tree delineation with low-density LiDAR data was examined using a point-density-based algorithm. First a high-resolution point density model (PDM) was developed from low-density LiDAR point cloud to locate individual trees through the horizontal spatial distribution of LiDAR points. Then, individual tree crowns and associated attributes were delineated with a 2D marker-controlled watershed segmentation. Additionally, the PDM-based approach was compared with a conventional canopy height model (CHM) based delineation. The results demonstrated that the PDM-based approach produced an 89% detection accuracy to identify deciduous trees in our study area. The tree attributes derived from the PDM-based algorithm explained 81% and 83% of tree height and crown width variations of forest stands, respectively. The conventional CHM-based tree attributes, on the other hand, could explain only 71% and 66% of tree height and crown width, respectively. Our results suggest that the application of the PDM-based individual tree identification in deciduous forests with low-density LiDAR data is feasible and has relatively high accuracy to predict tree height and crown width, which are highly desired in large-scale forest inventory and analysis.

Assessment of sampling strategies utilizing auxiliary information in large-scale forest inventory

The National Forest Inventory of Finland (NFI) produces national- and regional-level statistics for sustainability assessment and strategical-level decision making. So far, the regional-level statistics are based on a systematic sampling design with geographical stratification. Auxiliary information such as remote sensing is not used for design or estimation at the regional level, but it is used at the small-area level, i.e., for municipality-level results. To improve the cost efficiency of the NFI, possibilities for using auxiliary data in both the design and estimation are of interest. We assessed the improvements obtainable by using an interpreted satellite image — the multisource NFI result from a previous NFI — as auxiliary information in the design phase. The results show that even though the multisource NFI map is not very accurate, significant improvements in efficiency can be obtained by using either the local pivotal method (LPM) or stratification. LPM improves efficiency by matching the sample distribution to population distribution. These results encourage us to further investigate (i) what would be the improvement with more accurate auxiliary information, for example, laser scanning data, and (ii) how LPM fits in a real-life situation where part of the plots are permanent and it would be used to select only the temporary plots.

Development of a System of Compatible Individual Tree Diameter and Aboveground Biomass Prediction Models Using Error-In-Variable Regression and Airborne LiDAR Data

Estimating individual tree diameters at breast height (DBH) from delineated crowns and tree heights on the basis of airborne light detection and ranging (LiDAR) data provides a good opportunity for large-scale forest inventory. Generally, ground-based measurements are more accurate, but LiDAR data and derived DBH values can be obtained over larger areas for a relatively smaller cost if a right procedure is developed. A nonlinear least squares (NLS) regression is not an appropriate approach to predict the aboveground biomass (AGB) of individual trees from the estimated DBH because both the response variable and the regressor are subject to measurement errors. In this study, a system of compatible individual tree DBH and AGB error-in-variable models was developed using error-in-variable regression techniques based on both airborne LiDAR and field-measured datasets of individual Picea crassifolia Kom. trees, collected in northwestern China. Two parameter estimation algorithms, i.e., the two-stage error-in-variable model (TSEM) and the nonlinear seemingly unrelated regression (NSUR), were proposed for estimating the parameters in the developed system of compatible individual tree DBH and AGB error-in-variable models. Moreover, two model structures were applied to estimate AGB for comparison purposes: NLS with AGB estimation depending on DBH (NLS&DD) and NLS with AGB estimation not depending on DBH (NLS&NDD). The results showed that both TSEM and NSUR led to almost the same parameter estimates for the developed system. Moreover, the developed system effectively accounted for the inherent correlation between DBH and AGB as well as for the effects of measurement errors in the DBH on the predictions of AGB, whereas NLS&DD and NLS&NDD did not. A leave-one-out cross-validation indicated that the prediction accuracy of the developed system of compatible individual tree DBH and AGB error-in-variable models with NSUR was the highest among those estimated by the four methods evaluated, but, statistically, the accuracy improvement was not significantly different from zero. The main reason might be that, except for the measurement errors, other source errors were ignored in the modeling. This study implies that, overall, the proposed method provides the potential to expand the estimations of both DBH and AGB from individual trees to stands by combining the error-in-variable modeling and LiDAR data and improve their estimation accuracies, but its application needs to be further validated by conducting a systematical uncertainty analysis of various source errors in a future study.

Designing Wood Supply Scenarios from Forest Inventories with Stratified Predictions

Forest growth and wood supply projections are increasingly used to estimate the future availability of woody biomass and the correlated effects on forests and climate. This research parameterizes an inventory-based business-as-usual wood supply scenario, with a focus on southwest Germany and the period 2002–2012 with a stratified prediction. First, the Classification and Regression Trees algorithm groups the inventory plots into strata with corresponding harvest probabilities. Second, Random Forest algorithms generate individual harvest probabilities for the plots of each stratum. Third, the plots with the highest individual probabilities are selected as harvested until the harvest probability of the stratum is fulfilled. Fourth, the harvested volume of these plots is predicted with a linear regression model trained on harvested plots only. To illustrate the pros and cons of this method, it is compared to a direct harvested volume prediction with linear regression, and a combination of logistic regression and linear regression. Direct harvested volume regression predicts comparable volume figures, but generates these volumes in a way that differs from business-as-usual. The logistic model achieves higher overall classification accuracies, but results in underestimations or overestimations of harvest shares for several subsets of the data. The stratified prediction method balances this shortcoming, and can be of general use for forest growth and timber supply projections from large-scale forest inventories.

Use of partial-coverage UAV data in sampling for large scale forest inventories

The use of image-based three-dimensional data from unmanned aerial vehicles (UAV) has proven effective for forest inventories. However, limitations in the range of UAV operations hinder their use in large scale applications. Use of partial-coverage UAV data in combination with field plots may increase precision of field-based estimates of forest resource parameters and may offer a cost-effective alternative to wall-to-wall acquisition. In this study, data from UAV collected in systematically distributed blocks and combined with field plots in a two-phase design with hybrid inference (UAVHYB) were used to estimate mean volume and its precision (standard error) for a 7330ha forest area in Norway. Because in UAVHYB the field data do not necessarily need to come from a probability sample, the approach offers great flexibility in field data collection. The estimate of precision using UAVHYB was compared to two alternative inventory approaches differing in terms of sampling design and mode of inference, i.e., (1) single-phase probabilistic sampling relying only on field samples for which the design-based approach to inference using simple random sampling estimators was adopted (FIELDDB) and (2) a model-based approach independent from design assumptions for which wall-to-wall airborne laser scanning data were applied (ALSMB). Relative efficiency (RE), calculated as the ratio between the estimated variances of two different inventory approaches, was used as measure for improvement in variance for one approach over the other. Comparison of UAVHYB against FIELDDB revealed that the use of the former was up to four times more efficient than the latter (RE=4.4). This translates to a need for 4.4 times as many field plots under simple random sampling for a FIELDDB estimate to be equally precise as the UAVHYB estimate. For ALSMB the increase in efficiency compared to UAVHYB was limited (RE=1.6). The study also demonstrated that the precision under UAVHYB can be improved when including additional field data from other inventories to enhance the model. Cost estimates for each inventory approach were compared, revealing that UAV may be a cost-effective tool for large scale forest resource assessments.