Search

Geolocation error in spaceborne sampling light detection and ranging (LiDAR) measurements of forest structure can compromise forest attribute estimates and degrade integration with georeferenced field measurements or other remotely sensed data. Data integration is especially problematic when geolocation error is not well quantified. We propose a general model that uses airborne laser scanning data to quantify and correct geolocation error in spaceborne sampling LiDAR. To illustrate the model, LiDAR data from NASA Goddard's LiDAR Hyperspectral and Thermal Imager (G‐LiHT) was used with a subset of LiDAR data from NASA's Global Ecosystem Dynamics Investigation (GEDI). The model accommodates multiple canopy height metrics derived from a simulated GEDI footprint kernel using spatially coincident G‐LiHT, and incorporates both additive and multiplicative mapping between the canopy height metrics generated from both datasets. A Bayesian implementation provides probabilistic uncertainty quantification in both parameter and geolocation error estimates. Results show a systematic geolocation error of 9.62 m in the southwest direction. In addition, estimated geolocation errors within GEDI footprints were highly variable, with results showing a ∼$$ \sim $$0.45 probability the true footprint center is within 20 m. Estimating and correcting geolocation error via the model outlined here can help inform subsequent efforts to integrate spaceborne LiDAR data, like GEDI, with other georeferenced data. [ABSTRACT FROM AUTHOR]

Light detection and ranging (LiDAR) is commonly used to predict forest inventory attributes of interest across large regions. Most studies utilize model-derived estimators whose performances are affected by training data, and give less attention to using design-derived estimators. The influence of sample design and estimation method is an important consideration for determining sample sizes or calibration plot densities; however, this has not been systematically explored, particularly in mixed-species forests. In this study, 10 sample selection methods (four equal probability and six variable probability selection methods) and six estimation techniques (two model-derived and four sample-derived estimators) across a range of sample sizes are evaluated using LiDAR-derived predictions of volume per ha. Results show that the use of variable probability selection methods combined with sample-derived estimation techniques are more efficient than using model-derived estimates. Estimation technique had a greater effect on sample efficiency than did selection method, though specific combinations were more efficient than others. For example, random forest imputation was the most efficient at the lowest sample sizes (n < 50); however, significant biases were obtained when used with variable probability selection methods. The required plot densities across the combinations of selection methods and estimation techniques used in this study ranged from one plot per 15.7–32.6 ha. Use of a variable probability selection method based on attributes derived directly from LiDAR point clouds coupled with a ratio or regression estimator was a very efficient LiDAR-assisted sampling design that should be considered more in the future. [ABSTRACT FROM AUTHOR]

Copyright of Canadian Journal of Remote Sensing is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

LiDAR has become a common means for providing area-based estimates of forest attributes. An important, yet understudied, factor influencing forest attribute estimation is the prediction cell size. In this analysis, the influence of 6 prediction cell sizes (2-, 4-, 10-, 25-, 50-, and 100-m squares) and 2 prediction methods (parametric vs. nonparametric) on LiDAR-derived, area-based, stand- and forest-level estimates of total volume at 2 separate locations in the Acadian Forest were evaluated and compared to field-based measurements. Statistically significant differences (p< 0.05) were observed between the field-based and some of the LiDAR-derived volumes. The stand-level percent differences between the various cell sizes ranged from < 0.1 to >38%, and were generally greatest at the 2 extreme cell sizes examined (2 m and 100 m). At the forest-level, all LiDAR-estimates were within the 95% confidence interval for the ground-based estimate, except for a few values from smaller cell sizes. There was also a general trend of smaller cell sizes producing lower values of volume, whereas larger cell sizes tended to give higher estimates. Overall, this study highlights the importance cell size influence on stand- and forest-level estimates and indicates that additional work is needed to better understand the trends identified in this analysis. [ABSTRACT FROM PUBLISHER]

Light detection and ranging (LiDAR) has become a common means for predicting key forest structural attributes, but comparisons of alternative statistical methods and the spatial extent of LiDAR metrics extraction on independent datasets have been minimal. The primary objective of this study was to assess the performance of local and non-local LiDAR aboveground biomass (AGB) prediction models at two locations in the Acadian Forest. Two common statistical techniques, nonlinear mixed e ects (NLME) and random forest (RF), were used to t the prediction models and compared. Finally, this study evaluated the inuence of alternative plot radii for LiDAR metrics extraction on model t and prediction accuracy. AGB models were independently developed at each forest and tested both locally (model applied to same forest used for development) and non-locally (model applied to di erent forest) using an extensive network of ground-based plots. In general, RF was found to outperform NLME when applied locally, but the di erences between the approaches were negligible when applied to the non-local dataset. NLME was found to perform equally well locally and non-locally. LiDAR extraction radius had very little inuence on model performance as well. Minimal di erences between models developed using xed- and variable-radius methods were found, while the optimal LiDAR extraction radius was not consistent among forests, statistical technique, or local vs. non-local. Overall, the results highlight the importance of a robust calibration dataset that covers the full range of observed variation for developing accurate prediction models based on remote sensing data. [ABSTRACT FROM AUTHOR]

The objective of this study was to evaluate the applicability of using a low-density (1-3 points m-2) discrete-return LiDAR (Light Detection and Ranging) for predicting maximum tree height, stem density, basal area, quadratic mean diameter and total volume. The research was conducted at the Penobscot Experimental Forest in central Maine, where a range of stand structures and species composition is present and generally representative of northern Maine's forests. Prediction models were developed utilizing the random forest algorithm that was calibrated using reference data collected in fixed radius circular plots. For comparison, the volume model used two sets of reference data, with one being fixed radius circular plots and the other variable radius plots. Prediction biases were evaluated with respect to five silvicultural treatments and softwood species composition based on the coefficient of determination (R²), root mean square error and mean bias, as well as residual scatter plots. Overall, this study found that LiDAR tended to underestimate maximum tree height and volume. The maximum tree height and volume models had R² values of 86.9% and 72.1%, respectively. The accuracy of volume prediction was also sensitive to the plot type used. While it was difficult to develop models with a high R², due to the complexities of Maine's forest structures and species composition, the results suggest that low density LiDAR can be used as a supporting tool in forest management for this region. [ABSTRACT FROM AUTHOR]

Light detection and ranging (LiDAR) has become a commonly-used tool for generating remotely-sensed forest inventories. However, LiDAR-derived forest inventories have remained uncommon at a regional scale due to varying parameters among LiDAR data acquisitions and the availability of sufficient calibration data. Here, we present a model using a 3-D convolutional neural network (CNN), a form of deep learning capable of scanning a LiDAR point cloud, combined with coincident satellite data (spectral, phenology, and disturbance history). We compared this approach to traditional modeling used for making forest predictions from LiDAR data (height metrics and random forest) and found that the CNN had consistently lower uncertainty. We then applied the CNN to public data over six New England states in the USA, generating maps of 14 forest attributes at a 10 m resolution over 85% of the region. Aboveground biomass estimates produced a root mean square error of 36 Mg ha−1 (44%) and were within the 97.5% confidence of independent county-level estimates for 33 of 38 or 86.8% of the counties examined. CNN predictions for stem density and percentage of conifer attributes were moderately successful, while predictions for detailed species groupings were less successful. The approach shows promise for improving the prediction of forest attributes from regional LiDAR data and for combining disparate LiDAR datasets into a common framework for large-scale estimation. [ABSTRACT FROM AUTHOR]

This study assesses univariate and multivariate spatial regression models for predicting individual tree structure variables using Light Detection And Ranging (LiDAR) covariates. Many studies have used covariates derived from LiDAR to help explain the variability in tree, stand, or forest variables at a fine spatial resolution across a specified domain. Few studies use regression models capable of accommodating residual spatial dependence between field measurements. Failure to acknowledge this spatial dependence can result in biased and perhaps misleading inference about the importance of LiDAR covariates and erroneous prediction. Accommodating residual spatial dependence, via spatial random effects, helps to meet basic model assumptions and, as illustrated in this study, can improve model fit and prediction. When multiple correlated tree structure variables are considered, it is attractive to specify joint models that are able to estimate the within tree covariance structure and use it for subsequent prediction for unmeasured trees. We capture within tree residual covariances by specifying a model with multivariate spatial random effects. The univariate and multivariate spatial random effects models are compared to those without random effects using a data set collected on the U.S. Forest Service Penobscot Experimental Forest, Maine. These data comprise individual tree measurements including geographic position, height, average crown length, average crown radius, and diameter at breast height. [ABSTRACT FROM PUBLISHER]

Fast-growing Eucalyptus spp. forest plantations and their resultant wood products are economically important and may provide a low-cost means to sequester carbon for greenhouse gas reduction. The development of advanced and optimized frameworks for estimating forest plantation attributes from lidar remote sensing data combined with statistical modeling approaches is a step towards forest inventory operationalization and might improve industry efficiency in monitoring and managing forest resources. In this study, we first developed and tested a framework for modeling individual tree attributes in fast-growing Eucalyptus forest plantation using airborne lidar data and linear mixed-effect models (LME) and assessed the gain in accuracy compared to a conventional linear fixed-effects model (LFE). Second, we evaluated the potential of using the tree-level estimates for determining tree attribute uniformity across different stand ages. In the field, tree measurements, such as tree geolocation, species, genotype, age, height (Ht), and diameter at breast height (dbh) were collected through conventional forest inventory practices, and tree-level aboveground carbon (AGC) was estimated using allometric equations. Individual trees were detected and delineated from lidar-derived canopy height models (CHM), and crown-level metrics (e.g., crown volume and crown projected area) were computed from the lidar 3-D point cloud. Field and lidar-derived crown metrics were combined for ht, dbh, and AGC modeling using an LME. We fitted a varying intercept and slope model, setting species, genotype, and stand (alone and nested) as random effects. For comparison, we also modeled the same attributes using a conventional LFE model. The tree attribute estimates derived from the best LME model were used for assessing forest uniformity at the tree level using the Lorenz curves and Gini coefficient (GC). We successfully detected 96.6% of the trees from the lidar-derived CHM. The best LME model for estimating the tree attributes was composed of the stand as a random effect variable, and canopy height, crown volume, and crown projected area as fixed effects. The %RMSE values for tree-level height, dbh, and AGC were 8.9%, 12.1%, and 23.7% for the LFE model and improved to 7.3%, 7.1%, and 13.6%, respectively, for the LME model. Tree attributes uniformity was assessed with the Lorenz curves and tree-level estimations, especially for the older stands. All stands showed a high level of tree uniformity with GC values approximately 0.2. This study demonstrates that accurate detection of individual trees and their associated crown metrics can be used to estimate Ht, dbh, and AGC stocks as well as forest uniformity in fast-growing Eucalyptus plantations forests using lidar data as inputs to LME models. This further underscores the high potential of our proposed approach to monitor standing stock and growth in Eucalyptus—and similar forest plantations for carbon dynamics and forest product planning. [ABSTRACT FROM AUTHOR]

Light detection and ranging (LiDAR) data have become almost ubiquitous as a remote sensing tool in forestry estimation and mapping applications. Such initiatives commonly rely on spatially aligned forest inventory plot measurements and LiDAR covariates to inform model-based estimators for small area estimation. There are many examples where such linking models provide the desired accuracy and precision of forest parameter estimates for small areas where paucity of inventory plot observations preclude design-based inference. This paper builds on previous small area estimation (SAE) work by linking LiDAR covariates with variable radius forest inventory plot measurements within a hierarchical Bayesian framework. Using this framework, we compare SAE of forest aboveground biomass using: i ) Fay-Herriot (FH); ii ) FH with conditional autoregressive random effects (FHCAR); and iii ) FHCAR with smoothed sampling variance (FHCAR-SMOOTH) models. Candidate models and the direct estimate based on plot measurements alone were compared using coefficient of variation (CV). On average, the FH model reduced the CV by 52.3% compared to the direct estimate. Incorporating spatial structure via the FHCAR model reduced the CV by 56.9% and 10.8% relative to the direct and the FH model estimates, respectively. Overall, these results illustrate the applicability and utility of using a SAE framework for linking LiDAR with typical forest inventory data. [ABSTRACT FROM AUTHOR]

Combining spatially-explicit long-term forest inventory and remotely sensed information from Light Detection and Ranging (LiDAR) datasets through statistical models can be a powerful tool for predicting and mapping above-ground biomass (AGB) at a range of geographic scales. We present and examine a novel modeling approach to improve prediction of AGB and estimate AGB growth using LiDAR data. The proposed model accommodates temporal misalignment between field measurements and remotely sensed data—a problem pervasive in such settings—by including multiple time-indexed measurements at plot locations to estimate AGB growth. We pursue a Bayesian modeling framework that allows for appropriately complex parameter associations and uncertainty propagation through to prediction. Specifically, we identify a space-varying coefficients model to predict and map AGB and its associated growth simultaneously. The proposed model is assessed using LiDAR data acquired from NASA Goddard's LiDAR, Hyper-spectral & Thermal imager and field inventory data from the Penobscot Experimental Forest in Bradley, Maine. The proposed model outperformed the time-invariant counterpart models in predictive performance as indicated by a substantial reduction in root mean squared error. The proposed model adequately accounts for temporal misalignment through the estimation of forest AGB growth and accommodates residual spatial dependence. Results from this analysis suggest that future AGB models informed using remotely sensed data, such as LiDAR, may be improved by adapting traditional modeling frameworks to account for temporal misalignment and spatial dependence using random effects. [ABSTRACT FROM AUTHOR]