LiDAR-derived Canopy Height Research Articles

Algorithm for improving the co-registration of LiDAR-derived digital canopy height models and field data

Improving trees location under LiDAR-derived digital canopy height models (DCMs) is of great interest as discrepancies between both dataset influence the accuracy of the estimations of forest attributes. A method is proposed for the co-registration of LiDAR-derived DCMs with local field positional measurements under a dense tree canopy. This approach consists of two main stages: (1) the assessment of the match between the LiDAR-derived digital terrain model and topographic surveying measurements when shifting the coordinates around a measured position; and (2) a comparison between the field height of selected trees and the LiDAR-derived DCM. Satisfactory results were obtained from geo-referencing field data and LiDAR models for characterizing the forest structure in heterogeneous Pinus sylvestris stands. Closure error of topographic surveying was 17.7 cm, and GPS accuracy to 95 % probability was below 10 cm, thus considerably lower than the resolution of the LiDAR models (1 m-pixel). The best co-location for field trees and LiDAR models provided a coefficient of determination of 0.56 between field-measured tree heights and LiDAR-derived DCM values.

Influence of Vegetation Structure on Lidar-derived Canopy Height and Fractional Cover in Forested Riparian Buffers During Leaf-Off and Leaf-On Conditions

Estimates of canopy height (H) and fractional canopy cover (FC) derived from lidar data collected during leaf-on and leaf-off conditions are compared with field measurements from 80 forested riparian buffer plots. The purpose is to determine if existing lidar data flown in leaf-off conditions for applications such as terrain mapping can effectively estimate forested riparian buffer H and FC within a range of riparian vegetation types. Results illustrate that: 1) leaf-off and leaf-on lidar percentile estimates are similar to measured heights in all plots except those dominated by deciduous compound-leaved trees where lidar underestimates H during leaf off periods; 2) canopy height models (CHMs) underestimate H by a larger margin compared to percentile methods and are influenced by vegetation type (conifer needle, deciduous simple leaf or deciduous compound leaf) and canopy height variability, 3) lidar estimates of FC are within 10% of plot measurements during leaf-on periods, but are underestimated during leaf-off periods except in mixed and conifer plots; and 4) depth of laser pulse penetration lower in the canopy is more variable compared to top of the canopy penetration which may influence within canopy vegetation structure estimates. This study demonstrates that leaf-off lidar data can be used to estimate forested riparian buffer canopy height within diverse vegetation conditions and fractional canopy cover within mixed and conifer forests when leaf-on lidar data are not available.

Measuring forest structure along productivity gradients in the Canadian boreal with small-footprint Lidar

The structure and productivity of boreal forests are key components of the global carbon cycle and impact the resources and habitats available for species. With this research, we characterized the relationship between measurements of forest structure and satellite-derived estimates of gross primary production (GPP) over the Canadian boreal. We acquired stand level indicators of canopy cover, canopy height, and structural complexity from nearly 25,000km of small-footprint discrete return Light Detection and Ranging (Lidar) data and compared these attributes to GPP estimates derived from the MODerate resolution Imaging Spectroradiometer (MODIS). While limited in our capacity to control for stand age, we removed recently disturbed and managed forests using information on fire history, roads, and anthropogenic change. We found that MODIS GPP was strongly linked to Lidar-derived canopy cover (r = 0.74, p < 0.01), however was only weakly related to Lidar-derived canopy height and structural complexity as these attributes are largely a function of stand age. A relationship was apparent between MODIS GPP and the maximum sampled heights derived from Lidar as growth rates and resource availability likely limit tree height in the prolonged absence of disturbance. The most structurally complex stands, as measured by the coefficient of variation of Lidar return heights, occurred where MODIS GPP was highest as productive boreal stands are expected to contain a wider range of tree heights and transition to uneven-aged structures faster than less productive stands. While MODIS GPP related near-linearly to Lidar-derived canopy cover, the weaker relationships to Lidar-derived canopy height and structural complexity highlight the importance of stand age in determining the structure of boreal forests. We conclude that an improved quantification of how both productivity and disturbance shape stand structure is needed to better understand the current state of boreal forests in Canada and how these forests are changing in response to changing climate and disturbance regimes.

Semi-Supervised Methods to Identify Individual Crowns of Lowland Tropical Canopy Species Using Imaging Spectroscopy and LiDAR

Our objective is to identify and map individuals of nine tree species in a Hawaiian lowland tropical forest by comparing the performance of a variety of semi-supervised classifiers. A method was adapted to process hyperspectral imagery, LiDAR intensity variables, and LiDAR-derived canopy height and use them to assess the identification accuracy. We found that semi-supervised Support Vector Machine classification using tensor summation kernel was superior to supervised classification, with demonstrable accuracy for at least eight out of nine species, and for all combinations of data types tested. We also found that the combination of hyperspectral imagery and LiDAR data usually improved species classification. Both LiDAR intensity and LiDAR canopy height proved useful for classification of certain species, but the improvements varied depending upon the species in question. Our results pave the way for target-species identification in tropical forests and other ecosystems.

Estimating aboveground biomass in interior Alaska with Landsat data and field measurements

Terrestrial plant biomass is a key biophysical parameter required for understanding ecological systems in Alaska. An accurate estimation of biomass at a regional scale provides an important data input for ecological modeling in this region. In this study, we created an aboveground biomass (AGB) map at 30-m resolution for the Yukon Flats ecoregion of interior Alaska using Landsat data and field measurements. Tree, shrub, and herbaceous AGB data in both live and dead forms were collected in summers and autumns of 2009 and 2010. Using the Landsat-derived spectral variables and the field AGB data, we generated a regression model and applied this model to map AGB for the ecoregion. A 3-fold cross-validation indicated that the AGB estimates had a mean absolute error of 21.8Mg/ha and a mean bias error of 5.2Mg/ha. Additionally, we validated the mapping results using an airborne lidar dataset acquired for a portion of the ecoregion. We found a significant relationship between the lidar-derived canopy height and the Landsat-derived AGB (R2=0.40). The AGB map showed that 90% of the ecoregion had AGB values ranging from 10Mg/ha to 134Mg/ha. Vegetation types and fires were the primary factors controlling the spatial AGB patterns in this ecoregion.

Estimation of stand variables in Pinus radiata D. Don plantations using different LiDAR pulse densities

This study aimed to verify the validity of linear, power function and exponential models to estimate stand variables in Galician Atlantic forests of Pinus radiata D. Don using Light Detection and Ranging (LiDAR). These forests differ in structure and species composition from boreal forests, in which this methodology is fully developed. The models tested use LiDAR-derived canopy height and intensity distribution metrics as explanatory variables to predict the following stand attributes: mean height, dominant height, stand basal area, stand volume, stand crown biomass, stand stem biomass and stand aboveground biomass. Exponential models performed best in most cases, with goodness-of-fit statistics similar to those reported in the international literature for boreal forests. The coefficient of determination ranged from 0.44 (for stand crown biomass) to 0.87 (for dominant height), and the root mean square error/mean·100 ranged from 8.2 per cent (for dominant height) to 31.6 per cent (for stand stem biomass). Model precision did not essentially vary after reducing 94 per cent of the original point cloud, i.e. when laser pulse density was reduced from 8 pulses m−2 to only 0.5 pulses m−2.

Maximum entropy modeling of mature hardwood forest distribution in four U.S. states

In the eastern United States, mature hardwood forest provides habitat for many species of native flora and fauna, but is much less common now than historically. This study examined the utility of maximum entropy modeling and spatial application to identify ecosystem types like mature hardwood forest. I performed pilot modeling in Charles County, Maryland, where I compared fine-scale geographic data available locally to coarse-scale data available nationally. As expected, a model constructed with the best locally available data, including LiDAR-derived canopy height and fine-scale soil maps, outperformed a model constructed with nationally consistent data. However, the model using national data nevertheless accurately identified most mature hardwood forest sites and excluded most young forest. I then applied the coarse-scale approach to four states: Pennsylvania, Ohio, Kentucky, and Tennessee. Average test AUC (area under the receiver operating curve) based on 10 replicates varied from 0.76 to 0.80 when comparing mature hardwood forest locations to general forest locations. The maximum training or test sensitivity plus specificity threshold, depending on the state, captured 78–79% of positive locations while rejecting 74–81% of negative locations. The maximum entropy approach is versatile, and can be applied to other ecosystems and species.

Assessing a Template Matching Approach for Tree Height and Position Extraction from Lidar-Derived Canopy Height Models of Pinus Pinaster Stands

In this paper, an assessment of a method using a correlation filter over a lidar-derived digital canopy height model (CHM) is presented. The objective of the procedure is to obtain stem density, position, and height values, on a stand with the following characteristics: ellipsoidal canopy shape (Pinus pinaster), even-aged and single-layer structure. The process consists of three steps: extracting a correlation map from CHM by applying a template whose size and shape resembles the canopy to be detected, applying a threshold mask to the correlation map to keep a subset of candidate-pixels, and then applying a local maximum filter to the remaining pixel groups. The method performs satisfactorily considering the experimental conditions. The mean tree extraction percentage is 65% with a coefficient of agreement of 0.4. The mean absolute error of height is ~0.5 m for all plots except one. It can be considered a valid approach for extracting tree density and height in regularly spaced stands (i.e., poplar plantations) which are fundamental for extracting related forest parameters such as volume and biomass.

Relationship between LiDAR-derived forest canopy height and Landsat images

The mean and standard deviation (SD) of light detection and ranging (LiDAR)-derived canopy height are related to forest structure. However, LiDAR data typically cover a limited area and have a high economic cost compared with satellite optical imagery. Optical images may be required to extrapolate LiDAR height measurements across a broad landscape. Different spectral indices were obtained from three Landsat scenes. The mean, median, SD and coefficient of variation (CV) of LiDAR canopy height measurements were calculated in 30-m square blocks corresponding with Landsat Enhanced Thematic Mapper Plus (ETM+) pixels. Correlation and forward stepwise regression analysis was applied to these data sets. Mean and median LiDAR height versus normalized difference vegetation index (NDVI), normalized difference moisture index (NDMI), normalized burn ratio (NBR) and wetness Tasseled Cap showed the best correlation coefficients (R 2 ranging between −0.62 and −0.76). Nineteen regression models were obtained (R 2 = 0.65–0.70). These results show that LiDAR-derived canopy height may be associated with Landsat spectral indices. This approach is of interest in sustainable forest management, although further research is required to improve accuracy.

Simulating the impact of discrete-return lidar system and survey characteristics over young conifer and broadleaf forests

article i nfo We present a model-based investigation of the effect of discrete-return lidar system and survey characteristics on the signal recorded over young forest environments. A Monte Carlo ray tracing (MCRT) model of canopy scattering was used to examine the sensitivity of model estimates of lidar-derived canopy height, hlidar to signal triggering method, canopy structure, footprint size, sampling density and scanning angle, for broadleaf and conifer canopies of varying density. Detailed 3D models of Scots pine (Pinus sylvestris) and Downy birch (Betula pubescens) were used to simulate lidar response, with minimal assumptions about canopy structure. Use of such models allowed the impact of lidar parameters on canopy height retrieval to be tested under a range of conditions typically not possible in practice. Retrieved hlidar was generally found to be an underestimate of 'true' canopy height, hcanopy, but with exceptions. Choice of signal triggering method caused hlidar to underestimate hcanopy by ∼4% for birch and ∼7% for pine (up to 66% in extreme cases). Variations in canopy structure resulted on average in underestimation of hcanopy by 13% for birch and between 29 and 48% for pine depending on age, but with over-estimates in some cases of up to 10%. Increasing footprint diameter from 0.1 to 1 m increased retrieved hlidar from significant underestimates of hcanopy to values indistinguishable from hcanopy. Increased sampling density led to slightly increased values of hlidar to close to hcanopy, but not significantly. Increasing scan angle increased hlidar by up to 8% for birch, and 19% for pine at a scan angle of 30°. The impact of scan angle was greater for conifers as a result of large variation in crown height. Results showed that interactions between physically modelled (hypothetical) within canopy returns are similar to findings made in other studies using actual lidar systems, and that these modelled returns can depend strongly on the type of canopy and the lidar acquisition characteristics, as well as interactions between these properties. Physical models of laser pulse/canopy interactions may provide additional information on pulse interactions within the canopy, but require validation and testing before they are applied to actual survey planning and logistics.

Lidar remote sensing of forest biomass: A scale-invariant estimation approach using airborne lasers

Researchers in lidar (Light Detection And Ranging) strive to search for the most appropriate laser-based metrics as predictors in regression models for estimating forest structural variables. Many previously developed models are scale-dependent that need to be fitted and then applied both at the same scale or pixel size. The objective of this paper is to develop methods for scale-invariant estimation of forest biomass using lidar data. We proposed two scale-invariant models for biomass: a linear functional model and an equivalent nonlinear model that use lidar-derived canopy height distributions (CHD) and canopy height quantile functions (CHQ) as predictors, respectively. The two models are called functional regression models because the predictors CHD and CHQ are themselves functions or functional data. The model formulation was justified mathematically under moderate assumptions. We also created a fine-resolution biomass map by mapping individual tree component biomass in a temperate forest of eastern Texas with a lidar tree-delineation approach. The map was used as reference data to synthesize training and test datasets at multiple scales for validating the two scale-invariant models. Results suggest that the models can accurately predict biomass and yield consistent predictive performances across a variety of scales with an R 2 ranging from 0.80 to 0.95 (RMSE: from 14. 3 Mg/ha to 33.7 Mg/ha) among all the fitted models. Results also show that a training data size of around 50 plots or less was enough to guarantee a good fitting of the linear functional model. Our findings demonstrate the effectiveness of CHD and CHQ as lidar metrics for estimating biomass as well as the capability of lidar for mapping biomass at a range of scales. The functional regression models of this study are useful for lidar-based forest inventory tasks where the analysis units vary in size and shape. They also hold promise for estimating other forest characteristics such as below-ground biomass, timber volume, crown fuel weight, and Leaf Area Index.

Estimating biomass of individual pine trees using airborne lidar

Airborne lidar (Light Detection And Ranging) is a proven technology that can be used to accurately assess aboveground forest biomass and bio-energy feedstocks. The overall goal of this study was to develop a method for assessing aboveground biomass and component biomass for individual trees using airborne lidar data in forest settings typical for loblolly pine stands (Pinus taeda L.) in the southeastern United States. More specific objectives included: (1) assessing the accuracy of estimating diameter at breast height (dbh) for individual pine trees using lidar-derived individual tree measurements, such as tree height and crown diameter, and (2) investigating the use of lidar-derived individual tree measurements with linear and nonlinear regression to estimate per tree aboveground biomass. In addition, the study presents a method for estimating the biomass of individual tree components, such as foliage, coarse roots, stem bark, and stem wood, as derived quantities from the aboveground biomass prediction. A lidar software application, TreeVaW, was used to extract forest inventory parameters at individual tree level from a lidar-derived canopy height model. Lidar-measured parameters at individual tree level, such as height and crown diameter, were used with regression models to estimate dbh, aboveground tree biomass, and tree-component biomass. Field measurements were collected for 45 loblolly pine trees over 0.1- and 0.01-acre plots. Linear regression models were able to explain 93% of the variability associated with individual tree biomass, 90% for dbh, and 79–80% for components biomass.

Forest Volume and Biomass Estimation Using Small-Footprint Lidar-Distributional Parameters on a Per-Segment Basis

Abstract This study assessed a lidar-based, object-oriented (segmentation) approach to forest volume and aboveground biomass modeling. The study area in the Piedmont physiographic region of Virginia is composed of temperate coniferous, deciduous, and mixed stands. Segmentation objects, hierarchical in terms of area and ranging from 0.035 to 5.632 ha/object, were created using a lidar-derived canopy height model. Horizontal point (basal area) samples were used to calculate volume and aboveground biomass. Per-object lidar point (per return height and intensity) distributional parameters were extracted from small-footprint lidar. Adjusted R 2 and Mallow's Cp metrics were used to select models for the range of segmentation results. Selected variables included intensity-based and structurally related first through fifth return height parameters. Object-based modeling (adjusted R 2 = 0.58–0.79; various object sizes) resulted in distinct improvements over stand-based attempts (adjusted R 2 = 0.40–0.73; majority adjusted R 2 &amp;lt; 0.50). Adjusted R 2 and RMSE values for deciduous volume (0.59; 51.15 m3/ha) and biomass (0.58; 37.41 Mg/ha) were better than those found for another, plot-based study in the study area. Coniferous R 2 values for volume (0.66) and biomass (0.59) were lower than previous studies, which was attributed to variability within the relatively narrow volume range (6.94–50.93 m3/ha).

Fusion of Small-Footprint Lidar and Multispectral Data to Estimate Plot- Level Volume and Biomass in Deciduous and Pine Forests in Virginia, USA

Abstract The principal study objective was to explore the feasibility of using small-footprint lidar data and multispectral imagery to estimate forest volume and biomass on small (0.017-ha) plots. In addition, the spatial dependency of residuals between ground-measured and lidar-estimated variables was investigated. The lidar data set was acquired over deciduous and pine stands in the southeastern United States. Individual trees were identified on the lidar-derived canopy height model by local maximum focal filtering with both square and circular windows of variable size. The size of the dynamically varying window was based on the height of the canopy and the taxonomic group as derived from coregistered multispectral optical data. Lidar-measured parameters at an individual tree level (height, crown diameter) were used with regression models and cross validation to estimate plot-level field inventory data, including volume, biomass, basal area, and diameter at breast height (dbh). Maximum R2 values for estimating biomass were 0.32 for deciduous trees (RMSE 44 Mg/ha) and 0.82 for pines (RMSE 29 Mg/ha). When estimating volume, maximum R2 values for deciduous trees were 0.39 (RMSE 52.84 m3/ha) and 0.83 for pines (RMSE 47.9 m3/ha). Calculation of Moran's I coefficient for each regression model revealed a lack of significant autocorellation of the residuals at 0.05 level. Both model fit and prediction for volume and biomass models for deciduous and pine plots indicated that the circular window shape is more appropriate to locate individual trees with lidar. Using the fused data set, lidar and optical imagery, as opposed to using lidar data alone, always improved biomass and volume estimates for pines and in some cases for deciduous plots as well. FOR. SCI. 50(4):551–565.