Comparison of Modeling Algorithms for Forest Canopy Structures Based on UAV-LiDAR: A Case Study in Tropical China

Knowledge of forest structure is vital for sustainable forest management decisions. Terrestrial laser scanning cannot describe the canopy trees in a large area, and it is unclear whether unmanned aerial vehicle-light detection and ranging (UAV-LiDAR) data have the ability to capture the forest canopy structural parameters in tropical forests. In this study, we estimated five forest canopy structures (stand density (N), basic area (G), above-ground biomass (AGB), Lorey’s mean height (HL), and under-crown height (hT)) with four modeling algorithms (linear regression (LR), bagged tree (BT), support vector regression (SVR), and random forest (RF)) based on UAV-LiDAR data and 60 sample plot data from tropical forests in Hainan and determined the optimal algorithms for the five canopy structures by comparing the performance of the four algorithms. First, we defined the canopy tree as a tree with a height ≥70% HL. Then, UAV-LiDAR metrics were calculated, and the LiDAR metrics were screened by recursive feature elimination (RFE). Finally, a prediction model of the five forest canopy structural parameters was established by the four algorithms, and the results were compared. The metrics’ screening results show that the most important LiDAR indexes for estimating HL, AGB, and hT are the leaf area index and some height metrics, while the most important indexes for estimating N and G are the kurtosis of heights and the coefficient of variation of height. The relative root mean squared error (rRMSE) of five structure parameters showed the following: when modeling HL, the rRMSEs (10.60%–12.05%) obtained by the four algorithms showed little difference; when N was modeled, BT, RF, and SVR had lower rRMSEs (26.76%–27.44%); when G was modeled, the rRMSEs of RF and SVR (15.37%–15.87%) were lower; when hT was modeled, BT, RF, and SVR had lower rRMSEs (10.24%–11.07%); when AGB was modeled, RF had the lowest rRMSE (26.75%). Our results will help facilitate choosing LiDAR indexes and modeling algorithms for tropical forest resource inventories.

Download Full-text

Estimation of diameter at breast height from mobile laser scanning data collected under a heavy forest canopy

Journal of Forest Science ◽

10.17221/28/2017-jfs ◽

2017 ◽

Vol 63 (No. 9) ◽

pp. 433-441 ◽

Cited By ~ 8

Author(s):

Čerňava Juraj ◽

Tuček Ján ◽

Koreň Milan ◽

Mokroš Martin

Keyword(s):

Laser Scanning ◽

Forest Canopy ◽

Mean Squared Error ◽

Satellite System ◽

Tree Level ◽

Efficient Technology ◽

Mobile Mapping ◽

Mobile Mapping System ◽

Global Navigation Satellite ◽

Short Time

Mobile laser scanning (MLS) is time-efficient technology of geospatial data collection that proved its ability to provide accurate measurements in many fields. Mobile innovation of the terrestrial laser scanning has a potential to collect forest inventory data on a tree level from large plots in a short time. Valuable data, collected using mobile mapping system (MMS), becomes very difficult to process when Global Navigation Satellite System (GNSS) outages become too long. A heavy forest canopy blocking the GNSS signal and limited accessibility can make mobile mapping very difficult. This paper presents processing of data collected by MMS under a heavy forest canopy. DBH was estimated from MLS point cloud using three different methods. Root mean squared error varied between 2.65 and 5.57 cm. Our research resulted in verification of the influence of MLS coverage of tree stem on the accuracy of DBH data.

Download Full-text

Optimizing the Remote Detection of Tropical Rainforest Structure with Airborne Lidar: Leaf Area Profile Sensitivity to Pulse Density and Spatial Sampling

Remote Sensing ◽

10.3390/rs11010092 ◽

2019 ◽

Vol 11 (1) ◽

pp. 92 ◽

Cited By ~ 20

Author(s):

Danilo Roberti Alves de Almeida ◽

Scott C. Stark ◽

Gang Shao ◽

Juliana Schietti ◽

Bruce Walker Nelson ◽

...

Keyword(s):

Grain Size ◽

Leaf Area ◽

Laser Scanning ◽

Forest Canopy ◽

Primary Source ◽

Airborne Lidar ◽

Area Index ◽

Pulse Density ◽

Optimal Values ◽

And Function

Airborne Laser Scanning (ALS) has been considered as a primary source to model the structure and function of a forest canopy through the indicators leaf area index (LAI) and vertical canopy profiles of leaf area density (LAD). However, little is known about the effects of the laser pulse density and the grain size (horizontal binning resolution) of the laser point cloud on the estimation of LAD profiles and their associated LAIs. Our objective was to determine the optimal values for reliable and stable estimates of LAD profiles from ALS data obtained over a dense tropical forest. Profiles were compared using three methods: Destructive field sampling, Portable Canopy profiling Lidar (PCL) and ALS. Stable LAD profiles from ALS, concordant with the other two analytical methods, were obtained when the grain size was less than 10 m and pulse density was high (>15 pulses m−2). Lower pulse densities also provided stable and reliable LAD profiles when using an appropriate adjustment (coefficient K). We also discuss how LAD profiles might be corrected throughout the landscape when using ALS surveys of lower density, by calibrating with LAI measurements in the field or from PCL. Appropriate choices of grain size, pulse density and K provide reliable estimates of LAD and associated tree plot demography and biomass in dense forest ecosystems.

Download Full-text

Using terrestrial laser scanning to constrain forest ecosystem structure and functions in the Ecosystem Demography model (ED2.2)

10.5194/gmd-2021-59 ◽

2021 ◽

Author(s):

Félicien Meunier ◽

Sruthi M. Krishna Moorthy ◽

Marc Peaucelle ◽

Kim Calders ◽

Louise Terryn ◽

...

Keyword(s):

Leaf Area ◽

Forest Structure ◽

Laser Scanning ◽

Initial Conditions ◽

Structural Parameters ◽

Terrestrial Laser Scanning ◽

Dimensional Structure ◽

Forest Site ◽

Ecosystem Structure ◽

Area Index

Abstract. Terrestrial Biosphere Modeling (TBM) is an invaluable approach for studying plant-atmosphere interactions at multiple spatial and temporal scales, as well as the global change impacts on ecosystems. Yet, TBM projections suffer from large uncertainties that limit their usefulness. A large part of this uncertainty arises from the empirical allometric (size-tomass) relationships that are used to represent forest structure in TBMs. Forest structure actually drives a large part of TBM uncertainty as it regulates key processes such as the transfer of carbon, energy, and water between the land and atmosphere, but remains challenging to measure and reliably represent. The poor representation of forest structure in TBMs results in models that are able to reproduce observed land fluxes, but which fail to realistically represent carbon pools, forest composition, and demography. Recent advances in Terrestrial Laser Scanning (TLS) techniques offer a huge opportunity to capture the three-dimensional structure of the ecosystem and transfer this information to TBMs in order to increase their accuracy. In this study, we quantified the impacts of integrating structural observations of individual trees (namely tree height, leaf area, woody biomass, and crown area) derived from TLS into the state-of-the-art Ecosystem Demography model (ED2.2) at a temperate forest site. We assessed the relative model sensitivity to initial conditions, allometric parameters, and canopy representation by changing them in turn from default configurations to site-specific, TLS-derived values. We show that forest demography and productivity as modelled by ED2.2 are sensitive to the imposed initial state, the model structural parameters, and the way canopy is represented. In particular, we show that: 1) the imposed openness of the canopy dramatically influenced the potential vegetation, the optimal ecosystem leaf area, and the vertical distribution of light in the forest, as simulated by ED2.2; 2) TLS-derived allometric parameters increased simulated leaf area index and aboveground biomass by 57 and 75 %, respectively; 3) the choice of model structure and allometric coefficient both significantly impacted the optimal set of parameters necessary to reproduce eddy covariance flux data.

Download Full-text

FEASIBILITY OF MACHINE LEARNING METHODS FOR SEPARATING WOOD AND LEAF POINTS FROM TERRESTRIAL LASER SCANNING DATA

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-annals-iv-2-w4-157-2017 ◽

2017 ◽

Vol IV-2/W4 ◽

pp. 157-164 ◽

Cited By ~ 6

Author(s):

D. Wang ◽

M. Hollaus ◽

N. Pfeifer

Keyword(s):

Machine Learning ◽

Laser Scanning ◽

Learning Algorithms ◽

Terrestrial Laser Scanning ◽

Gaussian Mixture ◽

Machine Learning Algorithms ◽

Support Vector ◽

Area Index ◽

Training Samples

Classification of wood and leaf components of trees is an essential prerequisite for deriving vital tree attributes, such as wood mass, leaf area index (LAI) and woody-to-total area. Laser scanning emerges to be a promising solution for such a request. Intensity based approaches are widely proposed, as different components of a tree can feature discriminatory optical properties at the operating wavelengths of a sensor system. For geometry based methods, machine learning algorithms are often used to separate wood and leaf points, by providing proper training samples. However, it remains unclear how the chosen machine learning classifier and features used would influence classification results. To this purpose, we compare four popular machine learning classifiers, namely Support Vector Machine (SVM), Na¨ıve Bayes (NB), Random Forest (RF), and Gaussian Mixture Model (GMM), for separating wood and leaf points from terrestrial laser scanning (TLS) data. Two trees, an <i>Erytrophleum fordii</i> and a <i>Betula pendula</i> (silver birch) are used to test the impacts from classifier, feature set, and training samples. Our results showed that RF is the best model in terms of accuracy, and local density related features are important. Experimental results confirmed the feasibility of machine learning algorithms for the reliable classification of wood and leaf points. It is also noted that our studies are based on isolated trees. Further tests should be performed on more tree species and data from more complex environments.

Download Full-text

Forest canopy structural parameters and Leaf Area Index retrieval using multi-sensors synergy observations

10.1117/12.825928 ◽

2009 ◽

Cited By ~ 1

Author(s):

Zhuo Fu ◽

Jindi Wang ◽

Jinling Song ◽

Hongmin Zhou ◽

Yong Pang ◽

...

Keyword(s):

Leaf Area Index ◽

Leaf Area ◽

Forest Canopy ◽

Structural Parameters ◽

Area Index

Download Full-text

Evaluation of the FluorWPS Model and Study of the Parameter Sensitivity for Simulating Solar-Induced Chlorophyll Fluorescence

Remote Sensing ◽

10.3390/rs13061091 ◽

2021 ◽

Vol 13 (6) ◽

pp. 1091

Author(s):

Chiming Tong ◽

Yunfei Bao ◽

Feng Zhao ◽

Chongrui Fan ◽

Zhenjiang Li ◽

...

Keyword(s):

Chlorophyll Fluorescence ◽

Radiative Transfer ◽

High Performance ◽

Forest Canopy ◽

Structural Parameters ◽

Coefficient Of Determination ◽

Leaf Angle ◽

Angle Distribution ◽

Structural Factors ◽

Area Index

Solar-induced chlorophyll fluorescence (SIF) has been used as an indicator for the photosynthetic activity of vegetation at regional and global scales. Canopy structure affects the radiative transfer process of SIF within canopy and causes the angular-dependencies of SIF. A common solution for interpreting these effects is the use of physically-based radiative transfer models. As a first step, a comprehensive evaluation of the three-dimensional (3D) radiative transfers is needed using ground truth biological and hyperspectral remote sensing measurements. Due to the complexity of forest modeling, few studies have systematically investigated the effect of canopy structural factors and sun-target-viewing geometry on SIF. In this study, we evaluated the capability of the Fluorescence model with the Weighted Photon Spread method (FluorWPS) to simulate at-sensor radiance and SIF at the top of canopy, and identified the influence of the canopy structural factors and sun-target-viewing geometry on the magnitude and directional response of SIF in deciduous forests. To evaluate the model, a 3D forest scene was first constructed from Goddard’s LiDAR Hyperspectral and Thermal (G-LiHT) LiDAR data. The reliability of the reconstructed scene was confirmed by comparing the calculated leaf area index with the measured ones from the scene, which resulted in a relative error of 3.5%. Then, the performance of FluorWPS was evaluated by comparing the simulated at-sensor radiance spectra with the spectra measured from the DUAL and FLUO spectrometer of HyPlant. The radiance spectra simulated by FluorWPS agreed well with the measured spectra by the two high-performance imaging spectrometers, with a coefficient of determination (R2) of 0.998 and 0.926, respectively. SIF simulated by the FluorWPS model agreed well with the values of the DART model. Furthermore, a sensitivity analysis was conducted to assess the effect of the canopy structural parameters and sun-target-viewing geometry on SIF. The maximum difference of the total SIF can be as large as 45% and 47% at the wavelengths of 685 nm and 740 nm for different foliage area volume densities (FAVDs), and 48% and 46% for fractional vegetation covers (FVCs), respectively. Leaf angle distribution has a markedly influence on the magnitude of SIF, with a ratio of emission part to SIF range from 0.48 to 0.72. SIF from the grass layer under the tree contributed 10%+ more to the top of canopy SIF even for a dense forest canopy (FAVD = 3.5 m−1, FVC = 76%). The red SIF at the wavelength of 685 nm had a similar shape to the far-red SIF at a wavelength of 740 nm but with higher variability in varying illumination conditions. The integration of the FluorWPS model and LiDAR modeling can greatly improve the interpretation of SIF at different scales and angular configurations.

Download Full-text

Estimation of LAI with the LiDAR Technology: A Review

Remote Sensing ◽

10.3390/rs12203457 ◽

2020 ◽

Vol 12 (20) ◽

pp. 3457

Author(s):

Yao Wang ◽

Hongliang Fang

Keyword(s):

Laser Scanning ◽

Large Scale ◽

Scale Validation ◽

Canopy Structure ◽

Impact Factors ◽

Future Research ◽

Lidar Data ◽

Biophysical Parameters ◽

Area Index ◽

Contact Frequency

Leaf area index (LAI) is an important vegetation parameter. Active light detection and ranging (LiDAR) technology has been widely used to estimate vegetation LAI. In this study, LiDAR technology, LAI retrieval and validation methods, and impact factors are reviewed. First, the paper introduces types of LiDAR systems and LiDAR data preprocessing methods. After introducing the application of different LiDAR systems, LAI retrieval methods are described. Subsequently, the review discusses various LiDAR LAI validation schemes and limitations in LiDAR LAI validation. Finally, factors affecting LAI estimation are analyzed. The review presents that LAI is mainly estimated from LiDAR data by means of the correlation with the gap fraction and contact frequency, and also from the regression of forest biophysical parameters derived from LiDAR. Terrestrial laser scanning (TLS) can be used to effectively estimate the LAI and vertical foliage profile (VFP) within plots, but this method is affected by clumping, occlusion, voxel size, and woody material. Airborne laser scanning (ALS) covers relatively large areas in a spatially contiguous manner. However, the capability of describing the within-canopy structure is limited, and the accuracy of LAI estimation with ALS is affected by the height threshold and sampling size, and types of return. Spaceborne laser scanning (SLS) provides the global LAI and VFP, and the accuracy of estimation is affected by the footprint size and topography. The use of LiDAR instruments for the retrieval of the LAI and VFP has increased; however, current LiDAR LAI validation studies are mostly performed at local scales. Future research should explore new methods to invert LAI and VFP from LiDAR and enhance the quantitative analysis and large-scale validation of the parameters.

Download Full-text

Classification of Mangrove Species Using Combined WordView-3 and LiDAR Data in Mai Po Nature Reserve, Hong Kong

Remote Sensing ◽

10.3390/rs11182114 ◽

2019 ◽

Vol 11 (18) ◽

pp. 2114 ◽

Cited By ~ 3

Author(s):

Qiaosi Li ◽

Frankie Kwan Kit Wong ◽

Tung Fung

Keyword(s):

Hong Kong ◽

Species Distribution ◽

Nature Reserve ◽

Canopy Height ◽

Recursive Feature Elimination ◽

Support Vector ◽

Svm Classifier ◽

Lidar Data ◽

Mangrove Species ◽

Spectral Bands

Mangroves have significant social, economic, environmental, and ecological values but they are under threat due to human activities. An accurate map of mangrove species distribution is required to effectively conserve mangrove ecosystem. This study evaluates the synergy of WorldView-3 (WV-3) spectral bands and high return density LiDAR-derived elevation metrics for classifying seven species in mangrove habitat in Mai Po Nature Reserve in Hong Kong, China. A recursive feature elimination algorithm was carried out to identify important spectral bands and LiDAR (Airborne Light Detection and Ranging) metrics whilst appropriate spatial resolution for pixel-based classification was investigated for discriminating different mangrove species. Two classifiers, support vector machine (SVM) and random forest (RF) were compared. The results indicated that the combination of 2 m resolution WV-3 and LiDAR data yielded the best overall accuracy of 0.88 by SVM classifier comparing with WV-3 (0.72) and LiDAR (0.79). Important features were identified as green (510–581 nm), red edge (705–745 nm), red (630–690 nm), yellow (585–625 nm), NIR (770–895 nm) bands of WV-3, and LiDAR metrics relevant to canopy height (e.g., canopy height model), canopy shape (e.g., canopy relief ratio), and the variation of height (e.g., variation and standard deviation of height). LiDAR features contributed more information than spectral features. The significance of this study is that a mangrove species distribution map with satisfactory accuracy can be acquired by the proposed classification scheme. Meanwhile, with LiDAR data, vertical stratification of mangrove forests in Mai Po was firstly mapped, which is significant to bio-parameter estimation and ecosystem service evaluation in future studies.

Download Full-text

Remote Sensing of Leaf Area Index from LiDAR Height Percentile Metrics and Comparison with MODIS Product in a Selectively Logged Tropical Forest Area in Eastern Amazonia

10.32920/14669124 ◽

2021 ◽

Author(s):

Yonghua Qu ◽

Ahmed Shaker ◽

Carlos Alberto Silva ◽

Carine Klauberg ◽

Ekena Rangel Pinagé

Keyword(s):

Tropical Forest ◽

Leaf Area Index ◽

Leaf Area ◽

Tropical Forests ◽

Airborne Lidar ◽

Lidar Data ◽

Area Index ◽

Dense Forest ◽

Modis Lai ◽

Airborne Lidar Data

Leaf area index (LAI) is an important parameter to describe the capacity of forests to intercept light and thus affects the microclimate and photosynthetic capacity of canopies. In general, tropical forests have a higher leaf area index and it is a challenge to estimate LAI in a forest with a very dense canopy. In this study, it is assumed that the traditional Light Detection and Ranging (LiDAR)-derived fractional vegetation cover (fCover) has weak relationship with leaf area index in a dense forest. We propose a partial least squares (PLS) regression model using the height percentile metrics derived from airborne LiDAR data to estimate the LAI of a dense forest. Ground inventory and airborne LiDAR data collected in a selectively logged tropical forest area in Eastern Amazonia are used to map LAI from the plot level to the landscape scale. The results indicate that the fCover, derived from the first return or the last return, has no significant correlations with the ground-based LAI. The PLS model evaluated by the leave-one-out validation shows that the estimated LAI is significantly correlated with the ground-based LAI with an R2 of 0.58 and a root mean square error (RMSE) of 1.13. A data comparison indicates that the Moderate Resolution Imaging Spectrometer (MODIS) LAI underestimates the landscape-level LAI by about 22%. The MODIS quality control data show that in the selected tile, the cloud state is not the primary factor affecting the MODIS LAI performance; rather, the LAI from the main radiative transfer (RT) algorithm contributes much to the underestimation of the LAI in the tropical forest. In addition, the results show that the LiDAR-based LAI has a better response to the logging activities than the MODIS-based LAI, and that the leaf area reduction caused by logging is about 13%. In contrast, the MODIS-based LAI exhibits no apparent spatial correlation with the LiDAR-based LAI. It is suggested that the main algorithm of MODIS should be improved with regard to tropical forests. The significance of this study is the proposal of a framework to produce ground-based LAI using forest inventory data and determine the plot-level LAI at the airborne and satellite scale using LiDAR data.

Download Full-text

Estimation of forest structure and canopy fuel parameters from small-footprint full-waveform LiDAR data

International Journal of Wildland Fire ◽

10.1071/wf13086 ◽

2014 ◽

Vol 23 (2) ◽

pp. 224 ◽

Cited By ~ 51

Author(s):

Txomin Hermosilla ◽

Luis A. Ruiz ◽

Alexandra N. Kazakova ◽

Nicholas C. Coops ◽

L. Monika Moskal

Keyword(s):

Forest Structure ◽

Forest Canopy ◽

Structural Parameters ◽

Mixed Forest ◽

Lidar Data ◽

Full Waveform ◽

Precise Knowledge ◽

Small Footprint ◽

Waveform Lidar ◽

Full Waveform Lidar

Precise knowledge of fuel conditions is important for predicting fire hazards and simulating fire growth and intensity across the landscape. We present a methodology to retrieve and map forest canopy fuel and other forest structural parameters using small-footprint full-waveform airborne light detection and ranging (LiDAR) data. Full-waveform LiDAR sensors register the complete returned backscattered signal through time and can describe physical properties of the intercepted objects. This study was undertaken in a mixed forest dominated by Douglas-fir, occasionally mixed with other conifers, in north-west Oregon (United States). We extracted two sets of LiDAR metrics using pulse detection and waveform modelling and then constructed several predictive models using forward stepwise multiple linear regression. The resulting models explained ~80% of the variability for many of the canopy fuel and forest structure parameters: aboveground biomass (R2=0.84), quadratic mean diameter (R2=0.82), canopy height (R2=0.79), canopy base height (R2=0.78) and canopy fuel load (R2=0.79). The lowest performing models included basal area (R2=0.76), stand volume (R2=0.73), canopy bulk density (R2=0.67) and stand density index (R2=0.66). Our results indicate that full-waveform LiDAR systems show promise in systematically characterising the structure and canopy fuel loads of forests, which may enable accurate fire behaviour forecasting that in turn supports the development of prevention and planning policies.

Download Full-text