scholarly journals Exploring the Relationship between Forest Canopy Height and Canopy Density from Spaceborne LiDAR Observations

2021 ◽  
Vol 13 (24) ◽  
pp. 4961
Author(s):  
Heather Kay ◽  
Maurizio Santoro ◽  
Oliver Cartus ◽  
Pete Bunting ◽  
Richard Lucas
Keyword(s):  
Forest Structure ◽  
Forest Canopy ◽  
Mean Squared Error ◽  
Canopy Height ◽  
Single Parameter ◽  
Maximum Height ◽  
Allometric Model ◽  
Laser Altimeter ◽  
Canopy Density ◽  
The Relationship

Forest structure is a useful proxy for carbon stocks, ecosystem function and species diversity, but it is not well characterised globally. However, Earth observing sensors, operating in various modes, can provide information on different components of forests enabling improved understanding of their structure and variations thereof. The Ice, Cloud and Elevation Satellite (ICESat) Geoscience Laser Altimeter System (GLAS), providing LiDAR footprints from 2003 to 2009 with close to global coverage, can be used to capture elements of forest structure. Here, we evaluate a simple allometric model that relates global forest canopy height (RH100) and canopy density measurements to explain spatial patterns of forest structural properties. The GLA14 data product (version 34) was applied across subdivisions of the World Wildlife Federation ecoregions and their statistical properties were investigated. The allometric model was found to correspond to the ICESat GLAS metrics (median mean squared error, MSE: 0.028; inter-quartile range of MSE: 0.022–0.035). The relationship between canopy height and density was found to vary across biomes, realms and ecoregions, with denser forest regions displaying a greater increase in canopy density values with canopy height, compared to sparser or temperate forests. Furthermore, the single parameter of the allometric model corresponded with the maximum canopy density and maximum height values across the globe. The combination of the single parameter of the allometric model, maximum canopy density and maximum canopy height values have potential application in frameworks that target the retrieval of above-ground biomass and can inform on both species and niche diversity, highlighting areas for conservation, and potentially enabling the characterisation of biophysical drivers of forest structure.

scholarly journals Regional Variation in Forest Canopy Height and Implications for Koala (Phascolarctos cinereus) Habitat Mapping and Forest Management

Forests ◽  
2021 ◽  
Vol 12 (11) ◽  
pp. 1494
Author(s):  
Dave L. Mitchell ◽  
Mariela Soto-Berelov ◽  
Simon D. Jones
Keyword(s):  
Habitat Management ◽  
Forest Canopy ◽  
Regional Scale ◽  
Tree Height ◽  
Canopy Height ◽  
Forest Community ◽  
Maximum Height ◽  
Phascolarctos Cinereus ◽  
Old Growth Forest ◽  
Eastern Australia

Previous research has shown that the Koala (Phascolarctos cinereus) prefers larger trees, potentially making this a key factor influencing koala habitat quality. Generally, tree height is considered at regional scales which may overlook variation at patch or local scales. In this study, we aimed to derive a set of parameters to assist in classifying koala habitat in terms of tree height, which can then be used as an overlay for existing habitat maps. To determine canopy height variation within a specific forest community across a broad area in eastern Australia, we used freely available Airborne Laser Scanning (ALS) data and adopted a straightforward approach by extracting maximum-height ALS returns within a total of 288 30 m × 30 m “virtual” ALS plots. Our findings show that while maximum tree heights generally fall within published regional-scale parameters (mean height 33.2 m), they vary significantly between subregions (mean height 28.8–39.0 m), within subregions (e.g., mean height 21.3–29.4 m), and at local scales, the tree heights vary in response to previous land-use (mean height 28.0–34.2 m). A canopy height dataset useful for habitat management needs to recognise and incorporate these variations. To examine how this information might be synthesised into a usable map, we used a wall-to-wall canopy height map derived from ALS to investigate spatial and nonspatial clustering techniques that capture canopy height variability at both intra-subregional (100s of hectares) and local (60 hectare) scales. We found that nonspatial K-medians clustering with three or four height classes is suited to intra-subregional extents because it allows for simultaneous assessment and comparison of multiple forest community polygons. Spatially constrained clustering algorithms are suited to individual polygons, and we recommend the use of the Redcap algorithm because it delineates contiguous height classes recognisable on a map. For habitat management, an overlay combining these height classification approaches as separate attributes would provide the greatest utility at a range of scales. In addition to koala habitat management, canopy height maps could also assist in managing other fauna; identifying forest disturbance, regenerating forest, and old-growth forest; and identifying errors in existing forest maps.

scholarly journals Using GEDI Waveforms for Improved TanDEM-X Forest Height Mapping: A Combined SINC + Legendre Approach

2021 ◽  
Vol 13 (15) ◽  
pp. 2882
Author(s):  
Hao Chen ◽  
Shane R. Cloude ◽  
Joanne C. White
Keyword(s):  
High Resolution ◽  
Forest Canopy ◽  
Test Site ◽  
Convergent Series ◽  
Canopy Height ◽  
Ecosystem Dynamics ◽  
Sinc Function ◽  
Vertical Profiles ◽  
Legendre Series ◽  
Vertical Profiling

In this paper, we consider a new method for forest canopy height estimation using TanDEM-X single-pass radar interferometry. We exploit available information from sample-based, space-borne LiDAR systems, such as the Global Ecosystem Dynamics Investigation (GEDI) sensor, which offers high-resolution vertical profiling of forest canopies. To respond to this, we have developed a new extended Fourier-Legendre series approach for fusing high-resolution (but sparsely spatially sampled) GEDI LiDAR waveforms with TanDEM-X radar interferometric data to improve wide-area and wall-to-wall estimation of forest canopy height. Our key methodological development is a fusion of the standard uniform assumption for the vertical structure function (the SINC function) with LiDAR vertical profiles using a Fourier-Legendre approach, which produces a convergent series of approximations of the LiDAR profiles matched to the interferometric baseline. Our results showed that in our test site, the Petawawa Research Forest, the SINC function is more accurate in areas with shorter canopy heights (<~27 m). In taller forests, the SINC approach underestimates forest canopy height, whereas the Legendre approach avails upon simulated GEDI forest structural vertical profiles to overcome SINC underestimation issues. Overall, the SINC + Legendre approach improved canopy height estimates (RMSE = 1.29 m) compared to the SINC approach (RMSE = 4.1 m).

scholarly journals Tropical forest canopy height estimation from combined polarimetric SAR and LiDAR using machine-learning

2021 ◽  
Vol 172 ◽  
pp. 79-94
Author(s):  
Maryam Pourshamsi ◽  
Junshi Xia ◽  
Naoto Yokoya ◽  
Mariano Garcia ◽  
Marco Lavalle ◽  
...  
Keyword(s):  
Machine Learning ◽  
Tropical Forest ◽  
Forest Canopy ◽  
Canopy Height ◽  
Polarimetric Sar ◽  
Height Estimation ◽  
Tropical Forest Canopy

scholarly journals Canopy Height Estimation at Landsat Resolution Using Convolutional Neural Networks

2020 ◽  
Vol 2 (1) ◽  
pp. 23-36
Author(s):  
Syed Aamir Ali Shah ◽  
Muhammad Asif Manzoor ◽  
Abdul Bais
Keyword(s):  
Neural Networks ◽  
Random Forest ◽  
Convolutional Neural Networks ◽  
Forest Structure ◽  
State Of The Art ◽  
Spatial Association ◽  
Canopy Height ◽  
Landsat Images ◽  
Structure Estimation ◽  
Vegetation Height

Forest structure estimation is very important in geological, ecological and environmental studies. It provides the basis for the carbon stock estimation and effective means of sequestration of carbon sources and sinks. Multiple parameters are used to estimate the forest structure like above ground biomass, leaf area index and diameter at breast height. Among all these parameters, vegetation height has unique standing. In addition to forest structure estimation it provides the insight into long term historical changes and the estimates of stand age of the forests as well. There are multiple techniques available to estimate the canopy height. Light detection and ranging (LiDAR) based methods, being the accurate and useful ones, are very expensive to obtain and have no global coverage. There is a need to establish a mechanism to estimate the canopy height using freely available satellite imagery like Landsat images. Multiple studies are available which contribute in this area. The majority use Landsat images with random forest models. Although random forest based models are widely used in remote sensing applications, they lack the ability to utilize the spatial association of neighboring pixels in modeling process. In this research work, we define Convolutional Neural Network based model and analyze that model for three test configurations. We replicate the random forest based setup of Grant et al., which is a similar state-of-the-art study, and compare our results and show that the convolutional neural networks (CNN) based models not only capture the spatial association of neighboring pixels but also outperform the state-of-the-art.

scholarly journals Retrieving Forest Canopy Elements Clumping Index Using ICESat GLAS Lidar Data

2021 ◽  
Vol 13 (5) ◽  
pp. 948
Author(s):  
Lei Cui ◽  
Ziti Jiao ◽  
Kaiguang Zhao ◽  
Mei Sun ◽  
Yadong Dong ◽  
...  
Keyword(s):  
Large Scale ◽  
Forest Canopy ◽  
Remote Sensing Data ◽  
Alternative Methods ◽  
High Signal ◽  
Laser Altimeter ◽  
Waveform Data ◽  
Forest Park ◽  
Clumping Index ◽  
The Impact

Clumping index (CI) is a canopy structural variable important for modeling the terrestrial biosphere, but its retrieval from remote sensing data remains one of the least reliable. The majority of regional or global CI products available so far were generated from multiangle optical reflectance data. However, these reflectance-based estimates have well-known limitations, such as the mere use of a linear relationship between the normalized difference hotspot and darkspot (NDHD) and CI, uncertainties in bidirectional reflectance distribution function (BRDF) models used to calculate the NDHD, and coarse spatial resolutions (e.g., hundreds of meters to several kilometers). To remedy these limitations and develop alternative methods for large-scale CI mapping, here we explored the use of spaceborne lidar—the Geoscience Laser Altimeter System (GLAS)—and proposed a semi-physical algorithm to estimate CI at the footprint level. Our algorithm was formulated to leverage the full vertical canopy profile information of the GLAS full-waveform data; it converted raw waveforms to forest canopy gap distributions and gap fractions of random canopies, which was used to estimate CI based on the radiative transfer theory and a revised Beer–Lambert model. We tested our algorithm over two areas in China—the Saihanba National Forest Park and Heilongjiang Province—and assessed its relative accuracies against field-measured CI and MODIS CI products. We found that reliable estimation of CI was possible only for GLAS waveforms with high signal-to-noise ratios (e.g., >65) and at gentle slopes (e.g., <12°). Our GLAS-based CI estimates for high-quality waveforms compared well to field-based CI (i.e., R2 = 0.72, RMSE = 0.07, and bias = 0.02), but they showed less correlation to MODIS CI (e.g., R2 = 0.26, RMSE = 0.12, and bias = 0.04). The difference highlights the impact of the scale effect in conducting comparisons of products with huge differences resolution. Overall, our analyses represent the first attempt to use spaceborne lidar to retrieve high-resolution forest CI and our algorithm holds promise for mapping CI globally.

Sign in / Sign up

Export Citation Format

Share Document