The potential of multifrequency SAR images for estimating forest biomass in Mediterranean areas

Santi, Emanuele; Paloscia, S.; Pettinato, S.; Fontanelli, Giacomo; Mura, Matteo; Zolli, Catherine; Maselli, Fabio; Chiesi, Marta; Bottai, Lorenzo; Chirici, Gherardo

doi:10.1016/j.rse.2017.07.038

Cited by 93 publications

(57 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, the widespread usage of optical data is limited by its poor penetration capacity through clouds and forest canopies, as well as data saturation problems [21]. Radar data, available internationally from airborne or space-borne systems with different frequency bands, polarizations, and variable imaging geometries, such as Terra-SAR (Terra-Synthetic Aperture Radar), advanced land observing satellite phased array L-band synthetic aperture radar (ALOS PALSAR), and Sentinel, have gained prominence for AGB estimation because of their better penetration ability and detailed vegetation structural information, but these still suffer from signal saturation problems [6,14,22]. LiDAR, an active remote-sensing technology, captures forests' vertical structures in great detail and provides 3D information, such as the geoscience laser altimeter system (GLAS), which has found favor in biomass estimation with an improved accuracy, but with complex data-processing, and the lack of space continuity problems [23,24].…”

mentioning

confidence: 99%

Estimation of Forest Above-Ground Biomass by Geographically Weighted Regression and Machine Learning with Sentinel Imagery

Lin

Ren

Zhang

et al. 2018

Forests

106

View full text Add to dashboard Cite

Accurate forest above-ground biomass (AGB) is crucial for sustaining forest management and mitigating climate change to support REDD+ (reducing emissions from deforestation and forest degradation, plus the sustainable management of forests, and the conservation and enhancement of forest carbon stocks) processes. Recently launched Sentinel imagery offers a new opportunity for forest AGB mapping and monitoring. In this study, texture characteristics and backscatter coefficients of Sentinel-1, in addition to multispectral bands, vegetation indices, and biophysical variables of Sentinal-2, based on 56 measured AGB samples in the center of the Changbai Mountains, China, were used to develop biomass prediction models through geographically weighted regression (GWR) and machine learning (ML) algorithms, such as the artificial neural network (ANN), support vector machine for regression (SVR), and random forest (RF). The results showed that texture characteristics and vegetation biophysical variables were the most important predictors. SVR was the best method for predicting and mapping the patterns of AGB in the study site with limited samples, whose mean error, mean absolute error, root mean square error, and correlation coefficient were 4 × 10 −3 , 0.07, 0.08 Mg·ha −1 , and 1, respectively. Predicted values of AGB from four models ranged from 11.80 to 324.12 Mg·ha −1 , and those for broadleaved deciduous forests were the most accurate, while those for AGB above 160 Mg·ha −1 were the least accurate. The study demonstrated encouraging results in forest AGB mapping of the normal vegetated area using the freely accessible and high-resolution Sentinel imagery, based on ML techniques.Traditional field-based measurements provide the most accurate AGB values, but they are destructive and spatially limited [10,11]. Uncertainty and bias in field measurements obviously exist, particularly those with large trees and tropical issues [4,5]. Combining remote sensing and sample plot data has become a popular method to generate spatially explicit estimations of forest AGB [12,13]. Various types of remote-sensing data are used for forest biomass estimation such as optical sensor data, radio detection and ranging (radar) data, and light detection and ranging (LiDAR) data, with each one having certain advantages over the others [14,15]. Optical sensors were first applied to retrieve the horizontal forest structure and AGB assessments through field sampling, due to their aggregate spectral signatures (reflectance or vegetation indices) with global coverage, repetitiveness, and cost-effectiveness [16,17]. Optical remote sensing data from a number of platforms, such as IKONOS, Quickbird, Worldview, ZY-3, systeme probatoire d'observation de la terre (SPOT), Sentinel, Landsat, and moderate-resolution imaging spectroradiometer (MODIS), with spatial resolutions varying from less than one meter to hundreds of meters, have been used by numerous researchers for biomass estimation [18][19][20]. However, the widespread usage of optical data is limited...

show abstract

mentioning

confidence: 99%

Estimation of Forest Above-Ground Biomass by Geographically Weighted Regression and Machine Learning with Sentinel Imagery

Lin

Ren

Zhang

et al. 2018

Forests

106

View full text Add to dashboard Cite

show abstract

“…Os estoques de carbono florestal são geralmente calculados utilizando a Biomassa Acima do Solo (AGB), assumindo que aproximadamente 50% da AGB é carbono (GOETZ & DUBAYAH, 2011). Com isso, o uso do sensoriamento remoto permite a quantificação da AGB e do carbono para uma larga escala espacial, com resultados promissores quando utilizadas imagens multiespectrais ou hiperespectrais (LAURIN et al, 2014;DEO et al, 2017;PHUA et al, 2017), imagens de Radar de Abertura Sintética (SAR) (CASTILLO et al, 2017;SANTI et al, 2017;NINGTHOUJAM et al, 2018) e dados Light Detection and Ranging (LiDAR) (D'OLIVEIRA et al, 2012;NIE et al, 2017;BAZEZEW et al, 2018).…”

Section: Introductionunclassified

Regressões Robusta E Linear Para Estimativa De Biomassa via Imagem Sentinel Em Uma Floresta Tropical

Debastiani¹,

Moura²,

Rex³

et al. 2019

BIOFIX

View full text Add to dashboard Cite

RESUMOA preocupação com as mudanças climáticas globais tem motivado diversos pesquisadores a encontrar métodos eficazes para a quantificação de biomassa florestal e carbono estocado em florestas tropicais, uma vez que, estas atuam de forma mitigatória e compensatória desses efeitos. O sensoriamento remoto tem sido utilizado de forma eficaz e com grande potencial para a estimativas em larga escala, com destaque para dados de Radar de Abertura Sintética (SAR) e imagens multiespectrais. Os estudos desenvolvidos com essa finalidade utilizaram diversas técnicas para associar a biomassa acima do solo (AGB) com os dados obtidos por sensoriamento remoto, entretanto, a aplicação da regressão robusta ainda não está sendo utilizada para tal finalidade. Sendo assim, o objetivo do presente estudo é avaliar o desempenho da regressão robusta comparando com a regressão linear, que é tradicionalmente utilizada, além de avaliar o potencial da utilização dos dados oriundos do satélite Sentinel 1 e 2 para a estimativa de AGB. Neste âmbito, foram utilizadas imagens multiespectrais (Sentinel 2), imagem SAR (Sentinel 1) e como variável resposta a AGB obtida a partir de dados Light Detection and Ranging (LiDAR). A AGB foi estimada por dois métodos de regressão: robusta e linear. Os modelos de regressão robusta e linear apresentaram desempenho semelhante, com R²aj. variando entre 0,33 a 0,34, erro padrão da estimativa de 48 Mg.ha -1 e raiz do erro médio quadrático de 16%. Conclui-se que não houve diferença significativa entre a regressão linear e a regressão robusta para esse conjunto de dados, indicando que a regressão robusta não se sobressaiu com a presença de outliers e que existe potencial na utilização de dados oriundos do satélite Sentinel. ABSTRACTConcern about global climate change has motivated several researchers to find effective methods for the quantification of forest biomass and carbon stored in tropical forests, since they act in a mitigating and compensatory way. Remote sensing has been used effectively and with great potential for large scale estimation, with emphasis on Synthetic Aperture Radar (SAR) data and multispectral images. The studies already developed for this purpose have used several techniques to associate above-ground biomass (AGB) with the data obtained by remote sensing, however, the application of robust regression is not yet being used for this purpose. Thus, the objective of the present study is to evaluate the robust regression performance in comparison with the linear regression, that is traditionally used, besides evaluating the potential of the data from Sentinel 1 and 2 satellite. In this context, multispectral images (Sentinel 2), SAR image (Sentinel 1) and as variable response to AGB obtained from data Light Detection And Ranging (LiDAR). AGB was estimated by two regression methods: robust and linear. The robust and linear regression models presented similar performance, with R²aj. ranging from 0.33 to 0.34, standard error of the estimate of 48 Mg.ha -1 and root mean square error of 16%. It is ...

show abstract

“…Not all vegetation indices are closely related with biomass. The widely used Synthetic Aperture Radar (SAR) (Santi, et al, 2017) and LIght Detection And Ranging (LIDAR) (Lei Zhang & Grift, 2012), (Andújar, et al, 2016) sensors offer a better tolerance to weather and light conditions and are capable of collecting three dimensional distribution of structures within vegetation. Thus, they allow precise analysis on the characteristics of vegetation including biomass.…”

Section: Introductionmentioning

confidence: 99%

Density Weighted Connectivity of Grass Pixels in image frames for biomass estimation

Li-gang¹,

Verma²,

Stockwell³

et al. 2018

Expert Systems with Applications

View full text Add to dashboard Cite

Accurate estimation of the biomass of roadside grasses plays a significant role in applications such as fire-prone region identification. Current solutions heavily depend on field surveys, remote sensing measurements and image processing using reference markers, which often demand big investments of time, effort and cost. This paper proposes Density Weighted Connectivity of Grass Pixels (DWCGP) to automatically estimate grass biomass from roadside image data. The DWCGP calculates the length of continuously connected grass pixels along a vertical orientation in each image column, and then weights the length by the grass density in a surrounding region of the column. Grass pixels are classified using feedforward artificial neural networks and the dominant texture orientation at every pixel is computed using multiorientation Gabor wavelet filter vote. Evaluations on a field survey dataset show that the DWCGP reduces Root-Mean-Square Error from 5.84 to 5.52 by additionally considering grass density on top of grass height. The DWCGP shows robustness to non-vertical grass stems and to changes of both Gabor filter parameters and surrounding region widths. It also has performance close to human observation and higher than eight baseline approaches, as well as promising results for classifying low vs. high fire risk and identifying fire-prone road regions.

show abstract

The potential of multifrequency SAR images for estimating forest biomass in Mediterranean areas

Cited by 93 publications

References 61 publications

Estimation of Forest Above-Ground Biomass by Geographically Weighted Regression and Machine Learning with Sentinel Imagery

Estimation of Forest Above-Ground Biomass by Geographically Weighted Regression and Machine Learning with Sentinel Imagery

Regressões Robusta E Linear Para Estimativa De Biomassa via Imagem Sentinel Em Uma Floresta Tropical

Density Weighted Connectivity of Grass Pixels in image frames for biomass estimation

Contact Info

Product

Resources

About