Stacked Sparse Autoencoder Modeling Using the Synergy of Airborne LiDAR and Satellite Optical and SAR Data to Map Forest Above-Ground Biomass

Timely, spatially complete, and reliable forest above-ground biomass (AGB) data are a prerequisite to support forest management and policy formulation. Traditionally, forest AGB is spatially estimated by integrating satellite images, in particular, optical data, with field plots from forest inventory programs. However, field data are limited in remote and unmanaged areas. In addition, optical reflectance usually saturates at high-density biomass level and is subject to cloud contaminations. Thus, this study aimed to develop a deep learning based workflow for mapping forest AGB by integrating Landsat 8 and Sentinel-1A images with airborne light detection and ranging (LiDAR) data. A reference AGB map was derived from the wall-to-wall LiDAR data and field measurements. The LiDAR plots—stratified random samples of forest biomass extracted from the LiDAR simulated strips in the reference map—were adopted as a surrogate for traditional field plots. In addition to the deep learning model, i.e., stacked sparse autoencoder network (SSAE), five different prediction techniques including multiple stepwise linear regressions, K-nearest neighbor, support vector machine, back propagation neural networks, and random forest were individually used to establish the relationship between LiDAR-derived forest biomass and the satellite predictors. Optical variables (Landsat 8 OLI), SAR variables (Sentinel-1A), and their combined variables were individually input to the six prediction models. Results showed that the SSAE model had the best performance for the forest biomass estimation. The combined optical and microwave dataset as explanatory variables improved the modeling performance compared to either the optical-only or microwave-only data, regardless of prediction algorithms. The best mapping accuracy was obtained by the SSAE model with inputs of optical and microwave integrated metrics that yielded $R^{2}$ of 0.812, root mean squared error (RMSE) of 21.753 Mg/ha, and relative RMSE (RMSEr) of 14.457%. Overall, the SSAE model with inputs of combined Landsat 8 OLI and Sentinel-1A information could result in accurate estimation of forest biomass by using the stratification-sampled and LiDAR-derived AGB as ground reference data. The modeling workflow has the potential to promote future forest growth monitoring and carbon stock assessment across large areas.

[1]  Kuan-Tsung Chang,et al.  Estimation of carbon sequestration by using vegetation indices , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[2]  Gherardo Chirici,et al.  Comparing echo-based and canopy height model-based metrics for enhancing estimation of forest aboveground biomass in a model-assisted framework , 2016 .

[3]  Gang Chen,et al.  When Big Data are Too Much: Effects of LiDAR Returns and Point Density on Estimation of Forest Biomass , 2016, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[4]  Jianping Guo,et al.  Estimation of forest above-ground biomass using multi-parameter remote sensing data over a cold and arid area , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[5]  Juergen Rossmann,et al.  Using Decision Tree Based Multiclass Support Vector Machines for Forest Mapping , 2011 .

[6]  Demetrios Gatziolis,et al.  CoMPArISon oF lIDAr- AnD PHoToInTerPreTATIon-BASeD eSTIMATeS oF CAnoPy Cover , 2012 .

[7]  C. Gleason,et al.  A Review of Remote Sensing of Forest Biomass and Biofuel: Options for Small-Area Applications , 2011 .

[8]  Matti Maltamo,et al.  Airborne discrete-return LIDAR data in the estimation of vertical canopy cover, angular canopy closure and leaf area index , 2011 .

[9]  Gang Wang,et al.  Deep Learning-Based Classification of Hyperspectral Data , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[10]  Bangqian Chen,et al.  Spatio-temporal prediction of leaf area index of rubber plantation using HJ-1A/1B CCD images and recurrent neural network , 2015 .

[11]  G. Hay,et al.  A Support Vector Regression Approach to Estimate Forest Biophysical Parameters at the Object Level Using Airborne Lidar Transects and QuickBird Data , 2011 .

[12]  E. Næsset.  Predicting forest stand characteristics with airborne scanning laser using a practical two-stage procedure and field data , 2002 .

[13]  Oleg Antropov,et al.  Improved Mapping of Tropical Forests With Optical and SAR Imagery, Part II: Above Ground Biomass Estimation , 2013, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[14]  Marc Simard,et al.  Radarsat-2 Backscattering for the Modeling of Biophysical Parameters of Regenerating Mangrove Forests , 2015, Remote. Sens..

[15]  Michael A. Wulder,et al.  Characterizing stand-level forest canopy cover and height using Landsat time series, samples of airborne LiDAR, and the Random Forest algorithm , 2015 .

[16]  M. Transport,et al.  Change detection of coral reef habitat using Landsat-5 TM, Landsat 7 ETM+ and Landsat 8 OLI data in the Red Sea (Hurghada, Egypt) , 2014 .

[17]  Michael A. Wulder,et al.  Integration of GLAS and Landsat TM data for aboveground biomass estimation , 2010 .

[18]  Gábor Lugosi,et al.  Introduction to Statistical Learning Theory , 2004, Advanced Lectures on Machine Learning.

[19]  Sudhanshu Sekhar Panda,et al.  Application of Vegetation Indices for Agricultural Crop Yield Prediction Using Neural Network Techniques , 2010, Remote. Sens..

[20]  W. Cohen,et al.  Characterizing 23 Years (1972–95) of Stand Replacement Disturbance in Western Oregon Forests with Landsat Imagery , 2002, Ecosystems.

[21]  Nicholas C. Coops,et al.  Estimating the probability of mountain pine beetle red-attack damage , 2006 .

[22]  Ross K. Meentemeyer,et al.  Effects of LiDAR point density and landscape context on estimates of urban forest biomass , 2015 .

[23]  Zheng Niu,et al.  Geostatistical modeling using LiDAR-derived prior knowledge with SPOT-6 data to estimate temperate forest canopy cover and above-ground biomass via stratified random sampling , 2015, Int. J. Appl. Earth Obs. Geoinformation.

[24]  Dan Zhang,et al.  Stacked Sparse Autoencoder in PolSAR Data Classification Using Local Spatial Information , 2016, IEEE Geoscience and Remote Sensing Letters.

[25]  Joanne C. White,et al.  Lidar sampling for large-area forest characterization: A review , 2012 .

[26]  Cheng Wang,et al.  Automated Detection of Three-Dimensional Cars in Mobile Laser Scanning Point Clouds Using DBM-Hough-Forests , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[27]  J. Carreiras,et al.  Understanding the relationship between aboveground biomass and ALOS PALSAR data in the forests of Guinea-Bissau (West Africa) , 2012 .

[28]  A. Hudak,et al.  Nearest neighbor imputation of species-level, plot-scale forest structure attributes from LiDAR data , 2008 .

[29]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[30]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[31]  Jun Wu,et al.  A Hierarchical Oil Tank Detector With Deep Surrounding Features for High-Resolution Optical Satellite Imagery , 2015, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[32]  K. V. S. Badarinath,et al.  Analysis of ENVISAT ASAR data for forest parameter retrieval and forest type classification—a case study over deciduous forests of central India , 2007 .

[33]  I. Woodhouse,et al.  Measuring biomass changes due to woody encroachment and deforestation/degradation in a forest-savanna boundary region of central Africa using multi-temporal L-band radar backscatter , 2011 .

[34]  Shengli Tao,et al.  Spatial distribution of forest aboveground biomass in China: Estimation through combination of spaceborne lidar, optical imagery, and forest inventory data , 2015 .

[35]  R. Fournier,et al.  A comparison of four methods to map biomass from Landsat-TM and inventory data in western Newfoundland , 2006 .

[36]  Na Yin,et al.  Estimating Forest Aboveground Biomass by Combining ALOS PALSAR and WorldView-2 Data: A Case Study at Purple Mountain National Park, Nanjing, China , 2014, Remote. Sens..

[37]  Barbara Koch,et al.  Mapping forest biomass from space - Fusion of hyperspectral EO1-hyperion data and Tandem-X and WorldView-2 canopy height models , 2015, Int. J. Appl. Earth Obs. Geoinformation.

[38]  Saso Dzeroski,et al.  Estimating vegetation height and canopy cover from remotely sensed data with machine learning , 2010, Ecol. Informatics.

[39]  Nicholas C. Coops,et al.  Integrating airborne LiDAR and space-borne radar via multivariate kriging to estimate above-ground biomass , 2013 .

[40]  Richard Gloaguen,et al.  Improving the Estimation of Above Ground Biomass Using Dual Polarimetric PALSAR and ETM+ Data in the Hyrcanian Mountain Forest (Iran) , 2014, Remote. Sens..

[41]  O. Mutanga,et al.  Investigating the robustness of the new Landsat-8 Operational Land Imager derived texture metrics in estimating plantation forest aboveground biomass in resource constrained areas , 2015 .

[42]  B. Koch,et al.  Non-parametric prediction and mapping of standing timber volume and biomass in a temperate forest: application of multiple optical/LiDAR-derived predictors , 2010 .

[43]  Sean P. Healey,et al.  Application of two regression-based methods to estimate the effects of partial harvest on forest structure using Landsat data , 2006 .

[44]  N. Coops,et al.  Update of forest inventory data with lidar and high spatial resolution satellite imagery , 2008 .

[45]  Jason C. Neff,et al.  Estimates of Aboveground Biomass from Texture Analysis of Landsat Imagery , 2014, Remote. Sens..

[46]  Jonathan Li,et al.  Learning Hierarchical Features for Automated Extraction of Road Markings From 3-D Mobile LiDAR Point Clouds , 2015, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[47]  John R. Schott,et al.  Leveraging EO-1 to Evaluate Capability of New Generation of Landsat Sensors for Coastal/Inland Water Studies , 2013, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[48]  Terje Gobakken,et al.  Improved estimates of forest vegetation structure and biomass with a LiDAR‐optimized sampling design , 2009 .

[49]  Qi Li,et al.  Hyperspectral Imagery Classification Using Sparse Representations of Convolutional Neural Network Features , 2016, Remote. Sens..

[50]  Kunwar K. Singh,et al.  The impact of urban residential development patterns on forest carbon density: An integration of LiDAR, aerial photography and field mensuration , 2015 .

[51]  W. Cohen,et al.  Integration of lidar and Landsat ETM+ data for estimating and mapping forest canopy height , 2002 .

[52]  Florian Hartig,et al.  Importance of sample size, data type and prediction method for remote sensing-based estimations of aboveground forest biomass , 2014 .

[53]  Juha Hyyppä,et al.  Advances in Forest Inventory Using Airborne Laser Scanning , 2012, Remote. Sens..

[54]  Zhenfeng Shao,et al.  Estimating Forest Aboveground Biomass by Combining Optical and SAR Data: A Case Study in Genhe, Inner Mongolia, China , 2016, Sensors.

[55]  M. Wulder,et al.  Forest inventory height update through the integration of lidar data with segmented Landsat imagery , 2003 .

[56]  Barbara Koch,et al.  Evaluation of most similar neighbour and random forest methods for imputing forest inventory variables using data from target and auxiliary stands , 2012 .

[57]  Luis Alonso,et al.  Machine learning regression algorithms for biophysical parameter retrieval: Opportunities for Sentinel-2 and -3 , 2012 .

[58]  Janne Heiskanen,et al.  Modelling lidar-derived boreal forest canopy cover with SPOT 4 HRVIR data , 2013 .

[59]  O. Mutanga,et al.  Evaluating the utility of the medium-spatial resolution Landsat 8 multispectral sensor in quantifying aboveground biomass in uMgeni catchment, South Africa , 2015 .

[60]  Seiho Uratsuka,et al.  Season-long daily measurements of multifrequency (Ka, Ku, X, C, and L) and full-polarization backscatter signatures over paddy rice field and their relationship with biological variables , 2002 .

[61]  Luis Alonso,et al.  Multioutput Support Vector Regression for Remote Sensing Biophysical Parameter Estimation , 2011, IEEE Geoscience and Remote Sensing Letters.

[62]  Jungho Im,et al.  Forest Biomass and Carbon Stock Quantification Using Airborne LiDAR Data: A Case Study Over Huntington Wildlife Forest in the Adirondack Park , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[63]  P. Gessler,et al.  Forest Service-- National AgroforestryCenter 1-1-2010 Landscape-scale parameterization of a tree-level forest growth model : a k-nearest neighbor imputation approach incorporating LiDAR data , 2013 .

[64]  S. Goetz,et al.  Importance of biomass in the global carbon cycle , 2009 .

[65]  A. Skidmore,et al.  Estimating tropical forest biomass more accurately by integrating ALOS PALSAR and Landsat-7 ETM+ data , 2013 .

[66]  Harold S. J. Zald,et al.  Influence of lidar, Landsat imagery, disturbance history, plot location accuracy, and plot size on accuracy of imputation maps of forest composition and structure , 2014 .

[67]  Erxue Chen,et al.  Estimating montane forest above-ground biomass in the upper reaches of the Heihe River Basin using Landsat-TM data , 2014 .

[68]  Zheng Niu,et al.  Estimating the Leaf Area Index, height and biomass of maize using HJ-1 and RADARSAT-2 , 2013, Int. J. Appl. Earth Obs. Geoinformation.

[69]  D. Lu.  The potential and challenge of remote sensing‐based biomass estimation , 2006 .

[70]  Joanne C. White,et al.  Integrating Landsat pixel composites and change metrics with lidar plots to predictively map forest structure and aboveground biomass in Saskatchewan, Canada , 2016 .

[71]  Wenjian Ni,et al.  Forest stand biomass estimation using ALOS PALSAR data based on LiDAR-derived prior knowledge in the Qilian Mountain, western China , 2012 .