Synergy of ICESat-2 and Landsat for Mapping Forest Aboveground Biomass with Deep Learning

Spatially continuous estimates of forest aboveground biomass (AGB) are essential to supporting the sustainable management of forest ecosystems and providing invaluable information for quantifying and monitoring terrestrial carbon stocks. The launch of the Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) on September 15th, 2018 offers an unparalleled opportunity to assess AGB at large scales using along-track samples that will be provided during its three-year mission. The main goal of this study was to investigate deep learning (DL) neural networks for mapping AGB with ICESat-2, using simulated photon-counting lidar (PCL)-estimated AGB for daytime, nighttime, and no noise scenarios, Landsat imagery, canopy cover, and land cover maps. The study was carried out in Sam Houston National Forest located in south-east Texas, using a simulated PCL-estimated AGB along two years of planned ICESat-2 profiles. The primary tasks were to investigate and determine neural network architecture, examine the hyper-parameter settings, and subsequently generate wall-to-wall AGB maps. A first set of models were developed using vegetation indices calculated from single-date Landsat imagery, canopy cover, and land cover, and a second set of models were generated using metrics from one year of Landsat imagery with canopy cover and land cover maps. To compare the effectiveness of final models, comparisons with Random Forests (RF) models were made. The deep neural network (DNN) models achieved R2 values of 0.42, 0.49, and 0.50 for the daytime, nighttime, and no noise scenarios respectively. With the extended dataset containing metrics calculated from Landsat images acquired on different dates, substantial improvements in model performance for all data scenarios were noted. The R2 values increased to 0.64, 0.66, and 0.67 for the daytime, nighttime, and no noise scenarios. Comparisons with Random forest (RF) prediction models highlighted similar results, with the same R2 and root mean square error (RMSE) range (15–16 Mg/ha) for daytime and nighttime scenarios. Findings suggest that there is potential for mapping AGB using a combinatory approach with ICESat-2 and Landsat-derived products with DL.

[1]  Lijuan Liu,et al.  A survey of remote sensing-based aboveground biomass estimation methods in forest ecosystems , 2016, Int. J. Digit. Earth.

[2]  S. Popescu.  Estimating biomass of individual pine trees using airborne lidar , 2007 .

[3]  Amy L. Neuenschwander,et al.  The ATL08 land and vegetation product for the ICESat-2 Mission , 2019, Remote Sensing of Environment.

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

[5]  D. Harding,et al.  ICESat waveform measurements of within‐footprint topographic relief and vegetation vertical structure , 2005 .

[6]  Paul R. Stysley,et al.  Laser production for NASA's Global Ecosystem Dynamics Investigation (GEDI) lidar , 2016, Defense + Security.

[7]  Yunjin Kim,et al.  An update on the NASA-ISRO dual-frequency DBF SAR (NISAR) mission , 2016, 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[8]  A. Baccini,et al.  Mapping forest canopy height globally with spaceborne lidar , 2011 .

[9]  Amy L. Neuenschwander,et al.  Estimating aboveground biomass and forest canopy cover with simulated ICESat-2 data , 2019, Remote Sensing of Environment.

[10]  Lana L. Narine,et al.  Mapping forest aboveground biomass with a simulated ICESat-2 vegetation canopy product and Landsat data , 2019, Annals of Forest Research.

[11]  Klaus Scipal,et al.  Coverage of high biomass forests by the ESA BIOMASS mission under defense restrictions , 2017 .

[12]  Guang-Bin Huang,et al.  Learning capability and storage capacity of two-hidden-layer feedforward networks , 2003, IEEE Trans. Neural Networks.

[13]  Sangmok Lee,et al.  Improved Prediction of Harmful Algal Blooms in Four Major South Korea’s Rivers Using Deep Learning Models , 2018, International journal of environmental research and public health.

[14]  Liang Wang,et al.  Forest Above Ground Biomass Estimation from Remotely Sensed Imagery in the Mount Tai Area Using the RBF ANN Algorithm , 2018 .

[15]  Maggi Kelly,et al.  Lidar with multi-temporal MODIS provide a means to upscale predictions of forest biomass , 2015 .

[16]  S. Popescu,et al.  Satellite lidar vs. small footprint airborne lidar: Comparing the accuracy of aboveground biomass estimates and forest structure metrics at footprint level , 2011 .

[17]  Lori A. Magruder,et al.  The Potential Impact of Vertical Sampling Uncertainty on ICESat-2/ATLAS Terrain and Canopy Height Retrievals for Multiple Ecosystems , 2016, Remote. Sens..

[18]  J. Blair,et al.  Modeling laser altimeter return waveforms over complex vegetation using high‐resolution elevation data , 1999 .

[19]  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 .

[20]  Michael A. Lefsky,et al.  Revised method for forest canopy height estimation from Geoscience Laser Altimeter System waveforms , 2007 .

[21]  Guoqing Sun,et al.  The uncertainty of biomass estimates from modeled ICESat-2 returns across a boreal forest gradient , 2015 .

[22]  Cardona Alzate,et al.  Predicción y selección de variables con bosques aleatorios en presencia de variables correlacionadas , 2020 .

[23]  Keirith A. Snyder,et al.  Comparison of Landsat and Land-Based Phenology Camera Normalized Difference Vegetation Index (NDVI) for Dominant Plant Communities in the Great Basin , 2019, Sensors.

[24]  Jin Liu,et al.  Mapping Global Forest Aboveground Biomass with Spaceborne LiDAR, Optical Imagery, and Forest Inventory Data , 2016, Remote. Sens..

[25]  C. Woodcock,et al.  Forest biomass estimation over regional scales using multisource data , 2004 .

[26]  Wei Yu,et al.  A Survey of Deep Learning: Platforms, Applications and Emerging Research Trends , 2018, IEEE Access.

[27]  Xiaoli Sun.  Space-based lidar systems , 2012, 2012 Conference on Lasers and Electro-Optics (CLEO).

[28]  José Cristóbal Riquelme Santos,et al.  A Preliminary Study of the Suitability of Deep Learning to Improve LiDAR-Derived Biomass Estimation , 2016, HAIS.

[29]  Xiao Xiang Zhu,et al.  Deep learning in remote sensing: a review , 2017, ArXiv.

[30]  S. Goetz,et al.  Advances in remote sensing technology and implications for measuring and monitoring forest carbon stocks and change , 2011 .

[31]  David J. Harding,et al.  Prospects of the ICESat-2 laser altimetry mission for savanna ecosystem structural studies based on airborne simulation data , 2016 .

[32]  Alicia Troncoso Lora,et al.  A comparison of machine learning regression techniques for LiDAR-derived estimation of forest variables , 2015, Neurocomputing.

[33]  Amy L. Neuenschwander,et al.  Applicability of an automatic surface detection approach to micro-pulse photon-counting lidar altimetry data: implications for canopy height retrieval from future ICESat-2 data , 2014 .

[34]  David J. Harding,et al.  The Ice, Cloud, and land Elevation Satellite-2 (ICESat-2): Science requirements, concept, and implementation , 2017 .

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

[36]  Arnt-Børre Salberg,et al.  Tree species classification in Norway from airborne hyperspectral and airborne laser scanning data , 2018 .

[37]  Hong Chi,et al.  National Forest Aboveground Biomass Mapping from ICESat/GLAS Data and MODIS Imagery in China , 2015, Remote. Sens..

[38]  M. Lefsky.  A global forest canopy height map from the Moderate Resolution Imaging Spectroradiometer and the Geoscience Laser Altimeter System , 2010 .

[39]  Yoshua. Bengio,et al.  Learning Deep Architectures for AI , 2007, Found. Trends Mach. Learn..

[40]  Sidik Mulyono,et al.  Multiple regularizations deep learning for paddy growth stages classification from LANDSAT-8 , 2016, 2016 International Conference on Advanced Computer Science and Information Systems (ICACSIS).

[41]  Jamal Ahmad Dargham,et al.  Finding the number of hidden neurons for an MLP neural network using coarse to fine search technique , 2010, 10th International Conference on Information Science, Signal Processing and their Applications (ISSPA 2010).

[42]  R. Nelson,et al.  Lidar-based estimates of aboveground biomass in the continental US and Mexico using ground, airborne, and satellite observations , 2017 .

[43]  Suming Jin,et al.  Completion of the 2011 National Land Cover Database for the Conterminous United States – Representing a Decade of Land Cover Change Information , 2015 .

[44]  Demetris Stathakis,et al.  How many hidden layers and nodes? , 2009 .

[45]  Amy L. Neuenschwander,et al.  Photon counting LiDAR: An adaptive ground and canopy height retrieval algorithm for ICESat-2 data , 2018 .

[46]  Xinghua Shi,et al.  A deep auto-encoder model for gene expression prediction , 2017, BMC Genomics.

[47]  Giles M. Foody,et al.  Mapping the biomass of Bornean tropical rain forest from remotely sensed data , 2001 .

[48]  Bin Zhang,et al.  Research on Object Detection Algorithm Based on PVANet , 2019 .

[49]  Nancy F. Glenn,et al.  Landsat 8 and ICESat-2: Performance and potential synergies for quantifying dryland ecosystem vegetation cover and biomass , 2016 .

[50]  Jindi Wang,et al.  Comparison and Evaluation of Three Methods for Estimating Forest above Ground Biomass Using TM and GLAS Data , 2017, Remote. Sens..

[51]  Mohammed A.-M. Salem,et al.  Deep Learning for Satellite Image Classification , 2018, AISI.

[52]  R. Houghton,et al.  Characterizing 3D vegetation structure from space: Mission requirements , 2011 .

[53]  Charmien Johnson Msfc.  Ice, Cloud, and Land Elevation Satellite , 2013 .

[54]  Nicolas Baghdadi,et al.  Testing Different Methods of Forest Height and Aboveground Biomass Estimations From ICESat/GLAS Data in Eucalyptus Plantations in Brazil , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[55]  Lei Wang,et al.  Stacked Sparse Autoencoder Modeling Using the Synergy of Airborne LiDAR and Satellite Optical and SAR Data to Map Forest Above-Ground Biomass , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[56]  Ricardo Alberto Cavieses Núñez,et al.  Deep learning models for the prediction of small-scale fisheries catches: finfish fishery in the region of “Bahía Magadalena-Almejas” , 2018, ICES Journal of Marine Science.

[57]  François Chollet,et al.  Deep Learning with Python , 2017 .