skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Arctic Vegetation Mapping Using Unsupervised Training Datasets and Convolutional Neural Networks

Abstract

Land cover datasets are essential for modeling and analysis of Arctic ecosystem structure and function and for understanding land–atmosphere interactions at high spatial resolutions. However, most Arctic land cover products are generated at a coarse resolution, often limited due to cloud cover, polar darkness, and poor availability of high-resolution imagery. A multi-sensor remote sensing-based deep learning approach was developed for generating high-resolution (5 m) vegetation maps for the western Alaskan Arctic on the Seward Peninsula, Alaska. The fusion of hyperspectral, multispectral, and terrain datasets was performed using unsupervised and supervised classification techniques over a ~343 km2 area, and a high-resolution (5 m) vegetation classification map was generated. An unsupervised technique was developed to classify high-dimensional remote sensing datasets into cohesive clusters. We employed a quantitative method to add supervision to the unlabeled clusters, producing a fully labeled vegetation map. We then developed convolutional neural networks (CNNs) using the multi-sensor fusion datasets to map vegetation distributions using the original classes and the classes produced by the unsupervised classification method. To validate the resulting CNN maps, vegetation observations were collected at 30 field plots during the summer of 2016, and the resulting vegetation products developed were evaluated against them for accuracy. Ourmore » analysis indicates the CNN models based on the labels produced by the unsupervised classification method provided the most accurate mapping of vegetation types, increasing the validation score (i.e., precision) from 0.53 to 0.83 when evaluated against field vegetation observations.« less

Authors:
 [1];  [1]; ORCiD logo [2];  [3];  [1]
  1. Univ. of Tennessee, Knoxville, TN (United States). Bredesen Center for Interdisciplinary Research and Graduate Education; Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). Environmental Sciences Division
  2. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). Computational Sciences and Engineering; Univ. of Tennessee, Knoxville, TN (United States). Dept. of Civil and Environmental Engineering
  3. Univ. of Alaska, Fairbanks, AK (United States). International Arctic Research Center
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Biological and Environmental Research (BER)
OSTI Identifier:
1489551
Grant/Contract Number:  
AC05-00OR22725
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Remote Sensing
Additional Journal Information:
Journal Volume: 11; Journal Issue: 1; Journal ID: ISSN 2072-4292
Publisher:
MDPI
Country of Publication:
United States
Language:
English
Subject:
54 ENVIRONMENTAL SCIENCES; 97 MATHEMATICS AND COMPUTING; hyperspectral; field-scale mapping; arctic; vegetation classification; convolutional neural network

Citation Formats

Langford, Zachary, Kumar, Jitendra, Hoffman, Forrest, Breen, Amy, and Iversen, Colleen. Arctic Vegetation Mapping Using Unsupervised Training Datasets and Convolutional Neural Networks. United States: N. p., 2019. Web. doi:10.3390/rs11010069.
Langford, Zachary, Kumar, Jitendra, Hoffman, Forrest, Breen, Amy, & Iversen, Colleen. Arctic Vegetation Mapping Using Unsupervised Training Datasets and Convolutional Neural Networks. United States. https://doi.org/10.3390/rs11010069
Langford, Zachary, Kumar, Jitendra, Hoffman, Forrest, Breen, Amy, and Iversen, Colleen. 2019. "Arctic Vegetation Mapping Using Unsupervised Training Datasets and Convolutional Neural Networks". United States. https://doi.org/10.3390/rs11010069. https://www.osti.gov/servlets/purl/1489551.
@article{osti_1489551,
title = {Arctic Vegetation Mapping Using Unsupervised Training Datasets and Convolutional Neural Networks},
author = {Langford, Zachary and Kumar, Jitendra and Hoffman, Forrest and Breen, Amy and Iversen, Colleen},
abstractNote = {Land cover datasets are essential for modeling and analysis of Arctic ecosystem structure and function and for understanding land–atmosphere interactions at high spatial resolutions. However, most Arctic land cover products are generated at a coarse resolution, often limited due to cloud cover, polar darkness, and poor availability of high-resolution imagery. A multi-sensor remote sensing-based deep learning approach was developed for generating high-resolution (5 m) vegetation maps for the western Alaskan Arctic on the Seward Peninsula, Alaska. The fusion of hyperspectral, multispectral, and terrain datasets was performed using unsupervised and supervised classification techniques over a ~343 km2 area, and a high-resolution (5 m) vegetation classification map was generated. An unsupervised technique was developed to classify high-dimensional remote sensing datasets into cohesive clusters. We employed a quantitative method to add supervision to the unlabeled clusters, producing a fully labeled vegetation map. We then developed convolutional neural networks (CNNs) using the multi-sensor fusion datasets to map vegetation distributions using the original classes and the classes produced by the unsupervised classification method. To validate the resulting CNN maps, vegetation observations were collected at 30 field plots during the summer of 2016, and the resulting vegetation products developed were evaluated against them for accuracy. Our analysis indicates the CNN models based on the labels produced by the unsupervised classification method provided the most accurate mapping of vegetation types, increasing the validation score (i.e., precision) from 0.53 to 0.83 when evaluated against field vegetation observations.},
doi = {10.3390/rs11010069},
url = {https://www.osti.gov/biblio/1489551}, journal = {Remote Sensing},
issn = {2072-4292},
number = 1,
volume = 11,
place = {United States},
year = {2019},
month = {1}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 2 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Classification for High Resolution Remote Sensing Imagery Using a Fully Convolutional Network
journal, May 2017


Road Segmentation of Remotely-Sensed Images Using Deep Convolutional Neural Networks with Landscape Metrics and Conditional Random Fields
journal, July 2017


Mapping Arctic Tundra Vegetation Communities Using Field Spectroscopy and Multispectral Satellite Data in North Alaska, USA
journal, November 2016


Parallel k-Means Clustering for Quantitative Ecoregion Delineation Using Large Data Sets
journal, January 2011


Finding Structure with Randomness: Probabilistic Algorithms for Constructing Approximate Matrix Decompositions
journal, January 2011


Deep learning
journal, May 2015


Vegetation and Permafrost Thaw Depth 10 Years after a Tundra Fire in 2002, Seward Peninsula, Alaska
journal, August 2015


Carbon cycling in the Arctic
journal, August 2014


Mapping Arctic Plant Functional Type Distributions in the Barrow Environmental Observatory Using WorldView-2 and LiDAR Datasets
journal, September 2016


Summer Differences among Arctic Ecosystems in Regional Climate Forcing
journal, June 2000


Projections of Twenty-First-Century Climate Extremes for Alaska via Dynamical Downscaling and Quantile Mapping
journal, September 2017


Hyperspectral and Multispectral Image Fusion via Deep Two-Branches Convolutional Neural Network
journal, May 2018


The Alaska Arctic Vegetation Archive (AVA-AK)
journal, September 2016


Differentiating among Four Arctic Tundra Plant Communities at Ivotuk, Alaska Using Field Spectroscopy
journal, January 2016


Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art
journal, June 2016


Regional Quantitative Cover Mapping of Tundra Plant Functional Types in Arctic Alaska
journal, October 2017


Convolutional Neural Network Approach for Mapping Arctic Vegetation Using Multi-Sensor Remote Sensing Fusion
conference, November 2017


Future Arctic climate changes: Adaptation and mitigation time scales
journal, February 2014


Shifts in Arctic vegetation and associated feedbacks under climate change
journal, March 2013


Multisource Remote Sensing Data Classification Based on Convolutional Neural Network
journal, February 2018


Estimating heterotrophic respiration at large scales: challenges, approaches, and next steps
journal, June 2016


Deep Fusion of Remote Sensing Data for Accurate Classification
journal, August 2017


Mapcurves: a quantitative method for comparing categorical maps
journal, May 2006


Hyperspectral Imagery Denoising by Deep Learning With Trainable Nonlinearity Function
journal, November 2017


Tundra shrubification and tree-line advance amplify arctic climate warming: results from an individual-based dynamic vegetation model
journal, August 2013


Remote Sensing-Based, 5-m, Vegetation Distributions, Kougarok Study Site, Seward Peninsula, Alaska, ca. 2009 - 2016
dataset, January 2019

  • Langford, Zachary; Kumar, Jitendra; Hoffman, Forrest
  • Next Generation Ecosystems Experiment - Arctic, Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (US); NGEE Arctic, Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
  • https://doi.org/10.5440/1418854

Response of subarctic vegetation to transient climatic change on the Seward Peninsula in north-west Alaska
journal, June 2000


Deriving Snow Cover Metrics for Alaska from MODIS
journal, September 2015


Frequent Fires in Ancient Shrub Tundra: Implications of Paleorecords for Arctic Environmental Change
journal, March 2008


Training deep neural networks on imbalanced data sets
conference, July 2016


Data Fusion and Remote Sensing: An ever-growing relationship
journal, December 2016