@article{Wu-2021-Assessment,
title = "Assessment of machine learning classifiers for global lake ice cover mapping from MODIS TOA reflectance data",
author = "Wu, Yuhao and
Duguay, Claude R. and
Xu, Linlin",
journal = "Remote Sensing of Environment, Volume 253",
volume = "253",
year = "2021",
publisher = "Elsevier BV",
url = "https://gwf-uwaterloo.github.io/gwf-publications/G21-29001",
doi = "10.1016/j.rse.2020.112206",
pages = "112206",
abstract = "The topic of satellite remote sensing of lake ice has gained considerable attention in recent years. Optical satellite data from the Moderate Resolution Imaging Spectroradiometer (MODIS) allow for the monitoring of lake ice cover (an Essential Climate Variable or ECV), and dates associated with ice phenology (freeze-up, break-up, and ice cover duration) over large areas in an era where ground-based observational networks have nearly vanished in many northern countries. Ice phenology dates as well as dates of maximum and minimum ice cover extent (for lakes that do not form a complete ice cover in winter or do not totally lose their ice cover in summer) are useful for assessing long-term trends and variability in climate, particularly due to their sensitivity to changes in near-surface air temperature. Existing knowledge-driven (threshold-based) retrieval algorithms for lake ice cover mapping that use top-of-atmosphere (TOA) reflectance products do not perform well under lower solar illumination conditions (i.e. large solar zenith angles), resulting in low TOA reflectance. This research assessed the capability of four machine learning classifiers (i.e. multinomial logistic regression, MLR; support vector machine, SVM; random forest, RF; gradient boosting trees, GBT) for mapping lake ice cover, water and cloud cover during both break-up and freeze-up periods using the MODIS/Terra L1B TOA (MOD02) product. The classifiers were trained and validated using samples collected from 17 large lakes across the Northern Hemisphere (Europe and North America); lakes that represent different characteristics with regards to area, latitude, freezing frequency, and ice duration. Following an accuracy assessment using random k-fold cross-validation (k = 100), all machine learning classifiers using a 7-band combination (visible, near-infrared and shortwave-infrared) were found to be able to produce overall classification accuracies above 94{\%}. Both RF and GBT provided overall and class-specific accuracies above 98{\%} and a more visually accurate depiction of lake ice, water and cloud cover. The two tree-based classifiers offered the most robust spatial transferability over the 17 lakes and performed consistently well across ice seasons. However, only RF was relatively insensitive to the choice of the hyperparameters compared to the other three classifiers. The results demonstrate the potential of RF for mapping lake ice cover globally from MODIS TOA reflectance data. {\mbox{$\bullet$}} This study assessed the capability of ML classifiers for lake ice mapping from MOIDS. {\mbox{$\bullet$}} RF and GBT produced the best performance in terms of classification accuracies. {\mbox{$\bullet$}} RF and GBT offered the most robust spatial and temporal transferability. {\mbox{$\bullet$}} RF was insensitive to the choice of the hyperparameters compared to other classifiers. {\mbox{$\bullet$}} The results show the potential of RF for mapping lake ice cover globally from MODIS.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="Wu-2021-Assessment">
<titleInfo>
<title>Assessment of machine learning classifiers for global lake ice cover mapping from MODIS TOA reflectance data</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yuhao</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Claude</namePart>
<namePart type="given">R</namePart>
<namePart type="family">Duguay</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Linlin</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<genre authority="bibutilsgt">journal article</genre>
<relatedItem type="host">
<titleInfo>
<title>Remote Sensing of Environment, Volume 253</title>
</titleInfo>
<originInfo>
<issuance>continuing</issuance>
<publisher>Elsevier BV</publisher>
</originInfo>
<genre authority="marcgt">periodical</genre>
<genre authority="bibutilsgt">academic journal</genre>
</relatedItem>
<abstract>The topic of satellite remote sensing of lake ice has gained considerable attention in recent years. Optical satellite data from the Moderate Resolution Imaging Spectroradiometer (MODIS) allow for the monitoring of lake ice cover (an Essential Climate Variable or ECV), and dates associated with ice phenology (freeze-up, break-up, and ice cover duration) over large areas in an era where ground-based observational networks have nearly vanished in many northern countries. Ice phenology dates as well as dates of maximum and minimum ice cover extent (for lakes that do not form a complete ice cover in winter or do not totally lose their ice cover in summer) are useful for assessing long-term trends and variability in climate, particularly due to their sensitivity to changes in near-surface air temperature. Existing knowledge-driven (threshold-based) retrieval algorithms for lake ice cover mapping that use top-of-atmosphere (TOA) reflectance products do not perform well under lower solar illumination conditions (i.e. large solar zenith angles), resulting in low TOA reflectance. This research assessed the capability of four machine learning classifiers (i.e. multinomial logistic regression, MLR; support vector machine, SVM; random forest, RF; gradient boosting trees, GBT) for mapping lake ice cover, water and cloud cover during both break-up and freeze-up periods using the MODIS/Terra L1B TOA (MOD02) product. The classifiers were trained and validated using samples collected from 17 large lakes across the Northern Hemisphere (Europe and North America); lakes that represent different characteristics with regards to area, latitude, freezing frequency, and ice duration. Following an accuracy assessment using random k-fold cross-validation (k = 100), all machine learning classifiers using a 7-band combination (visible, near-infrared and shortwave-infrared) were found to be able to produce overall classification accuracies above 94%. Both RF and GBT provided overall and class-specific accuracies above 98% and a more visually accurate depiction of lake ice, water and cloud cover. The two tree-based classifiers offered the most robust spatial transferability over the 17 lakes and performed consistently well across ice seasons. However, only RF was relatively insensitive to the choice of the hyperparameters compared to the other three classifiers. The results demonstrate the potential of RF for mapping lake ice cover globally from MODIS TOA reflectance data. \bullet This study assessed the capability of ML classifiers for lake ice mapping from MOIDS. \bullet RF and GBT produced the best performance in terms of classification accuracies. \bullet RF and GBT offered the most robust spatial and temporal transferability. \bullet RF was insensitive to the choice of the hyperparameters compared to other classifiers. \bullet The results show the potential of RF for mapping lake ice cover globally from MODIS.</abstract>
<identifier type="citekey">Wu-2021-Assessment</identifier>
<identifier type="doi">10.1016/j.rse.2020.112206</identifier>
<location>
<url>https://gwf-uwaterloo.github.io/gwf-publications/G21-29001</url>
</location>
<part>
<date>2021</date>
<detail type="volume"><number>253</number></detail>
<detail type="page"><number>112206</number></detail>
</part>
</mods>
</modsCollection>
%0 Journal Article
%T Assessment of machine learning classifiers for global lake ice cover mapping from MODIS TOA reflectance data
%A Wu, Yuhao
%A Duguay, Claude R.
%A Xu, Linlin
%J Remote Sensing of Environment, Volume 253
%D 2021
%V 253
%I Elsevier BV
%F Wu-2021-Assessment
%X The topic of satellite remote sensing of lake ice has gained considerable attention in recent years. Optical satellite data from the Moderate Resolution Imaging Spectroradiometer (MODIS) allow for the monitoring of lake ice cover (an Essential Climate Variable or ECV), and dates associated with ice phenology (freeze-up, break-up, and ice cover duration) over large areas in an era where ground-based observational networks have nearly vanished in many northern countries. Ice phenology dates as well as dates of maximum and minimum ice cover extent (for lakes that do not form a complete ice cover in winter or do not totally lose their ice cover in summer) are useful for assessing long-term trends and variability in climate, particularly due to their sensitivity to changes in near-surface air temperature. Existing knowledge-driven (threshold-based) retrieval algorithms for lake ice cover mapping that use top-of-atmosphere (TOA) reflectance products do not perform well under lower solar illumination conditions (i.e. large solar zenith angles), resulting in low TOA reflectance. This research assessed the capability of four machine learning classifiers (i.e. multinomial logistic regression, MLR; support vector machine, SVM; random forest, RF; gradient boosting trees, GBT) for mapping lake ice cover, water and cloud cover during both break-up and freeze-up periods using the MODIS/Terra L1B TOA (MOD02) product. The classifiers were trained and validated using samples collected from 17 large lakes across the Northern Hemisphere (Europe and North America); lakes that represent different characteristics with regards to area, latitude, freezing frequency, and ice duration. Following an accuracy assessment using random k-fold cross-validation (k = 100), all machine learning classifiers using a 7-band combination (visible, near-infrared and shortwave-infrared) were found to be able to produce overall classification accuracies above 94%. Both RF and GBT provided overall and class-specific accuracies above 98% and a more visually accurate depiction of lake ice, water and cloud cover. The two tree-based classifiers offered the most robust spatial transferability over the 17 lakes and performed consistently well across ice seasons. However, only RF was relatively insensitive to the choice of the hyperparameters compared to the other three classifiers. The results demonstrate the potential of RF for mapping lake ice cover globally from MODIS TOA reflectance data. \bullet This study assessed the capability of ML classifiers for lake ice mapping from MOIDS. \bullet RF and GBT produced the best performance in terms of classification accuracies. \bullet RF and GBT offered the most robust spatial and temporal transferability. \bullet RF was insensitive to the choice of the hyperparameters compared to other classifiers. \bullet The results show the potential of RF for mapping lake ice cover globally from MODIS.
%R 10.1016/j.rse.2020.112206
%U https://gwf-uwaterloo.github.io/gwf-publications/G21-29001
%U https://doi.org/10.1016/j.rse.2020.112206
%P 112206
Markdown (Informal)
[Assessment of machine learning classifiers for global lake ice cover mapping from MODIS TOA reflectance data](https://gwf-uwaterloo.github.io/gwf-publications/G21-29001) (Wu et al., GWF 2021)
ACL
- Yuhao Wu, Claude R. Duguay, and Linlin Xu. 2021. Assessment of machine learning classifiers for global lake ice cover mapping from MODIS TOA reflectance data. Remote Sensing of Environment, Volume 253, 253:112206.
