Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data

Shaojia Ge; Erkki Tomppo; Yrjo Rauste; Weimin Su; Hong Gu; Jaan Praks; Oleg Antropov

doi:10.1109/IGARSS39084.2020.9324212

Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data

Shaojia Ge, Erkki Tomppo, Yrjo Rauste, Weimin Su, Hong Gu, Jaan Praks, Oleg Antropov

BA6408 Remote sensing

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

3 Citations (Scopus)

Abstract

In this study, we assess the potential of long time series of Sentinel-1 SAR data in forest growing stock volume (GSV) estimation. The study site with 17,762 forest stands is located near the Hyytiälä forestry field station in Finland and represents the boreal coniferous forest. Altogether 96 images spanning more than three years of observations have been studied using linear and random forest regression approaches. Our analysis demonstrates considerable decrease in the prediction errors of GSV as the the number of input scenes increases. The use of feature extraction and dimensionality reduction techniques allows to achieve to nearly optimal performance already with 10 scenes. While the GSV prediction errors using individual Sentinel-1 scenes varied considerably from 86 to 93 m³/ha, the prediction accuracy with combined scenes improved to 76 m³/ha (44.9%) RMSE.

Original language	English
Title of host publication	2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020
Subtitle of host publication	Proceedings
Publisher	IEEE Institute of Electrical and Electronic Engineers
Pages	4509-4512
ISBN (Electronic)	978-1-7281-6374-1
ISBN (Print)	978-1-7281-6375-8
DOIs	https://doi.org/10.1109/IGARSS39084.2020.9324212
Publication status	Published - 17 Feb 2020
MoE publication type	A4 Article in a conference publication
Event	IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020: Online - Virtual, Waikoloa, United States Duration: 26 Sept 2020 → 2 Oct 2020

Publication series

Series	IEEE International Geoscience and Remote Sensing Symposium Proceedings
ISSN	2153-6996

Conference

Conference	IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020
Country/Territory	United States
City	Waikoloa
Period	26/09/20 → 2/10/20

Keywords

boreal forest
forest stem volume
random forest regression
Sentinel-1
support vector regression
synthetic aperture radar

Access to Document

10.1109/IGARSS39084.2020.9324212

Cite this

Ge, S., Tomppo, E., Rauste, Y., Su, W., Gu, H., Praks, J., & Antropov, O. (2020). Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data. In 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020: Proceedings (pp. 4509-4512). Article 9324212 IEEE Institute of Electrical and Electronic Engineers. https://doi.org/10.1109/IGARSS39084.2020.9324212

Ge, Shaojia ; Tomppo, Erkki ; Rauste, Yrjo et al. / Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data. 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020: Proceedings. IEEE Institute of Electrical and Electronic Engineers, 2020. pp. 4509-4512 (IEEE International Geoscience and Remote Sensing Symposium Proceedings).

@inproceedings{4842b425f57c482e9dcd9a893c89bd37,

title = "Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data",

abstract = "In this study, we assess the potential of long time series of Sentinel-1 SAR data in forest growing stock volume (GSV) estimation. The study site with 17,762 forest stands is located near the Hyyti{\"a}l{\"a} forestry field station in Finland and represents the boreal coniferous forest. Altogether 96 images spanning more than three years of observations have been studied using linear and random forest regression approaches. Our analysis demonstrates considerable decrease in the prediction errors of GSV as the the number of input scenes increases. The use of feature extraction and dimensionality reduction techniques allows to achieve to nearly optimal performance already with 10 scenes. While the GSV prediction errors using individual Sentinel-1 scenes varied considerably from 86 to 93 m3/ha, the prediction accuracy with combined scenes improved to 76 m3/ha (44.9%) RMSE.",

keywords = "boreal forest, forest stem volume, random forest regression, Sentinel-1, support vector regression, synthetic aperture radar",

author = "Shaojia Ge and Erkki Tomppo and Yrjo Rauste and Weimin Su and Hong Gu and Jaan Praks and Oleg Antropov",

note = "Funding Information: This study was supported by the Nationa l Natural Science Foundation of China (Grant No. 61671246,61801221), and by the Natural Science Foundation of Jiangs u Province (Grant No. BK20170855). Publisher Copyright: {\textcopyright} 2020 IEEE. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.; IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020 : Online ; Conference date: 26-09-2020 Through 02-10-2020",

year = "2020",

month = feb,

day = "17",

doi = "10.1109/IGARSS39084.2020.9324212",

language = "English",

isbn = "978-1-7281-6375-8",

series = "IEEE International Geoscience and Remote Sensing Symposium Proceedings",

publisher = "IEEE Institute of Electrical and Electronic Engineers",

pages = "4509--4512",

booktitle = "2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020",

address = "United States",

}

Ge, S, Tomppo, E, Rauste, Y, Su, W, Gu, H, Praks, J & Antropov, O 2020, Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data. in 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020: Proceedings., 9324212, IEEE Institute of Electrical and Electronic Engineers, IEEE International Geoscience and Remote Sensing Symposium Proceedings, pp. 4509-4512, IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020, Waikoloa, Hawaii, United States, 26/09/20. https://doi.org/10.1109/IGARSS39084.2020.9324212

Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data. / Ge, Shaojia; Tomppo, Erkki; Rauste, Yrjo et al.
2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020: Proceedings. IEEE Institute of Electrical and Electronic Engineers, 2020. p. 4509-4512 9324212 (IEEE International Geoscience and Remote Sensing Symposium Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data

AU - Ge, Shaojia

AU - Tomppo, Erkki

AU - Rauste, Yrjo

AU - Su, Weimin

AU - Gu, Hong

AU - Praks, Jaan

AU - Antropov, Oleg

N1 - Funding Information: This study was supported by the Nationa l Natural Science Foundation of China (Grant No. 61671246,61801221), and by the Natural Science Foundation of Jiangs u Province (Grant No. BK20170855). Publisher Copyright: © 2020 IEEE. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2020/2/17

Y1 - 2020/2/17

N2 - In this study, we assess the potential of long time series of Sentinel-1 SAR data in forest growing stock volume (GSV) estimation. The study site with 17,762 forest stands is located near the Hyytiälä forestry field station in Finland and represents the boreal coniferous forest. Altogether 96 images spanning more than three years of observations have been studied using linear and random forest regression approaches. Our analysis demonstrates considerable decrease in the prediction errors of GSV as the the number of input scenes increases. The use of feature extraction and dimensionality reduction techniques allows to achieve to nearly optimal performance already with 10 scenes. While the GSV prediction errors using individual Sentinel-1 scenes varied considerably from 86 to 93 m3/ha, the prediction accuracy with combined scenes improved to 76 m3/ha (44.9%) RMSE.

AB - In this study, we assess the potential of long time series of Sentinel-1 SAR data in forest growing stock volume (GSV) estimation. The study site with 17,762 forest stands is located near the Hyytiälä forestry field station in Finland and represents the boreal coniferous forest. Altogether 96 images spanning more than three years of observations have been studied using linear and random forest regression approaches. Our analysis demonstrates considerable decrease in the prediction errors of GSV as the the number of input scenes increases. The use of feature extraction and dimensionality reduction techniques allows to achieve to nearly optimal performance already with 10 scenes. While the GSV prediction errors using individual Sentinel-1 scenes varied considerably from 86 to 93 m3/ha, the prediction accuracy with combined scenes improved to 76 m3/ha (44.9%) RMSE.

KW - boreal forest

KW - forest stem volume

KW - random forest regression

KW - Sentinel-1

KW - support vector regression

KW - synthetic aperture radar

UR - http://www.scopus.com/inward/record.url?scp=85101971910&partnerID=8YFLogxK

U2 - 10.1109/IGARSS39084.2020.9324212

DO - 10.1109/IGARSS39084.2020.9324212

M3 - Conference article in proceedings

AN - SCOPUS:85101971910

SN - 978-1-7281-6375-8

T3 - IEEE International Geoscience and Remote Sensing Symposium Proceedings

SP - 4509

EP - 4512

BT - 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020

PB - IEEE Institute of Electrical and Electronic Engineers

T2 - IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020

Y2 - 26 September 2020 through 2 October 2020

ER -

Ge S, Tomppo E, Rauste Y, Su W, Gu H, Praks J et al. Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data. In 2020 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2020: Proceedings. IEEE Institute of Electrical and Electronic Engineers. 2020. p. 4509-4512. 9324212. (IEEE International Geoscience and Remote Sensing Symposium Proceedings). doi: 10.1109/IGARSS39084.2020.9324212

Predicting Growing Stock Volume of Boreal Forests Using Very Long Time Series of Sentinel-1 Data

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this