TY - JOUR

T1 - Improving the accuracy of land cover classification in cloud persistent areas using optical and radar satellite image time series

AU - Lopes, Mailys

AU - Frison, Pierre-Louis

AU - Crowson, Merry

AU - Warren-Thomas, Eleanor

AU - Hariyadi, Bambang

AU - Kartika, Winda D.

AU - Agus, Fahmuddin

AU - Hamer, Keith C.

AU - Stringer, Lindsay

AU - Hill, Jane K.

AU - Pettorelli, Nathalie

PY - 2020/4/1

Y1 - 2020/4/1

N2 - 1. The recent availability of high spatial and temporal resolution optical and radar satellite imagery has dramatically increased opportunities for mapping land coverat fine scales. Fusion of optical and radar images has been found useful in tropical areas affected by cloud cover because of their complementarity. However,the multitemporal dimension these data now offer is often neglected becausethese areas are primarily characterized by relatively low levels of seasonality andbecause the consideration of multitemporal data requires more processing time.Hence, land cover mapping in these regions is often based on imagery acquired fora single date or on an average of multiple dates.2. The aim of this work is to assess the added value brought by the temporal dimension of optical and radar time series when mapping land cover in tropical environments. Specifically, we compared the accuracies of classifications based on (a)optical time series, (b) their temporal average, (c) radar time series, (d) their temporal average, (e) a combination of optical and radar time series and (f) a combinationof their temporal averages for mapping land cover in Jambi province, Indonesia,using Sentinel-1 and Sentinel-2 imagery.3. Using the full information contained in the time series resulted in significantlyhigher classification accuracies than using temporal averages (+14.7% forSentinel-1, +2.5% for Sentinel-2 and +2% combining Sentinel-1 and Sentinel-2).Overall, combining Sentinel-2 and Sentinel-1 time series provided the highest accuracies (Kappa = 88.5%).4. Our study demonstrates that preserving the temporal information provided bysatellite image time series can significantly improve land cover classifications intropical biodiversity hotspots, improving our capacity to monitor ecosystems of high conservation relevance such as peatlands. The proposed method is reproducible, automated and based on open-source tools satellite imagery.

AB - 1. The recent availability of high spatial and temporal resolution optical and radar satellite imagery has dramatically increased opportunities for mapping land coverat fine scales. Fusion of optical and radar images has been found useful in tropical areas affected by cloud cover because of their complementarity. However,the multitemporal dimension these data now offer is often neglected becausethese areas are primarily characterized by relatively low levels of seasonality andbecause the consideration of multitemporal data requires more processing time.Hence, land cover mapping in these regions is often based on imagery acquired fora single date or on an average of multiple dates.2. The aim of this work is to assess the added value brought by the temporal dimension of optical and radar time series when mapping land cover in tropical environments. Specifically, we compared the accuracies of classifications based on (a)optical time series, (b) their temporal average, (c) radar time series, (d) their temporal average, (e) a combination of optical and radar time series and (f) a combinationof their temporal averages for mapping land cover in Jambi province, Indonesia,using Sentinel-1 and Sentinel-2 imagery.3. Using the full information contained in the time series resulted in significantlyhigher classification accuracies than using temporal averages (+14.7% forSentinel-1, +2.5% for Sentinel-2 and +2% combining Sentinel-1 and Sentinel-2).Overall, combining Sentinel-2 and Sentinel-1 time series provided the highest accuracies (Kappa = 88.5%).4. Our study demonstrates that preserving the temporal information provided bysatellite image time series can significantly improve land cover classifications intropical biodiversity hotspots, improving our capacity to monitor ecosystems of high conservation relevance such as peatlands. The proposed method is reproducible, automated and based on open-source tools satellite imagery.

KW - cloud persistent areas

KW - conservation

KW - data combination

KW - land cover classification

KW - remote sensing

KW - satellite image time series

KW - Sentinel-1

KW - Sentinel-2

U2 - 10.1111/2041-210X.13359

DO - 10.1111/2041-210X.13359

M3 - Editorial

VL - 11

SP - 532

EP - 541

JO - Methods in Ecology and Evolution

JF - Methods in Ecology and Evolution

SN - 2041-210X

IS - 4

ER -