Production of global daily seamless data cubes and quantification of global land cover change from 1985 to 2020 - iMap World 1.0

Liu, Han; Gong, Peng; Wang, Jie; Wang, Xi; Ning, Grant; Xu, Bing

doi:10.1016/j.rse.2021.112364

Cited by 111 publications

(69 citation statements)

References 117 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The full Landsat-5 L1T-level surface reflectance archive [63] covering China with a cloud score of less than 60 was preprocessed and downloaded effortlessly from GEE, a cloud-based platform for processing petabyte-scale geospatial datasets [61]. The L1T-level products have undergone geometric, radiation, and atmospheric corrections and are ready for use [64,65]. After masking clouds and shadows using Landsat quality flag information [66], a composite for a given year was produced in the form of a median mosaic of all available Landsat scenes.…”

Section: Landsat-5 Rs Imagerymentioning

confidence: 99%

Mapping Multi-Temporal Population Distribution in China from 1985 to 2010 Using Landsat Images via Deep Learning

Zhuang

Yan

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Fine knowledge of the spatiotemporal distribution of the population is fundamental in a wide range of fields, including resource management, disaster response, public health, and urban planning. The United Nations’ Sustainable Development Goals also require the accurate and timely assessment of where people live to formulate, implement, and monitor sustainable development policies. However, due to the lack of appropriate auxiliary datasets and effective methodological frameworks, there are rarely continuous multi-temporal gridded population data over a long historical period to aid in our understanding of the spatiotemporal evolution of the population. In this study, we developed a framework integrating a ResNet-N deep learning architecture, considering neighborhood effects with a vast number of Landsat-5 images from Google Earth Engine for population mapping, to overcome both the data and methodology obstacles associated with rapid multi-temporal population mapping over a long historical period at a large scale. Using this proposed framework in China, we mapped fine-scale multi-temporal gridded population data (1 km × 1 km) of China for the 1985–2010 period with a 5-year interval. The produced multi-temporal population data were validated with available census data and achieved comparable performance. By analyzing the multi-temporal population grids, we revealed the spatiotemporal evolution of population distribution from 1985 to 2010 in China with the characteristic of concentration of the population in big cities and the contraction of small- and medium-sized cities. The framework proposed in this study demonstrates the feasibility of mapping multi-temporal gridded population distribution at a large scale over a long period in a timely and low-cost manner, which is particularly useful in low-income and data-poor areas.

show abstract

Section: Landsat-5 Rs Imagerymentioning

confidence: 99%

Mapping Multi-Temporal Population Distribution in China from 1985 to 2010 Using Landsat Images via Deep Learning

Zhuang

Yan

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Land cover change can influence the energy balance and biogeochemical cycles ( Claussen, Brovkin & Ganopolski, 2001 ; DeFries et al, 1999 ) and it can further affect climate change, surface characteristics and the provision of ecosystem services ( Pielke, 2005 ; Reyers et al, 2009 ; Zhao, Pitman & Chase, 2001 ). Therefore, better frequent land cover observations are desirable for understanding global environmental change ( Liu et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

“…For instance, global land cover maps with 30 m-resolution based on Landsat images (Finer Resolution Observation and Monitoring of Global Land Cover, FROM-GLC) ( Gong et al, 2013 ), and 10-m resolution, based on Sentinel 2 (FROM-GLC10), were developed recently ( Gong et al, 2019 ). However, those finer resolution land cover products are hard to be updated to cover long time series due to low data availability for Landsat in the past ( Liu et al, 2021 ; Yu, Shi & Gong, 2015 ). By aggregating and fusing Landsat and MODIS images, which have different spatial resolutions and observation frequencies, researchers have made progress on land cover mapping by improving their accuracy ( Yu, Wang & Gong, 2013 ) and increasing their observation frequency very recently ( Liu et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

“…However, those finer resolution land cover products are hard to be updated to cover long time series due to low data availability for Landsat in the past ( Liu et al, 2021 ; Yu, Shi & Gong, 2015 ). By aggregating and fusing Landsat and MODIS images, which have different spatial resolutions and observation frequencies, researchers have made progress on land cover mapping by improving their accuracy ( Yu, Wang & Gong, 2013 ) and increasing their observation frequency very recently ( Liu et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Towards an open and synergistic framework for mapping global land cover

Zhao

Liu

et al. 2021

PeerJ

Self Cite

View full text Add to dashboard Cite

Global land-cover datasets are key sources of information for understanding the complex inter-actions between human activities and global change. They are also among the most critical variables for climate change studies. Over time, the spatial resolution of land cover maps has increased from the kilometer scale to 10-m scale. Single-type historical land cover datasets, including for forests, water, and impervious surfaces, have also been developed in recent years. In this study, we present an open and synergy framework to produce a global land cover dataset that combines supervised land cover classification and aggregation of existing multiple thematic land cover maps with the Google Earth Engine (GEE) cloud computing platform. On the basis of this method of classification and mosaicking, we derived a global land cover dataset for 6 years over a time span of 25 years. The overall accuracies of the six maps were around 75% and the accuracy for change area detection was over 70%. Our product also showed good similarity with the FAO and existing land cover maps.

show abstract

“…the 1-km International Geosphere-Biosphere Programme data and information system cover (IGBP-DISCover) map (Loveland et al, 2000), the 1-km University of Maryland (UMD) land cover map (Hansen, DeFries, Townshend, & Sohlberg, 2000), the 1-km global land cover classification product (GLC2000) (Bartholome & Belward, 2005), the 1-km and 500-m Moderate Resolution Imaging Spectrometer (MODIS) land cover maps (Friedl et al, 2010;Tateishi et al, 2011), the 300-m global land cover map (GlobCover) derived from Medium Resolution Imaging Spectrometer (MERIS) dataset (Arino et al, 2012), the 300-m European Space Agency (ESA) Climate Change Initiative (CCI) land cover maps from 1992 to 2015 (UCL-Geomatics, 2017), the 100-m ESA Copernicus Global Land Service Land Cover Map (CGLS-LC100) (Buchhorn et al, 2020), and the 100-m global land cover fraction map (Masiliūnas et al, 2021); (ii) fine-resolution ones from 10 m to 30 m, e.g. the 30-m finer resolution observation and monitoring of global land cover (FROM-GLC30) (Gong et al, 2013), the 30-m global land cover data product (GlobeLand30) , the 30-m global land-cover product with fine classification system (GLC_FCS30) (Zhang et al, 2020a), the most recent 30-m intelligent mapping of global land cover (iMap World 1.0) (Liu et al, 2021), the 20-m ESA CCI Sentinel-2 prototype land cover map of Africa in 2016 (Lesiv et al, 2017), and the 10-m finer resolution observation and monitoring of global land cover (FROM-GLC10) (Gong et al, 2019); and (iii) high-resolution one less than 10 m, e.g. the recent 3-m national land cover map in China based on Planet Imagery (Dong et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Mapping essential urban land use categories (EULUC) using geospatial big data: Progress, challenges, and opportunities

Чэн

Gong

2021

Big Earth Data

Self Cite

View full text Add to dashboard Cite

Urban land use information that reflects socio-economic functions and human activities is critically essential for urban planning, landscape design, environmental management, health promotion, and biodiversity conservation. Land-use maps outlining the distribution, pattern, and composition of essential urban land use categories (EULUC) have facilitated a wide spectrum of applications and further triggered new opportunities in urban studies. New and improved Earth observations, algorithms, and advanced products for extracting thematic urban information, in association with emerging social sensing big data and auxiliary crowdsourcing datasets, all together offer great potentials to mapping fine-resolution EULUC from regional to global scales. Here we review the advances of EULUC mapping research and practices in terms of their data, methods, and applications. Based on the historical retrospect, we summarize the challenges and limitations of current EULUC studies regarding sample collection, mixed land use problem, data and model generalization, and large-scale mapping efforts. Finally, we propose and discuss future opportunities, including cross-scale mapping, optimal integration of multi-source features, global sample libraries from crowdsourcing approaches, advanced machine learning and ensembled classification strategy, open portals for data visualization and sharing, multi-temporal mapping of EULUC change, and implications in urban environmental studies, to facilitate multi-scale fine-resolution EULUC mapping research.

show abstract

Production of global daily seamless data cubes and quantification of global land cover change from 1985 to 2020 - iMap World 1.0

Cited by 111 publications

References 117 publications

Mapping Multi-Temporal Population Distribution in China from 1985 to 2010 Using Landsat Images via Deep Learning

Mapping Multi-Temporal Population Distribution in China from 1985 to 2010 Using Landsat Images via Deep Learning

Towards an open and synergistic framework for mapping global land cover

Mapping essential urban land use categories (EULUC) using geospatial big data: Progress, challenges, and opportunities

Contact Info

Product

Resources

About