簡易檢索 / 詳目顯示
| 研究生： |
林佩玟 Lin, Pei-Wen |
|---|---|
| 論文名稱： |
臺灣西部紅樹林時空變遷與土壤碳存量之遙測研究 Remote Sensing of Spatiotemporal Change of Mangrove Forests and Soil Carbon Stocks in Western Taiwan |
| 指導教授： |
范噶色
Stephan van Gasselt |
| 口試委員: |
范噶色
Stephan van Gasselt 溫在弘 Wen, Tzai-Hung 黃春嘉 Huang, Chun-Jia |
| 學位類別： |
碩士
Master |
| 系所名稱： |
社會科學學院 - 地政學系 Department of Land Economics |
| 論文出版年： | 2026 |
| 畢業學年度： | 114 |
| 語文別： | 英文 |
| 論文頁數： | 149 |
| 中文關鍵詞： | 紅樹林 、藍碳 、土壤碳存量 、長時序遙測 、光譜指數 、美國大地衛星 (Landsat) |
| 外文關鍵詞： | mangroves, blue carbon, soil carbon stock, long-term remote sensing, spectral index, Landsat |
| 相關次數： | 點閱：5 下載：0 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
全球暖化與氣候變遷已加劇海平面上升、極端氣候及沿海生態系統所承受之環境壓力，使自然生態系統在氣候減緩與調適中的功能日益受到重視。紅樹林具有高碳儲存能力，能將大量碳儲存於植被與土壤中，同時具有海岸保護、生物多樣性維持及環境調節等多重生態系服務，因此被視為重要的濱海藍碳生態系與自然為本解決方案。紅樹林的保育與長期監測亦與聯合國永續發展目標密切相關，包括 SDG 13「氣候行動」所強調之氣候變遷減緩與調適、SDG 14「保育及永續利用海洋生態系」及 SDG 15「陸域生態」之濕地與生態系保育，並可作為 SDG 6.6.1 水相關生態系面積變化監測的重要對象。因此，掌握紅樹林長期分布變遷及其碳存量，不僅有助於理解沿海生態系對氣候變遷的回應，亦是評估藍碳資源及支持永續海岸管理的重要基礎。
然而，在區域尺度的紅樹林碳存量評估中，常以紅樹林分布面積結合單位面積碳密度進行推估。既有藍碳研究多著重於提升植被生物量、地下部碳庫及土壤碳密度參數之精度，相較之下，遙測判釋所造成的紅樹林面積不確定性及其如何傳遞至碳存量估算，受到的關注相對有限。實際上，在「面積 × 碳密度」的估算架構下，分布面積與碳密度皆會直接影響最終碳存量。尤其臺灣西部紅樹林多呈狹長、破碎且受潮汐、泥灘地、淺水及周邊植被干擾的空間型態，使長期衛星影像判釋容易產生混合像元與邊界誤差。因此，若未充分評估面積判釋的不確定性，即使碳密度參數具有高度精度，最終碳存量仍可能產生系統性偏差。
本研究利用 Google Earth Engine 處理 Landsat 5、7、8 與 9 地表反射率影像，重建 1990 至 2025 年臺灣西部 12 處樣區之逐年紅樹林分布，並評估判釋誤差對面積型碳存量估算之影響。各年度影像以移動時間窗進行中位數合成，採一致之 Normalized Mangrove Vegetation Index（MVIn）固定閾值，搭配地形限制與最小圖斑篩選，以維持長時序比較之一致性；序列末端年份因時間窗不完整，視為推估結果。結果顯示，臺灣西部紅樹林自 1990 年代以來整體呈擴張趨勢，但近年部分地區出現下降，且各樣區變遷幅度與時間並不一致。固定閾值可掌握整體長期趨勢，但判釋表現具有區域差異。敏感度分析亦顯示，空間精度與面積偏差未必同步，參數調整可能明顯影響判釋面積及碳存量估算。因此，長期紅樹林藍碳監測除碳密度不確定性外，亦應考量分布判釋誤差及其對碳存量估算之傳遞效應。
Global warming and climate change have intensified sea-level rise, extreme weather events, and environmental pressures on coastal ecosystems, increasing recognition of the role of natural ecosystems in climate change mitigation and adaptation. Mangrove forests have a high capacity for carbon storage, sequestering substantial amounts of carbon in both vegetation and soils while providing multiple ecosystem services, including coastal protection, biodiversity conservation, and environmental regulation. They are therefore regarded as important coastal blue carbon ecosystems and nature-based solutions. Mangrove conservation and long-term monitoring are also closely related to the United Nations Sustainable Development Goals (SDGs), particularly SDG 13 (Climate Action), which emphasizes climate change mitigation and adaptation; SDG 14 (Life Below Water); and SDG 15 (Life on Land), which addresses wetland and ecosystem conservation. Mangroves can also serve as an important target for monitoring changes in the extent of water-related ecosystems under SDG Indicator 6.6.1. Understanding long-term changes in mangrove distribution and carbon stocks is therefore essential not only for evaluating the responses of coastal ecosystems to climate change, but also for assessing blue carbon resources and supporting sustainable coastal management.
However, regional-scale assessments of mangrove carbon stocks commonly estimate total carbon storage by combining mapped mangrove area with carbon density per unit area. Previous blue carbon studies have largely focused on improving the accuracy of carbon-density parameters, including aboveground biomass, belowground carbon pools, and soil carbon density. In comparison, uncertainties in mangrove area derived from remote-sensing classification, and the manner in which these uncertainties propagate into carbon-stock estimates, have received relatively limited attention. Under an ``area $\times$ carbon density'' framework, both mapped area and carbon density directly affect the resulting carbon-stock estimate. This issue is particularly important along the western coast of Taiwan, where mangroves often occur in narrow and fragmented patches and are affected by tides, mudflats, shallow water, and surrounding vegetation. These conditions increase the likelihood of mixed pixels and boundary-related classification errors in long-term satellite observations. Consequently, even when carbon-density parameters are accurately estimated, insufficient consideration of uncertainty in mapped mangrove area may still lead to systematic bias in carbon-stock estimates.
This study used Google Earth Engine to process Landsat 5, 7, 8, and 9 surface reflectance imagery and reconstruct annual mangrove distributions from 1990 to 2025 across 12 study sites along the western coast of Taiwan. Classification errors and their effects on area-based carbon-stock estimates were further evaluated. Annual imagery was composited using a moving temporal window and pixel-wise median values. A fixed threshold of the Normalized Mangrove Vegetation Index (MVIn), consistently applied across all years and study sites, was combined with topographic constraints and minimum-patch filtering to maintain methodological consistency throughout the long-term time series. Because complete temporal windows were unavailable for the final years of the series, estimates for these years were treated as provisional. The results indicate that mangrove area along the western coast of Taiwan has generally expanded since the 1990s, although declines have occurred in several areas in recent years. The direction, magnitude, and timing of peak mangrove extent varied among study sites, indicating that regional-scale trends represent the combined effects of localized expansion and contraction. Accuracy and sensitivity analyses showed that the fixed-threshold approach was capable of capturing the overall long-term trend, but classification performance varied spatially, indicating limited transferability of a single classification rule across different regions. In addition, spatial overlap accuracy and total-area bias did not necessarily change in parallel. Some parameter adjustments produced only minor differences in spatial accuracy while substantially altering the estimated mangrove area, thereby affecting carbon-stock estimates.
This study demonstrates that, in addition to uncertainty in carbon density, classification errors in mapped mangrove area represent an important source of uncertainty in blue carbon assessments. Long-term mangrove blue carbon monitoring should therefore consider spatial classification accuracy, area bias, and the propagation of these uncertainties into carbon-stock estimates simultaneously.
誌謝 i
Declaration ii
Abstract iii
摘要 v
Contents vi
List of Figures x
List of Tables xii
1 Introduction 1
1.1 Background and Motivation 1
1.1.1 The Carbon Function of Mangroves and the Management Context in Taiwan 1
1.1.2 Extent Mapping as a Source of Uncertainty in Carbon Stock Estimation 4
1.1.3 Choice of Method for a Long Time Series 5
1.2 Research Objectives 7
1.3 Scope and Limitations 8
1.4 Research Workflow and Thesis Structure 13
2 Literature Review 15
2.1 Mangrove Ecosystems and Blue Carbon 15
2.1.1 Local Carbon Coefficients for Taiwan 17
2.2 Remote Sensing Methods for Mangrove Mapping 20
2.2.1 Spectral Index Thresholding 20
2.2.2 Supervised Machine Learning 25
2.2.3 Deep Learning 27
2.2.4 Criteria for Method Selection 29
2.3 Existing Mangrove Distribution Datasets 30
2.4 Approaches to Soil Carbon Stock Estimation 31
2.5 Accuracy Assessment and Area Estimation 32
2.6 Research Gaps 34
3 Study Area and Data 35
3.1 Study Area 35
3.1.1 Definition of the Area 35
3.1.2 Species Composition and Heterogeneity of Carbon Coefficients 37
3.1.3 Coverage of the Study Sites Relative to the National Inventory 38
3.1.4 Criteria for Site Subsets 39
3.2 Data Sources 42
3.3 Coordinate System and Computing Environment 45
4 Methodology 47
4.1 Study Sites and Analysis Data 49
4.2 GEE Pre-processing and Annual Compositing 52
4.3 MVIn Computation and Classification 56
4.3.1 The Normalized Mangrove Vegetation Index 56
4.3.2 Threshold Setting 60
4.4 Terrain Constraint and Spatial Post-processing 62
4.5 Analysis Extent and Sensitivity 63
4.6 Local Processing 65
4.7 Accuracy Assessment 66
4.7.1 Pixel-level Internal Validation 67
4.7.2 External Validation 67
4.7.3 Area Ratio and Relative Area Bias 68
4.7.4 Official Area Validation and Reference Quality 69
4.7.5 Cross-comparison with the MDC Product 71
4.8 Spatial Pattern Metrics 72
4.9 Soil Carbon Stock Estimation 73
4.9.1 Look-up Carbon Density Parameters 74
4.9.2 The Maxwell SOC Spatial Layer 75
4.9.3 Species-weighted Carbon Coefficients 76
4.10 Statistical Methods 79
4.11 Software, Outputs and Reproducibility 80
5 Results 82
5.1 Mapping Accuracy 82
5.1.1 Pixel-level Internal Validation 82
5.1.2 External Validation 84
5.1.3 Threshold Sensitivity 86
5.1.4 Error Against Official Area Baselines 88
5.1.5 Effect of the Analysis Extent 89
5.2 Distribution and Change 91
5.2.1 Regional Area Series 91
5.2.2 Change by Site 93
5.2.3 Spatial Pattern of Change 94
5.2.4 Annual Comparison with MDC 98
5.3 Soil Carbon Stocks 101
5.3.1 Carbon Stock Series and Its Change 101
5.3.2 Uncertainty from the Distribution Source 104
5.3.3 Spatial Variation in Carbon Density 105
5.3.4 Carbon Density by Change Type 106
6 Discussion 108
6.1 Temporal Consistency of the Compositing Window 108
6.2 Decoupling of Spatial Agreement and Area Bias 111
6.3 Spatial Structure of the Error 114
6.4 Comparison with an Existing Dataset 116
6.5 Bias Assessment and Sensitivity of the Carbon Estimate 117
6.6 Limitations 123
7 Conclusions and Recommendations 126
7.1 Main Findings 126
7.2 Implications for Policy and Method 131
7.3 Recommendations for Further Work 133
References 135
A Program Register and Data Correspondence 143
B Terrain Constraint and Post-processing by Site 147
Alongi, D. M. (2014). Carbon cycling and storage in mangrove forests. Annual Review of Marine Science, 6:195–219.
Baloloy, A. B., Blanco, A. C., Sta. Ana, R. R. C., and Nadaoka, K. (2020). Development and application of a new mangrove vegetation index (MVI) for rapid and accurate mangrove mapping. ISPRS Journal of Photogrammetry and Remote Sensing, 166:95–117.
Belgiu, M. and Drăguţ, L. (2016). Random forest in remote sensing: A review of applications and future directions. ISPRS Journal of Photogrammetry and Remote Sensing, 114:24–31.
Breiman, L. (2001). Random forests. Machine Learning, 45(1):5–32.
Bunting, P., Rosenqvist, A., Hilarides, L., Lucas, R. M., Thomas, N., Tadono, T., Worthington, T. A., Spalding, M., Murray, N. J., and Rebelo, L.-M. (2022). Global mangrove extent change 1996–2020: Global Mangrove Watch version 3.0. Remote Sensing, 14(15):3657.
Bunting, P., Rosenqvist, A., Lucas, R. M., Rebelo, L.-M., Hilarides, L., Thomas, N., Hardy, A., Itoh, T., Shimada, M., and Finlayson, C. M. (2018). The Global Mangrove Watch—a new 2010 global baseline of mangrove extent. Remote Sensing, 10(10):1669.
Card, D. H. (1982). Using known map category marginal frequencies to improve estimates of thematic map accuracy. Photogrammetric Engineering and Remote Sensing, 48(3):431–439.
Donato, D. C., Kauffman, J. B., Murdiyarso, D., Kurnianto, S., Stidham, M., and Kanninen, M. (2011). Mangroves among the most carbon-rich forests in the tropics. Nature Geoscience, 4(5):293–297.
Foody, G. M. (2002). Status of land cover classification accuracy assessment. Remote Sensing of Environment, 80(1):185–201.
Giri, C., Ochieng, E., Tieszen, L. L., Zhu, Z., Singh, A., Loveland, T., Masek, J., and Duke, N. (2011). Status and distribution of mangrove forests of the world using earth observation satellite data. Global Ecology and Biogeography, 20(1):154–159.
Gómez, C., White, J. C., and Wulder, M. A. (2016). Optical remotely sensed time series data for land cover classification: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 116:55–72.
Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., and Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of Environment, 202:18–27.
Guo, Y., Liao, J., and Shen, G. (2021). Mapping large-scale mangroves along the Maritime Silk Road from 1990 to 2015 using a novel deep learning model and Landsat data. Remote Sensing, 13(2):245.
Gupta, K., Mukhopadhyay, A., Giri, S., Chanda, A., Datta Majumdar, S., Samanta, S., Mitra, D., Samal, R. N., Pattnaik, A. K., and Hazra, S. (2018). An index for discrimination of mangroves from non-mangroves using LANDSAT 8 OLI imagery. MethodsX, 5:1129–1139.
Hengl, T., Mendes de Jesus, J., Heuvelink, G. B. M., Ruiperez Gonzalez, M., Kilibarda, M., Blagotić, A., Shangguan, W., Wright, M. N., Geng, X., Bauer-Marschallinger, B., Guevara, M. A., Vargas, R., MacMillan, R. A., Batjes, N. H., Leenaars, J. G. B., Ribeiro, E., Wheeler, I., Mantel, S., and Kempen, B. (2017). SoilGrids250m: Global gridded soil information based on machine learning. PLOS ONE, 12(2):e0169748.
Hsinchu City Government (2024). Mangrove management at Xiangshan wetland. Xiangshan Wetland Ecology Portal. In Chinese. 新竹市政府，香山濕地生態網。Records clearance from 2007, full-area clearance of 346 ha from 2015, and a cumulative cleared area of about 450 ha by 2019. https://ssw.hccg.gov.tw/faqs, accessed 2026-08-24.
Huete, A. R. (1988). A soil-adjusted vegetation index (SAVI). Remote Sensing of Environment, 25(3):295–309.
Intergovernmental Panel on Climate Change (2014). 2013 supplement to the 2006 IPCC guidelines for national greenhouse gas inventories: Wetlands. IPCC, Geneva, Switzerland. Edited by T. Hiraishi, T. Krug, K. Tanabe, N. Srivastava, J. Baasansuren, M. Fukuda, & T. G. Troxler.
Lim, H. S., Lee, Y., Lin, M. H., and Chia, W. C. (2024). Mangrove species detection using YOLOv5 with RGB imagery from consumer unmanned aerial vehicles (UAVs). The Egyptian Journal of Remote Sensing and Space Sciences, 27(4):645–655.
Lin, H.-J. (2024). Survey and assessment of potential restoration sites for marine carbon sinks in Taiwan: Results report. Commissioned research results report 111-C-70, Ocean Conservation Administration, Ocean Affairs Council, Kaohsiung, Taiwan. 240 pp. In Chinese. 林幸助，臺灣海洋碳匯潛力復育點調查與評估計畫成果報告。Executing institution: National Chung Hsing University.
Lin, H.-J., Chen, K.-Y., Kao, Y.-C., Lin, W.-J., Lin, C.-W., and Ho, C.-W. (2023). Assessing coastal blue carbon sinks in Taiwan. Marine Research, 3(2):1–17.
Lin, H.-J., Ho, C.-W., and Chen, T.-Y. (2019). Mangrove ecosystem survey project, 2019: Results report. Commissioned research results report, Ocean Conservation Administration, Ocean Affairs Council, Kaohsiung, Taiwan. 166 pp. In Chinese. 林幸助、何瓊紋、陳彤昀，108 年紅樹林生態系調查計畫成果報告書，海洋委員會海洋保育署。Executing institution: National Chung Hsing University.
Maxwell, T. L., Hengl, T., Parente, L. L., Minarik, R., Worthington, T. A., Bunting, P., Smart, L. S., Spalding, M. D., and Landis, E. (2023). Global mangrove soil organic carbon stocks dataset at 30 m resolution for the year 2020 based on spatiotemporal predictive machine learning. Data in Brief, 50:109621. Data descriptor for the soil organic carbon layer used here. https://doi.org/10.1016/j.dib.2023.109621.
McFeeters, S. K. (1996). The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. International Journal of Remote Sensing, 17(7):1425–1432.
McKee, K. L., Cahoon, D. R., and Feller, I. C. (2007). Caribbean mangroves adjust to rising sea level through biotic controls on change in soil elevation. Global Ecology and Biogeography, 16(5):545–556.
McLeod, E., Chmura, G. L., Bouillon, S., Salm, R., Björk, M., Duarte, C. M., Lovelock, C. E., Schlesinger, W. H., and Silliman, B. R. (2011). A blueprint for blue carbon: Toward an improved understanding of the role of vegetated coastal habitats in sequestering CO2. Frontiers in Ecology and the Environment, 9(10):552–560.
Millennium Ecosystem Assessment (2005). Ecosystems and human well-being: Synthesis. Island Press, Washington, DC.
NASA JPL (2020). NASADEM merged DEM global 1 arc second V001. [Data set]. NASA EOSDIS Land Processes DAAC. Accessed through Google Earth Engine.
Olofsson, P., Foody, G. M., Herold, M., Stehman, S. V., Woodcock, C. E., and Wulder, M. A. (2014). Good practices for estimating area and assessing accuracy of land change. Remote Sensing of Environment, 148:42–57.
Olofsson, P., Foody, G. M., Stehman, S. V., and Woodcock, C. E. (2013). Making better use of accuracy data in land change studies: Estimating accuracy and area and quantifying uncertainty using stratified estimation. Remote Sensing of Environment, 129:122–131.
Pal, M. and Mather, P. M. (2005). Support vector machines for classification in remote sensing. International Journal of Remote Sensing, 26(5):1007–1011.
Pendleton, L., Donato, D. C., Murray, B. C., Crooks, S., Jenkins, W. A., Sifleet, S., Craft, C., Fourqurean, J. W., Kauffman, J. B., Marbà, N., Megonigal, P., Pidgeon, E., Herr, D., Gordon, D., and Baldera, A. (2012). Estimating global “blue carbon” emissions from conversion and degradation of vegetated coastal ecosystems. PLoS ONE, 7(9):e43542.
Pham, T. D., Yoshino, K., Le, N. N., and Bui, D. T. (2018). Estimating aboveground biomass of a mangrove plantation on the northern coast of Vietnam using machine learning techniques with an integration of ALOS-2 PALSAR-2 and Sentinel-2A data. International Journal of Remote Sensing, 39(22):7761–7788.
Prayudha, B., Ulumuddin, Y. I., Siregar, V., Suyarso, Agus, S. B., Prasetyo, L. B., Suyadi, Avianto, P., and Ramadhani, M. R. (2024). Enhanced mangrove index: A spectral index for discrimination understorey, nypa, and mangrove trees. MethodsX, 12:102778.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention — MICCAI 2015, pages 234–241, Cham, Switzerland. Springer.
Rouse, J. W., Haas, R. H., Schell, J. A., and Deering, D. W. (1974). Monitoring vegetation systems in the Great Plains with ERTS. In Third Earth Resources Technology Satellite-1 Symposium, volume 1 of NASA SP-351, pages 309–317, Washington, DC. NASA.
Roy, D. P., Kovalskyy, V., Zhang, H. K., Vermote, E. F., Yan, L., Kumar, S. S., and Egorov, A. (2016). Characterization of Landsat-7 to Landsat-8 reflective wavelength and normalized difference vegetation index continuity. Remote Sensing of Environment, 185:57–70.
Sharp, R., Tallis, H. T., Ricketts, T., Guerry, A. D., Wood, S. A., Chaplin-Kramer, R., Nelson, E., Ennaanay, D., Wolny, S., Olwero, N., et al. (2020). InVEST user’s guide. Technical report, The Natural Capital Project, Stanford University, University of Minnesota, The Nature Conservancy, and World Wildlife Fund. Version 3.8; Coastal Blue Carbon model documentation.
Stehman, S. V. and Foody, G. M. (2019). Key issues in rigorous accuracy assessment of land cover products. Remote Sensing of Environment, 231:111199.
Suyarso and Avianto, P. (2022). AMMI automatic mangrove map and index: Novelty for efficiently monitoring mangrove changes with the case study in Musi Delta, South Sumatra, Indonesia. International Journal of Forestry Research, 2022:1–14.
Tomlinson, P. B. (2016). The botany of mangroves. Cambridge University Press, Cambridge, England, 2nd edition.
Tran, T. V., Reef, R., Zhu, X., and Gunn, A. (2024). Characterising the distribution of mangroves along the southern coast of Vietnam using multi-spectral indices and a deep learning model. Science of the Total Environment, 923:171367.
Tuia, D., Persello, C., and Bruzzone, L. (2016). Domain adaptation for the classification of remote sensing data: An overview of recent advances. IEEE Geoscience and Remote Sensing Magazine, 4(2):41–57.
U.S. Geological Survey (2024). Landsat 8–9 OLI/TIRS Collection 2 Level 2 science product guide. Technical Report LSDS-1619, U.S. Geological Survey, Earth Resources Observation and Science (EROS) Center, Sioux Falls, SD. Level-2 surface reflectance scaling factors and QA_PIXEL bit definitions. The corresponding document for Landsat 4–7 is LSDS-1618.
Wang, H.-H., Fu, S.-W., Teng, K.-C., Hung, H.-C., and Liao, H.-Y. (2015). Changes in mangrove area and species composition in Taiwan. Taiwan Forestry Journal, 41(2):47–51. In Chinese. 王相華、傅淑瑋、鄧國禎、洪西洲、廖學儀，臺灣紅樹林面積變遷及物種組成現況，台灣林業 41(2)：47–51.
Wang, X., Li, H., Jia, M., Zhang, R., Zhao, C., Lu, C., and Wang, Z. (2026). Significant increase in China’s mangrove cover fraction during 1990 to 2023. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 19:3625–3636.
Win, K. and Sasaki, J. (2024). The change detection of mangrove forests using deep learning with medium-resolution satellite imagery: A case study of Wunbaik mangrove forest in Myanmar. Remote Sensing, 16(21):4077.
Winarso, G., Purwanto, A. D., and Yuwono, D. M. (2014). New mangrove index as degradation health indicator using remote sensing data: Segara Anakan and Alas Purwo case study. In Proceedings of the 12th Biennial Conference of Pan Ocean Remote Sensing Conference (PORSEC 2014), Bali, Indonesia.
World Meteorological Organization (2025). WMO confirms 2024 warmest year on record, about 1.55 °C above pre-industrial level. Press release. Published 10 January 2025. https://wmo.int/news/media-centre/wmo-confirms-2024-warmest-year-record-about-155degc-above-pre-industrial-level, accessed 5 January 2026.
Yang, G., Huang, K., Sun, W., Wang, L., and Chen, B. (2024). Mangrove dynamics in China (MDC): Accurate and annual distribution maps from 1990 to 2020 (version 1). [Data set]. Science Data Bank. Version 1, released 30 July 2024. https://doi.org/10.57760/sciencedb.11480, accessed 5 January 2026.
Yang, J. and Huang, X. (2021). The 30 m annual land cover dataset and its dynamics in China from 1990 to 2019. Earth System Science Data, 13(8):3907–3925.
