高时空分辨率绿洲土地覆被地图，对厘清干旱区生态演化与社会发展过程具有重要价值。然而绿洲相关研究起步较晚，加之绿洲景观高度破碎、土地覆被类型转换频繁，构建高时空分辨率绿洲土地覆被数据集面临诸多挑战，因此，专门针对绿洲的此类数据产品仍然匮乏。\n
基于此，本文构建一套逐年 30 米土地覆被制图框架，融合 Landsat 卫星影像、机器学习算法、时间序列分割模型（LandTrendr）与主成分分析方法。基于该框架，依托谷歌地球引擎（GEE）平台生成 1987—2024 年中国绿洲 30 米逐年土地覆被数据集（OasisMap30）。基于 6300 余个目视解译样本的精度验证表明， OasisMap30 整体精度高于 90%。结合目视解译样本与第三方验证数据开展多产品交叉对比，OasisMap30 在分类精度与误差控制方面优势显著。此外，将该数据集与多款 30 米不透水面、耕地、地表水体专题产品对比，发现 OasisMap30 与现有数据一致性良好。\n
基于 OasisMap30 数据集，本文解析了中国绿洲土地覆被格局演变特征。结果显示：1987—2024 年我国绿洲总面积扩张 45.87%，增量达 775 万公顷，扩张核心驱动力为耕地开垦与草地修复。其中，404 万公顷荒漠转化为草地，319 万公顷荒漠开垦为耕地；不透水面扩张 58 万公顷，地表水体增加 35 万公顷；同时存在大量地类内部转换，例如 312 万公顷草地转为耕地。\n
综上，这套时序连续、高分辨率的 OasisMap30 数据集可为绿洲景观格局演变、社会 - 生态响应、空间格局优化等研究提供有力支撑，助力绿洲区域可持续发展。\n
数据集共划分 7 类土地覆被，数据中代码及其所代表的土地覆被类型如下：1：不透水面、2：地表水体、3：耕地、4：灌木、5：林地、6：草地、7：裸地。", "ds_source": "", "ds_process_way": "", "ds_quality": "", "ds_acq_start_time": "1987-01-01 00:00:00", "ds_acq_end_time": "2024-12-31 00:00:00", "ds_acq_place": "中国", "ds_acq_lon_east": null, "ds_acq_lat_south": null, "ds_acq_lon_west": null, "ds_acq_lat_north": null, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "open-access", "ds_total_size": 6752658049, "ds_files_count": 2, "ds_format": "GeoTIFF", "ds_space_res": "30m", "ds_time_res": "年", "ds_coordinate": "无", "ds_projection": "Albers等面积投影", "ds_thumbnail": "3808cedc-34aa-4244-8651-8faa911c6e71.png", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": null, "organization_id": "a4dd5849-78f2-44c5-b0f1-3450e952b2a2", "ds_serv_man": null, "ds_serv_phone": null, "ds_serv_mail": null, "doi_value": "", "subject_codes": [ "170.4510" ], "quality_level": 0, "publish_time": "2026-07-01 09:49:43", "last_updated": "2026-07-20 17:35:28", "protected": false, "protected_to": null, "lang": "zh", "cstr": "11738.11.ncdc.landcover.db7473.2026", "i18n": { "en": { "title": "OasisMap30: a 30 m annual land cover dataset of China's oases from 1987 to 2024", "ds_format": "GeoTIFF", "ds_source": "", "ds_quality": "", "ds_ref_way": "", "ds_abstract": "
High-temporal and spatial resolution oasis land cover maps are of great value in clarifying the ecological evolution and social development process of arid areas. However, research on oases started late, coupled with the high fragmentation of oasis landscapes and frequent conversion of land cover types, constructing high-temporal and spatial resolution oasis land cover datasets faces many challenges. Therefore, such data products specifically for oases are still scarce.\r\n
Based on this, this paper builds a 30-meter land cover mapping framework every year, integrating Landsat satellite images, machine learning algorithms, time series segmentation model (LandTrendr) and principal component analysis methods. Based on this framework, the 30-meter annual land cover data set (OasisMap30) of China oases from 1987 to 2024 is generated based on the Google Earth Engine (GEE) platform. Accuracy verification based on more than 6300 visual interpretation samples shows that the overall accuracy of OasisMap30 is higher than 90%. Combining visual interpretation samples with third-party verification data to conduct cross-comparison among multiple products, OasisMap30 has significant advantages in classification accuracy and error control. In addition, comparing this dataset with a variety of 30-meter special products on impervious surface, cultivated land, and surface water bodies, it was found that OasisMap30 is in good agreement with existing data.\r\n
Based on the OasisMap30 dataset, this paper analyzes the evolution characteristics of land cover pattern in oases in China. The results show that from 1987 to 2024, the total area of oases in my country expanded by 45.87%, with an increase of 7.75 million hectares. The core driving force for expansion was cultivated land reclamation and grassland restoration. Among them, 4.04 million hectares of desert were transformed into grassland, and 3.19 million hectares of desert were reclaimed into cultivated land; impervious surface expanded by 580,000 hectares, and surface water bodies increased by 350,000 hectares; at the same time, there were a large number of internal land types, such as 3.12 million hectares of grassland converted into cultivated land.\r\n
In summary, this set of continuous time-series and high-resolution OasisMap30 dataset can provide strong support for research on oasis landscape pattern evolution, social-ecological response, spatial pattern optimization, etc., and help the sustainable development of oasis regions.\r\n
The dataset is divided into 7 types of land cover. The codes in the data and the types of land cover they represent are as follows: 1: impervious surface, 2: surface water, 3: cultivated land, 4: shrubs, 5: forest land, 6: grassland, 7: bare land.", "ds_time_res": "", "ds_acq_place": "China", "ds_space_res": "", "ds_projection": "", "ds_process_way": "", "ds_ref_instruction": "" } }, "submit_center_id": "ncdc", "data_level": 0, "recommendation_value": 0, "license_type": "https://creativecommons.org/licenses/by/4.0/", "doi_reg_from": "reg_local", "cstr_reg_from": "reg_local", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "is_paper_in_submitting": false, "belong_to_nieer": false, "allow_update_data": false, "ds_topic_tags": [ "绿洲", "土地覆被", "LandTrendr", "干旱区" ], "ds_subject_tags": [ "自然地理学" ], "ds_class_tags": [], "ds_locus_tags": [ "中国" ], "ds_time_tags": [ 1987, 1988, 1989, 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023, 2024 ], "ds_contributors": [ { "true_name": "陈鹏", "email": "pengchen24@swu.edu.cn", "work_for": "西南大学地理科学学院", "country": "中国" }, { "true_name": "唐强", "email": "qiangtang@swu.edu.cn", "work_for": "西南大学地理科学学院", "country": "中国" }, { "true_name": "刘焱序", "email": "yanxuliu@bnu.edu.cn", "work_for": "北京师范大学地理科学学部 陆地表层系统科学与可持续发展研究院", "country": "中国" }, { "true_name": "王帅", "email": "shuaiwang@bnu.edu.cn", "work_for": "北京师范大学地理科学学部 陆地表层系统科学与可持续发展研究院", "country": "中国" } ], "ds_meta_authors": [ { "true_name": "陈鹏", "email": "pengchen24@swu.edu.cn", "work_for": "西南大学地理科学学院", "country": "中国" } ], "ds_managers": [ { "true_name": "陈鹏", "email": "pengchen24@swu.edu.cn", "work_for": "西南大学地理科学学院", "country": "中国" } ], "category": "基础地理" }