The floodplain of the lower Yellow River is an important part of the Yellow River Basin. Establishing a long-term sequence and high-precision land use/cover change dataset for the floodplain of the lower Yellow River，reconstructing its historical trajectory, and grasping the current pattern are fundamental tasks of great significance for the high-quality development of the floodplain of the lower Yellow River and scientific research on related issues. This paper selects 36 scenes of Landsat5 TM imagery from 1985 to 2010, 8 scenes of GF-1 WFV imagery and 12 scenes of Landsat8 OLI imagery from 2015 to 2020. Using methods such as Random Forest, human-computer interaction interpretation, and accuracy evaluation with confusion matrices, we have developed land use/cover change data for the floodplain of the lower Yellow River from 1985 to 2020, with a total of 8 periods every 5 years. The spatial resolution of the data is 30m, with an overall accuracy exceeding 90% and a Kappa coefficient higher than 0.85. This dataset can provide scientific data support for research on regional ecological environmental protection, territorial spatial planning, and the coupling processes, mechanisms, and measurements of natural-human systems in typical regions of the Yellow River Basin.
This dataset contains a total of 8 periods of land use/cover change data for the years 1985, 1990, 1995, 2000, 2005, 2010, 2015, and 2020, with a spatial resolution of 30m. The land use classification standard for the floodplain of the lower Yellow River includes five land use/cover types: cultivated land, water bodies, construction land, forest and grassland, and unused land. The coordinate system adopted is WGS_1984 UTM zone_50N. A data sample is provided for the land use/cover change data of the floodplain of the lower Yellow River in 2020.
The development of the 30m resolution LUCC (Land Use/Cover Change) data for the floodplain of the lower Yellow River from 1985 to 2020 was primarily carried out using ENVI 5.3 and ArcGIS 10.5 software platforms. Multi-source remote sensing imagery, such as Landsat5 TM, Landsat8 OLI, and GF-1 WFV, was used as the data source, and human-computer interaction interpretation methods were employed to complete the task. The main operational procedures included: preprocessing of remote sensing data, determination of land use classification standards, selection of interpretation markers and establishment of sample databases, interpretation of remote sensing imagery, and accuracy assessment.
The data collection, selection, and preprocessing of remote sensing imagery, as well as the selection and establishment of interpretation markers and sample databases, all conform to operational specifications with strict quality control. Additionally, a confusion matrix was used to verify the accuracy of the produced dataset. Both the overall accuracy and Kappa coefficient are above 85%. The overall data quality is good.
This work is licensed under a
Creative
Commons Attribution 4.0 International License.
CSTR:11738.11.NCDC.YRIVER.DB3984.2023
10.57760/sciencedb.j00001.00781