Spatiotemporal pattern analysis of land use in Jiangsu Province based on long-term time series remote sensing images
Studying spatiotemporal patterns of land use is crucial for optimal land resource allocation and sustainable development. This study utilizes the Google Earth Engine (GEE) platform and long-term remote sensing imagery data, selecting Jiangsu Province as a case study area. Principal Component Analysi...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tamkang University Press
2025-06-01
|
| Series: | Journal of Applied Science and Engineering |
| Subjects: | |
| Online Access: | http://jase.tku.edu.tw/articles/jase-202601-29-01-0001 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839655280336437248 |
|---|---|
| author | Zhendong Ji Lingzhi Yin Jinhong Wang |
| author_facet | Zhendong Ji Lingzhi Yin Jinhong Wang |
| author_sort | Zhendong Ji |
| collection | DOAJ |
| description | Studying spatiotemporal patterns of land use is crucial for optimal land resource allocation and sustainable development. This study utilizes the Google Earth Engine (GEE) platform and long-term remote sensing imagery data, selecting Jiangsu Province as a case study area. Principal Component Analysis (PCA) was applied to reduce feature dimensionality, and the Random Forest classification algorithm was optimized with Bayesian Optimization and Tree-structured Parzen Estimators (TPE) for improved performance. The classification achieved an overall accuracy of 92% and a Kappa coefficient of 0.89. Spatiotemporal clustering was conducted at the optimal scale, determined using landscape pattern indices, to analyze the land use evolution from 2000 to 2020. The study results indicate that: (1) PCA effectively reduced feature redundancy, enabling a more robust classification process, while Bayesian optimization improved the model’s predictive performance. (2) Cropland area continuously declined, built-up land expanded significantly, waterbody areas decreased slightly,
and forest coverage remained stable. The main transitions occurred between built-up land and cropland, as well as between waterbodies and both cropland and built-up land. (3) From 2000 to 2010, rapid urbanization led to substantial expansion of built-up land, particularly in coastal areas, south of the Yangtze River, and northern cities, causing significant cropland loss and ecological degradation. Post-2010, land use policies helped curb cropland loss. These findings offer valuable insights into land use patterns in Jiangsu, supporting effective land resource management and planning. |
| format | Article |
| id | doaj-art-f71d35d6a2904b9c9ecf9c7d271dacb8 |
| institution | Matheson Library |
| issn | 2708-9967 2708-9975 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Tamkang University Press |
| record_format | Article |
| series | Journal of Applied Science and Engineering |
| spelling | doaj-art-f71d35d6a2904b9c9ecf9c7d271dacb82025-06-25T10:53:43ZengTamkang University PressJournal of Applied Science and Engineering2708-99672708-99752025-06-0129111210.6180/jase.202601_29(1).0001Spatiotemporal pattern analysis of land use in Jiangsu Province based on long-term time series remote sensing imagesZhendong Ji0Lingzhi Yin1Jinhong Wang2College of computer science and technology, Zhejiang Sci-Tech University, Hangzhou, ChinaCollege of computer science and technology, Zhejiang Sci-Tech University, Hangzhou, ChinaZhejiang academy of surveying and mapping, Hangzhou, ChinaStudying spatiotemporal patterns of land use is crucial for optimal land resource allocation and sustainable development. This study utilizes the Google Earth Engine (GEE) platform and long-term remote sensing imagery data, selecting Jiangsu Province as a case study area. Principal Component Analysis (PCA) was applied to reduce feature dimensionality, and the Random Forest classification algorithm was optimized with Bayesian Optimization and Tree-structured Parzen Estimators (TPE) for improved performance. The classification achieved an overall accuracy of 92% and a Kappa coefficient of 0.89. Spatiotemporal clustering was conducted at the optimal scale, determined using landscape pattern indices, to analyze the land use evolution from 2000 to 2020. The study results indicate that: (1) PCA effectively reduced feature redundancy, enabling a more robust classification process, while Bayesian optimization improved the model’s predictive performance. (2) Cropland area continuously declined, built-up land expanded significantly, waterbody areas decreased slightly, and forest coverage remained stable. The main transitions occurred between built-up land and cropland, as well as between waterbodies and both cropland and built-up land. (3) From 2000 to 2010, rapid urbanization led to substantial expansion of built-up land, particularly in coastal areas, south of the Yangtze River, and northern cities, causing significant cropland loss and ecological degradation. Post-2010, land use policies helped curb cropland loss. These findings offer valuable insights into land use patterns in Jiangsu, supporting effective land resource management and planning.http://jase.tku.edu.tw/articles/jase-202601-29-01-0001land use classificationlong-term time seriesspatiotemporal pattern evolutiongoogle earth engine |
| spellingShingle | Zhendong Ji Lingzhi Yin Jinhong Wang Spatiotemporal pattern analysis of land use in Jiangsu Province based on long-term time series remote sensing images Journal of Applied Science and Engineering land use classification long-term time series spatiotemporal pattern evolution google earth engine |
| title | Spatiotemporal pattern analysis of land use in Jiangsu Province based on long-term time series remote sensing images |
| title_full | Spatiotemporal pattern analysis of land use in Jiangsu Province based on long-term time series remote sensing images |
| title_fullStr | Spatiotemporal pattern analysis of land use in Jiangsu Province based on long-term time series remote sensing images |
| title_full_unstemmed | Spatiotemporal pattern analysis of land use in Jiangsu Province based on long-term time series remote sensing images |
| title_short | Spatiotemporal pattern analysis of land use in Jiangsu Province based on long-term time series remote sensing images |
| title_sort | spatiotemporal pattern analysis of land use in jiangsu province based on long term time series remote sensing images |
| topic | land use classification long-term time series spatiotemporal pattern evolution google earth engine |
| url | http://jase.tku.edu.tw/articles/jase-202601-29-01-0001 |
| work_keys_str_mv | AT zhendongji spatiotemporalpatternanalysisoflanduseinjiangsuprovincebasedonlongtermtimeseriesremotesensingimages AT lingzhiyin spatiotemporalpatternanalysisoflanduseinjiangsuprovincebasedonlongtermtimeseriesremotesensingimages AT jinhongwang spatiotemporalpatternanalysisoflanduseinjiangsuprovincebasedonlongtermtimeseriesremotesensingimages |