Spatiotemporal dynamics of the Liaohe Delta over the past 30 years (1987– 2017) based on an integrated classification and preferred features approach

Author:

Sun Jihong1,Su Guohui1,Song Huairong1,Wei Helong1,Liu Jingpeng1,Lin Wenrong1

Affiliation:

1. Qingdao Institute of Marine Geology

Abstract

Abstract Mapping wetlands and monitoring spatiotemporal variabilities in wetland regions are useful for providing basic ecosystem-monitoring data that are necessary for the protection and management of wetlands. The main objective of this work was to propose a new approach for monitoring the spatiotemporal patterns and reclamation of coastal wetlands in the Liaohe Delta region from 1987 to 2017. With the proposed approach, we aimed to improve the classification accuracy by using integrated classification and a preferred features method. First, after preprocessing the remote sensing data representing the four years of 1987, 1997, 2007 and 2017, we extracted the first component of the principal component analysis (PCA1), normalized difference vegetation index (NDVI), modified normalized difference water index (MNDWI), soil index (SI), index-based build-up index (IBI), and tasseled cap transformation (TCT) values of the characteristic parameters, such as the brightness component, and then used the maximum likelihood classifier (MLC) and decision tree (DT) methods to classify the preprocessed image landscapes. Finally, we combined the results of the two classification methods with the optimal characteristic parameter band to form new data images and applied the MLC method to perform landscape classification. The analytical results showed that the proposed method can obtain a high average accuracy of 87.71% and a kappa coefficient of 0.85, reflecting a 16.50% higher average accuracy and a 20.72% higher kappa coefficient than the MLC results (average accuracy of 75.29% and kappa coefficient of 0.71). These results indicate that the proposed method is effective and feasible for long-term landscape dynamics research. By using this method, the landscape distributions of the Liaohe Delta wetlands in 4 periods were obtained. We found that although the area of reed wetlands in the Liaohe Delta region was reduced from 1987–1997 (from 1284.44 km2 in 1987 to 1006.70 km2 in 1997), the results were very good in the later periods, indicating optimized wetland protection (from 1040.20 km2 in 2007 to 1275.53 km2 in 2017). The coastal zone changed significantly throughout the study period, especially from 2007–2017; during this period, the coastline was significantly affected by human activities, and large areas of tidal flats and coastal suaeda were converted into salt pans and aquaculture areas, while ports, piers, and urban construction areas also continued to extend to the shallow-sea areas (resulting in the coastline land area increasing by 263.24 km2).

Publisher

Research Square Platform LLC

Reference32 articles.

1. Contrast analysis of supervised classification and decision tree method to extract Qinzhou Bay wetland;Xiaoli Chang;Chinese Agricultural Science Bulletin,2014

2. Spatiotemporal dynamics of coastal wetlands and reclamation in the Yangtze Estuary during past 50 years(1960s–2015);Chen Lin Ren;Chinese Geographical Science,2018

3. Principal component analysis applied to a time series of MODIS images: the spatio-temporal variability of the Pantanal wetland, Brazil;Almeida TIR;Wetlands Ecol Manage,2015

4. Comparison of machine learning methods of land use/land cover classification in the complicated terrain regions;Gu Xiaotian Gao;Remote Sensing Technology and Application,2019

5. Transformation of vegetation structure in China’s Liaohe Delta during 1988–2006;Ji;Chinese Journal of Plant Ecology,2010

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3