Use of Ensemble Model for Modeling the Larval Fish Habitats of Different Ecological Guilds in the Yangtze Estuary

Author:

Wan Rong12,Song Pengbo1,Li Zengguang234,Long Xiangyu5,Wang Dong2,Zhai Lu2

Affiliation:

1. Fisheries College, Ocean University of China, Qingdao 266000, China

2. National Engineering Research Center for Oceanic Fisheries, Key Laboratory of Sustainable Exploitation of Oceanic Fisheries Resources (Ministry of Education), College of Marine Sciences, Shanghai Ocean University, Shanghai 201306, China

3. Key Laboratory of Marine Ecological Monitoring and Restoration Technologies, Shanghai 201306, China

4. Joint Laboratory for Monitoring and Conservation of Aquatic Living Resources in the Yangtze Estuary, Shanghai 201306, China

5. Scientific Observing and Experimental Station of Fishery Resources for Key Fishing Grounds, Ministry of Agriculture, Key Laboratory of Sustainable Utilization of Technology Research for Fisheries Resources of Zhejiang Province, Marine Fisheries Research Institute of Zhejiang, Zhoushan 316000, China

Abstract

Estuaries provide nursery habitats for a variety of fish species of different ecological guilds and have complicated environmental conditions. In this study, we applied ensemble species distribution models (SDMs) to four abundant and different ecological guild larval fish species (Hemiculter bleekeri (freshwater guild), Pseudolaubuca sinensis (freshwater guild), Coilia mystus (brackish water guild), and Engraulis japonicus (marine guild)) to explore their suitable habitats and the fish–habitat relationships in the Yangtze Estuary. The results showed that random forest (RF) demonstrated the best performance in all single algorithms and the surface range envelop (SRE) model was scarcely accurate. The ensemble SDMs demonstrated a superior predictive ability compared with any single algorithm, with the true skill statistic (TSS) and the area under the receiver operating characteristic curve (AUC) scores being above 0.899 and 0.641, respectively. Binary presence–absence maps showed the different spatial distribution patterns of the four species. We primarily found the freshwater species (P. sinensis and H. bleekeri) present in the South Branch (west of 122° E), whereas we found the marine species (E. japonicus) anywhere except inside the South Branch (west of 121.8° E). The area for P. sinensis (1615.93 km2) was relatively larger than that for H. bleekeri (1136.87 km2). We predicted that the brackish water species (C. mystus) would most likely be present inside the North Branch (west of 122° E), Eastern Chongming, and outside the South Branch (east of 121.8° E). Salinity, as a key environmental variable, contributed to the spatial variability. A low salinity (sea surface salinity (SSS) < 3) was beneficial for P. sinensis and H. bleekeri but was not suitable for E. japonicus. The SSS suitable range for C. mystus was 5–10 and 12–20. Multiple ecological guild species dwelled in the confluence of salt and fresh water. Our results will play an important role in the design of specific conservation strategies for fishery resources in this area.

Funder

Shanghai Municipal Science and Technology Commission Local Capacity Construction Project

Science and Technology Commission of Shanghai Municipality

Publisher

MDPI AG

Subject

Ecology,Aquatic Science,Ecology, Evolution, Behavior and Systematics

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

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