...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Estuarine Coastal and Shelf Science >Applying multibeam sonar and mathematical modeling for mapping seabed substrate and biota of offshore shallows
【24h】

Applying multibeam sonar and mathematical modeling for mapping seabed substrate and biota of offshore shallows

机译：应用多波束声纳和数学建模绘制海上浅层海底基质和生物区系

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

6 Both basic science and marine spatial planning are in a need of high resolution spatially continuous data on seabed habitats and biota. As conventional point-wise sampling is unable to cover large spatial extents in high detail, it must be supplemented with remote sensing and modeling in order to fulfill the scientific and management needs. The combined use of in situ sampling, sonar scanning, and mathematical modeling is becoming the main method for mapping both abiotic and biotic seabed features. Further development and testing of the methods in varying locations and environmental settings is essential for moving towards unified and generally accepted methodology. To fill the relevant research gap in the Baltic Sea, we used multibeam sonar and mathematical modeling methods generalized additive models (GAM) and random forest (RF) together with underwater video to map seabed substrate and epibenthos of offshore shallows. In addition to testing the general applicability of the proposed complex of techniques, the predictive power of different sonar-based variables and modeling algorithms were tested. Mean depth, followed by mean backscatter, were the most influential variables in most of the models. Generally, mean values of sonar-based variables had higher predictive power than their standard deviations. The predictive accuracy of RF was higher than that of GAM. To conclude, we found the method to be feasible and with predictive accuracy similar to previous studies of sonar-based mapping. (C) 2017 Elsevier Ltd. All rights reserved.
机译：6基础科学和海洋空间规划都需要有关海底栖息地和生物群的高分辨率空间连续数据。由于常规的逐点采样无法详细涵盖较大的空间范围，因此必须补充遥感和建模才能满足科学和管理需求。原位采样，声纳扫描和数学建模的结合使用正成为绘制非生物和生物海底特征的主要方法。在朝着统一和普遍接受的方法学发展的过程中，在不同位置和环境设置中进一步开发和测试方法至关重要。为了填补波罗的海的相关研究空白，我们使用了多波束声纳和数学建模方法，广义加性模型（GAM）和随机森林（RF）以及水下视频，以绘制海底基质和近海浅滩的底栖动物。除了测试所提出的复杂技术的一般适用性之外，还测试了基于声纳的不同变量和建模算法的预测能力。在大多数模型中，平均深度及其后的反向散射是影响最大的变量。通常，基于声纳的变量的平均值具有比其标准差更高的预测能力。 RF的预测准确性高于GAM。总而言之，我们发现该方法是可行的，并且具有与以前基于声纳测绘的研究相似的预测准确性。 （C）2017 Elsevier Ltd.保留所有权利。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号