School of Biological and Marine Sciences, University of Plymouth, Drakes Circus, Plymouth, PL4 8AA, UK.
Department of Animal Science, University of California, Davis, CA, 95616, USA.
Environ Monit Assess. 2019 Dec 5;192(1):11. doi: 10.1007/s10661-019-7980-4.
Engineered structures in the open ocean are becoming more frequent with the expansion of the marine renewable energy industry and offshore marine aquaculture. Floating engineered structures function as artificial patch reefs providing novel and relatively stable habitat structure not otherwise available in the pelagic water column. The enhanced physical structure can increase local biodiversity and benefit fisheries yet can also facilitate the spread of invasive species. Clear evidence of any ecological consequences will inform the design and placement of structures to either minimise negative impacts or enhance ecosystem restoration. The development of rapid, cost-effective and reliable remote underwater monitoring methods is crucial to supporting evidence-based decision-making by planning authorities and developers when assessing environmental risks and benefits of offshore structures. A novel, un-baited midwater video system, PelagiCam, with motion-detection software (MotionMeerkat) for semi-automated monitoring of mobile marine fauna, was developed and tested on the UK's largest offshore rope-cultured mussel farm in Lyme Bay, southwest England. PelagiCam recorded Atlantic horse mackerel (Trachurus trachurus), garfish (Belone belone) and two species of jellyfish (Chrysaora hysoscella and Rhizostoma pulmo) in open water close to the floating farm structure. The software successfully distinguished video frames where fishes were present versus absent. The PelagiCam system provides a cost-effective remote monitoring tool to streamline biological data acquisition in impact assessments of offshore floating structures. With the rise of sophisticated artificial intelligence for object recognition, the integration of computer vision techniques should receive more attention in marine ecology and has great potential to revolutionise marine biological monitoring.
随着海洋可再生能源产业和海洋水产养殖的发展,海洋中越来越多地出现了工程结构。浮式工程结构充当人工斑块礁,提供了新的、相对稳定的生境结构,而这些结构在水层中是不存在的。增强的物理结构可以增加当地生物多样性,有利于渔业,但也可以促进入侵物种的传播。任何生态后果的明确证据都将为设计和布置结构提供信息,以尽量减少负面影响或增强生态系统恢复。开发快速、经济高效和可靠的远程水下监测方法对于支持规划当局和开发者在评估海上结构的环境风险和效益时基于证据的决策至关重要。一种新型的、无诱饵的中层视频系统 PelagiCam 与运动检测软件(MotionMeerkat)一起,用于半自动化监测移动海洋动物群,在英格兰西南部莱姆湾的英国最大的海上绳索养殖贻贝养殖场进行了开发和测试。PelagiCam 在靠近浮动养殖场结构的开阔水域中记录了大西洋鲭鱼(Trachurus trachurus)、鲱鱼(Belone belone)和两种水母(Chrysaora hysoscella 和 Rhizostoma pulmo)。该软件成功地区分了有鱼和无鱼的视频帧。PelagiCam 系统提供了一种具有成本效益的远程监测工具,可简化海上浮动结构影响评估中的生物数据采集。随着用于物体识别的复杂人工智能的兴起,计算机视觉技术的集成应该在海洋生态学中得到更多关注,并且具有彻底改变海洋生物监测的巨大潜力。