Seaquest DSV: a compact Deep-water Sonar and Visual sampler for exploring the marine twilight zone

深海中的动物，包括各种各样的鱼类、鱿鱼和浮游动物，很难取样，但它们在海洋生态系统功能（例如，它们是金枪鱼和一些鲸目动物等物种的食物）、生物地球化学循环（例如，帮助将大气中的碳输送到深海，缓冲气候变化）中发挥着重要作用，可能是渔民的直接目标。从根本上说，我们需要很好地了解哪些物种在哪里，数量有多大。我们提出了一种声学和光学采样装置，将有助于解决这一问题，通过克服目前的一些采样困难，在中层区域（200至1000米深度范围）打开一个新的窗口。传统的渔网调查表明，中远洋区约有10亿公吨的鱼类。2014年，一个有争议的国际团队提出，这些旧的估计数字太低了一个数量级，实际上可能有超过1万吨的中远洋鱼类。他们的估计是利用单次环球航行收集的单频（38千赫）科学回声测深仪数据得出的。他们假设所有来自中上层的回声能量都来自鱼类，但没有任何净样本来证实这一点。不同大小的鱼返回不同强度的回声能量，一些被认为在中远洋中丰富的浮游动物（虹吸管）具有含气的气团，可以返回比某些鱼类更强的回声。因此，在缺乏物种或大小信息的情况下，任何“鱼类”生物量估算都存在相当大的不确定性，这是由回声强度对鱼类生物量的整体缩放引起的。由于这一总体数字，现在人们对中浮游生物作为一种潜在的主要蛋白质来源的商业兴趣越来越大。作为一个科学界，我们需要更好地了解中上层生物群落的组成，这样我们才能更好地向社会通报生活在那里的生物的生态系统服务以及它们的收获潜力。基本声学理论（例如[2]），我们自己的工作[3]和同事的[4]专注于中上层，表明鱼类和虹吸管不能通过单频采样区分。然而，多频数据可以提供关于大小的信息，在某些情况下，可以实现物种的分离。用于鱼类/浮游动物识别/分级的典型频率范围从几十到几百千赫。声音传播的物理特性限制了该频谱的有效范围在海水中只有几十米，因此为了对中上层进行声学采样，我们需要将回声测深仪降低到深水中。我们提议的工具将使之成为可能。此外，我们将使用立体视频来捕捉一些我们通过声学检测到的生物体的图像。这将使我们能够在频谱范围内确定已知大小（大小影响TS）的物种的声目标强度（TS，从目标反向散射的声能比例的比率测量），从而能够对声学调查数据进行定量评估，并朝着更好地了解全球生物量分布的方向取得进展。声学和立体光学的结合提供了一种创新的、世界领先的中层取样新方法。李建平，陈晓明等。2014。远洋大型中上层鱼类的生物量和营养效率。网络通信，5:3271。2. Simmonds， E.和MacLennan， D. 2005。渔业声学。Blackwell科学有限公司Proud， R.等人。2018。从虹吸管到深散射层：估算全球中远洋鱼类生物量的不确定性范围。冰JMS.4。klose， R. J. et al. 2016。深散射层，气囊密度和尺寸估计使用双频声学和光学探头。冰JMS。73: 2037 - 2048。5. Brierley， A. S.等。1998。南大洋浮游动物的声学识别。DSR第二部分：TSIO。45: 1155 - 1173。

Animals in the deep sea, including a diverse array of fish, squid and zooplankton, are hard to sample, but play important roles in ocean ecosystem function (e.g. they are food for species such as tuna and some cetaceans), biogeochemical cycling (e.g. helping transport atmospheric carbon to the deep sea, buffering climate change), and may be targeted directly by fishers. We need fundamentally to gain a good understanding of which species are where, and in what abundance. We propose an acoustic and optical sampling device that will help with this, opening a new window on the mesopelagic zone (200 to 1,000 meter depth range) by overcoming some present day sampling difficulties.Traditional net surveys suggest that there are about 1,000 million metric tonnes (MT) of fish in the mesopelagic zone. In 2014, an international team suggested, controversially, that these old estimates were an order of magnitude too low, and that there may in fact be more than 10,000 MT of mesopelagic fish [1]. Their estimate was made using single-frequency (38 kHz) scientific echosounder data collected on a single circumnavigation of the globe. They assumed that all of the echo energy from the mesopelagic came from fish but did not have any net samples to confirm this. Different sized fish return different intensities of echo energy, and some zooplankton thought to be abundant in the mesopelagic (siphonophores) have gas-bearing pneumatophores that can return stronger echoes than some fish. In the absence of species or size information, therefore, there is scope for considerable uncertainty in any 'fish' biomass estimate arising from a blanket scaling of echo intensity to fish biomass. Due to this headline figure, there is now growing commercial interest in mesopelagic biomass as a potential major source of protein. We need as a scientific community to better understand mesopelagic community composition so we can better inform society of the ecosystem services of the organisms that live there and their potential for harvest.Basic acoustic theory (e.g. [2]), our own work [3] and that of colleagues [4] focusing on the mesopelagic, has shown that fish and siphonophores cannot be differentiated by single frequency sampling. Multiple frequency data can however give information on size and, in some circumstances, can enable separation of species [5]. Typical ranges of frequencies used for fish/zooplankton identification/sizing range from tens to several hundred kHz. The physics of sound propagation limits the effective range of the high end of this spectrum to a few tens of m in seawater, so in order to acoustically sample the mesopelagic we need to lower the echosounder into deep water. The instrument we propose will enable this. Furthermore, we will use stereo video to capture images of some of the organisms we detect acoustically. This will enable us to determine the acoustic target strength (TS, a ratio measure of the proportion of sound energy backscattered from a target) of species of known size (size influences TS) across a spectrum of frequencies and so enable quantitative evaluation of acoustic survey data and progress towards better understanding of global biomass distribution. Combining acoustic and stereo optics provides an innovative and world-leading new way to sample the mesopelagic.1. Irigoien, X. et al. 2014. Large mesopelagic fishes biomass and trophic efficiency in the open ocean. Nat Comm, 5: 3271. 2. Simmonds, E., and MacLennan, D. 2005. Fisheries acoustics. Blackwell Science Ltd. 3. Proud, R., et al. 2018. From siphonophores to deep scattering layers: uncertainty ranges for the estimation of global mesopelagic fish biomass. ICES JMS.4. Kloser, R. J. et al. 2016. Deep-scattering layer, gas-bladder density, and size estimates using a two-frequency acoustic and optical probe. ICES JMS. 73: 2037-2048. 5. Brierley, A. S. et al. 1998. Acoustic discrimination of Southern Ocean zooplankton. DSR Part II:TSIO. 45: 1155-1173.