Jennifer M. Durden

Biological responses to disturbance from simulated deep-sea polymetallic nodule mining

Commercial-scale mining for polymetallic nodules could have a major impact on the deep-sea environment, but the effects of these mining activities on deep-sea ecosystems are very poorly known. The first commercial test mining for polymetallic nodules was carried out in 1970. Since then a number of small-scale commercial test mining or scientific disturbance studies have been carried out. Here we evaluate changes in faunal densities and diversity of benthic communities measured in response to these 11 simulated or test nodule mining disturbances using meta-analysis techniques. We find that impacts are often severe immediately after mining, with major negative changes in density and diversity of most groups occurring. However, in some cases, the mobile fauna and small-sized fauna experienced less negative impacts over the longer term. At seven sites in the Pacific, multiple surveys assessed recovery in fauna over periods of up to 26 years. Almost all studies show some recovery in faunal density and diversity for meiofauna and mobile megafauna, often within one year. However, very few faunal groups return to baseline or control conditions after two decades. The effects of polymetallic nodule mining are likely to be long term. Our analyses show considerable negative biological effects of seafloor nodule mining, even at the small scale of test mining experiments, although there is variation in sensitivity amongst organisms of different sizes and functional groups, which have important implications for ecosystem responses. Unfortunately, many past studies have limitations that reduce their effectiveness in determining responses. We provide recommendations to improve future mining impact test studies. Further research to assess the effects of test-mining activities will inform ways to improve mining practices and guide effective environmental management of mining activities.

Landscape-scale spatial heterogeneity in phytodetrital cover and megafauna biomass in the abyss links to modest topographic variation

Sinking particulate organic matter (POM, phytodetritus) is the principal limiting resource for deep-sea life. However, little is known about spatial variation in POM supply to the abyssal seafloor, which is frequently assumed to be homogenous. In reality, the abyss has a highly complex landscape with millions of hills and mountains. Here, we show a significant increase in seabed POM % cover (by ~1.05 times), and a large significant increase in megafauna biomass (by ~2.5 times), on abyssal hill terrain in comparison to the surrounding plain. These differences are substantially greater than predicted by current models linking water depth to POM supply or benthic biomass. Our observed variations in POM % cover (phytodetritus), megafauna biomass, sediment total organic carbon and total nitrogen, sedimentology, and benthic boundary layer turbidity, all appear to be consistent with topographically enhanced current speeds driving these enhancements. The effects are detectable with bathymetric elevations of only 10 s of metres above the surrounding plain. These results imply considerable unquantified heterogeneity in global ecology.

High resolution study of the spatial distributions of abyssal fishes by autonomous underwater vehicle

On abyssal plains, demersal fish are believed to play an important role in transferring energy across the seafloor and between the pelagic and benthic realms. However, little is known about their spatial distributions, making it difficult to quantify their ecological significance. To address this, we employed an autonomous underwater vehicle to conduct an exceptionally large photographic survey of fish distributions on the Porcupine Abyssal Plain (NE Atlantic, 4850 m water depth) encompassing two spatial scales (1–10 km2) on and adjacent to a small abyssal hill (240 m elevation). The spatial distributions of the total fish fauna and that of the two dominant morphotypes (Coryphaenoides sp. 1 and C. profundicolus) appeared to be random, a result contrary to common expectation but consistent with previous predictions for these fishes. We estimated total fish density on the abyssal plain to be 723 individuals km−2 (95% CI: 601–844). This estimate is higher, and likely more precise, than prior estimates from trawl catch and baited camera techniques (152 and 188 individuals km−2 respectively). We detected no significant difference in fish density between abyssal hill and plain, nor did we detect any evidence for the existence of fish aggregations at any spatial scale assessed.

Perspectives in visual imaging for marine biology and ecology: from acquisition to understanding

1National Oceanography Centre, European Way, Southampton, United Kingdom Email: Jennifer.Durden@noc.soton.ac.uk 2Ocean and Earth Science, University of Southampton, National Oceanography Centre Southampton, European Way, Southampton, United Kingdom 3GEOMAR Helmholtz Centre for Ocean Research Kiel, Kiel, Germany 4CSIRO (Oceans & Atmosphere Flagship), Hobart, Australia 5Australian Centre for Field Robotics, University of Sydney, Sydney, Australia 6Girona University, Girona, Spain 7Life Sciences Department, Natural History Museum, Cromwell Road, London, United Kingdom 8Monterey Bay Aquarium Research Institute, Moss Landing, California, USA 9Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USA 10Japan Agency for MarineEarth Science and Technology, Natsushimacho, Yokosuka, Japan 11Biodata Mining Group, Faculty of Technology, Bielefeld University, Bielefeld, Germany 12Geoscience Australia, Symonston, Australia

Ranking Color Correction Algorithms Using Cluster Indices

In the field of underwater imaging it is often necessary to pre-process images to homogenize image quality across a whole image transect and to compensate variations in imaging conditions. A variety of pre-processing methods have been developed in recent years to overcome different problems occurring in underwater imaging, hence performing different on different image sets. Protocols for an objective comparison and scoring of those methods are needed. Here we show how to use cluster indices to rank the differently methods regarding their performances on different image sets. Our results show different rankings for four preprocessing methods for two chosen sets of benthic images from the deep seafloor.