5
views
0
recommends
+1 Recommend
0 collections
    0
    shares
      • Record: found
      • Abstract: not found
      • Article: not found

      Global patterns of diel vertical migration times and velocities from acoustic data : Global patterns of diel vertical migration

      1 , 2 , 2
      Limnology and Oceanography
      Wiley

      Read this article at

      ScienceOpenPublisher
      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Related collections

          Most cited references46

          • Record: found
          • Abstract: found
          • Article: not found

          Large mesopelagic fishes biomass and trophic efficiency in the open ocean

          Mesopelagic fishes—the small fishes living in the ocean’s twilight zone—form one of the most characteristic features of the open ocean: the deep scattering layer at depths between 200 and 1,000 m, visible in the echosounder display of vessels sailing all oceans1. Whereas the mesopelagic fish genus Cyclothone sp. is likely the most abundant vertebrate on earth2, mesopelagic fishes remain one of the least investigated components of the open-ocean ecosystem, with major gaps in our knowledge of their biology and adaptations, and even major uncertainties about their global biomass. Trawling estimates suggest that the biomass of mesopelagic fishes is ~1,000 million tons3 4, a number commonly used in assessments of ecosystem function and the biogeochemistry of the global ocean5 6. However, even for the original estimate it was stated that ‘most of the gear used to obtain the available information obviously underestimate the biomass present’3, and the efficiency of different types of nets to capture mesopelagic organisms has been further questioned recently, with inter-calibration exercises showing order-of-magnitude differences in the captured biomass depending on the type of gear7. Moreover, trawling-based biomass estimates are systematically below acoustic estimates7 8 9, as mesopelagic fishes have been shown to exhibit escape reactions to nets, rendering trawling data suspect of gross underestimation10. Here we combine a sensitivity analysis of acoustic data collected during Malaspina 2010, the Spanish Circumnavigation Expedition (December 2010–July 2011, Fig. 1a), and modelling, to show that mesopelagic fishes biomass in the open ocean is about one order of magnitude higher than previous estimates. We furthermore examine the mesopelagic fishes biomass relative to primary production (PP) and consider the implications of these estimates for the functioning of the open-ocean ecosystem and biogeochemical cycles. Results Acoustic biomass estimates A Simrad EK60 echosounder operating at 38 kHz frequency was used to obtain data throughout the 32,000-mile voyage (Fig. 1a). We used data obtained during the daytime from 200–1,000 m depths, comprising the main diurnal habitat of mesopelagic fishes, to calculate fish biomass and considering the different sources of uncertainties involved (Methods). The average (±s.d.) nautical area scattering coefficient (s A, m2 nmi−2) in the 200–1,000 m layer for the whole cruise was 1,864±1,341, with individual estimates ranging from 158–7,617 m2 n mi−2 (N=209, Fig. 1a,b). The s A was significantly correlated with the 2010 satellite-derived average daily PP (mg C m−2 d−1)11, (Fig. 2, Pearson’s r=0.77, F corrected for spatial autocorrelation=26, D.F. corrected for spatial autocorrelation=18, P 0.1 m−1 (Supplementary Table 2). Thus, although increased PP in productive areas is likely to increase production at TL2 and TL3, it is also likely, beyond a certain point where high plankton biomass leads to high light attenuation and visual constraints become severe, to decrease the transfer efficiency between mesozooplankton and visual predators34 relative to the transfer efficiency in clear waters at the oligotrophic ocean. The evidence that mesopelagic fishes biomass, and consequently the total fishes biomass, is 10-fold higher, or even more, than previously assumed has important implications for our understanding of the carbon fluxes in the ocean. A discrepancy in the estimates of export production using 234Th:238U disequilibria and shallow sediment traps has been systematically reported35. This discrepancy has generally been attributed to artifacts in the traps36 or to the episodic nature of the sinking events37. A comparison between reverse modelling estimated C export fluxes and sediment traps suggested that shallow traps ( 1,000 m) measured fluxes higher than the modelled ones38. An order of magnitude higher biomass of mesopelagic fishes might explain the difference. Mesopelagic fishes perform diel vertical migration, feeding at night in the upper layers (euphotic) and excreting and respiring at depth at day. This implies that mesopelagic fishes drive a vertical flux from the surface to the mesopelagic layer, bypassing the detection capacity of sediment traps. At the same time, by defaecating in deep waters, faeces produced with surface organic matter start sinking at depths of ~500–700 m, bypassing consumption in a large fraction of the water column and increasing the observed flux in deep traps. Basically, mesopelagic fishes accelerate the flux by actively transporting organic matter in the top layer of the water column, where most organic carbon is lost from the sedimentary particle flux. Del Giorgio and Duarte39, using fishes production estimates derived from fisheries landings (mesopelagic fishes are not commercially fished), considered fishes respiration of little significance for the global ocean. However, a substantial revision of the mesopelagic fishes biomass upwards renders fishes respiration relevant. Assuming fishes respiration to be nine times production39 and half of it to happen in deep waters, up to 10% of the PP could be respired by mesopelagic fishes in deep waters (Table 4). This estimate is again in agreement with the local estimate in the northeast Pacific23. The estimate needs refinement in terms of time occupied and the respiration rates in deep layers (usually more than half of the day, but also at low temperatures), but indicates that, in deep layers, the sum of MFP and respiration is in the order of magnitude needed to explain the discrepancies between 234Th:238U disequilibria and shallow sediment traps. Moreover, the excretion in deep layers of materials ingested by mesopelagic fishes in the surface might partly explain the unexpectedly large microbial respiration in the deep ocean40. Our results strengthen the previous claim that mesopelagic fishes are the most abundant fishes and, indeed, the most abundant vertebrates in the biosphere2. The results from the survey presented here suggest that trophic transfer efficiency from primary producers to fishes has been underestimated in the oligotrophic ocean, with the high transfer efficiency from primary producers to fishes associated with warm water temperatures and extreme water transparency, maximizing prey capture by visual predators. As many mesopelagic fishes, dominant in oceanic areas, are strong vertical migrators, feeding in the upper water column and excreting at depth, these results have important implications for the biogeochemical cycles of the ocean, as these animals provide trophic connectivity and transport organic carbon between the surface and the mesopelagic ocean, and could help explain existing discrepancies between flux estimates obtained by the 234Th:238U method and sediment traps35, as well as the unexpectedly large microbial respiration in deep water40. Even with the current 109 tons estimate mesopelagic fishes are considered to play a key role in the world’s oceans as a link between plankton and top predators41 and in the oxygen depletion of the open ocean deep layers32. With the 10-fold higher biomass found in this study and in recent local studies23 the conclusion about potential impacts of harvesting mesopelagic fishes extends to the global biogeochemical cycles. This finding calls for an effort to improve the accuracy of the estimates of the biomass and composition of the mesopelagic community. A more accurate estimate will require technological developments to increase the capturability of mesopelagic fishes and obtain detailed target strengths, as well as coordinated cruises across representative areas of the world ocean with sufficient resolution to address mesoscale structures. Methods Echosounder Continuous acoustic measurements were made with a calibrated42 Simrad EK60 echosounder (7° beam width), operating at a frequency of 38 kHz and with a ping rate of 1 transmitted pulse per 2 s. The data were stored for later analysis, carried out using the LSSS software43. The echosounder data were episodically affected by noise from various sources; consequently, prior to import into the LSSS software for post-processing, the data were subjected to a series of filters to remove bad data. These filters introduced a bias by removing the highest intensity data, however. The backscatter estimates, which are the basis of our biomass estimates, are therefore conservative. The filters worked by comparing the integrated backscatter over a depth range with the background backscatter over the same depth range. Background intensities were detected using a median filter that was 400 pings wide, updated every 100 pings. Pings affected by attenuation were defined as pings with backscatter >6 dB below the median in either of the depth ranges of 50–600 m or 600–1,000 m. Periods with backscatter >4 dB above background levels in the depth range 800–1,000 m were also marked. Pings tagged by these filters were excluded in further analyses. Lastly, a simple 9-point running median (horizontal) removed shorter irregular spikes. After manual scrutiny of the remaining data, the data were integrated in 2-minute-by-2-metre bins at a threshold of −90 dB. After integration in LSSS, data were imported into R44 for further analysis. Acoustic results were split into day, night and crepuscular data, using the function ‘sunriset’ from the ‘maptools’ package, with crepuscular periods defined as sunset/sunrise±1 h. To plot the echogram (Fig. 1b), the processed LSSS data were exported in 10-minute-by-1-metre bins to Matlab, where the nautical area scattering coefficient (s A, m2 nmi−2) was converted to volume backscattering strength (S v, dB re 1 m−1). The daytime data were extracted (2 h after sunrise and 2 h before sunset) and interpolated to remove the night-time gaps. Longer distances with missing or removed data (between cruise legs) were plotted as white. Satellite data The satellite data we used were all annual averages for the year 2010. Annual averages of PP for 2010 were generated by averaging monthly data for PP downloaded from the Ocean Productivity website ( http://www.science.oregonstate.edu/ocean.productivity/index.php)11. Cruise segments were then generated by combining all position fixes within a startpoint±(8 × 4.6) km in north, south, east and west directions. The startpoint of the next segment was the first position registered outside this box. Alignment of in situ and satellite data was done by selecting the 64 (8 × 8, size per bin ~4.6 × 4.6 km) chlorophyll-a bins along the segments that were closest to the midpoint (median position) within a cruise segment, with the added restriction that no chlorophyll-a bin could be used twice. Values for these 64 bins, corresponding to an area of ~37 × 37 km were averaged. For the other satellite-derived measurements, the maximum and minimum positions of the chlorophyll-a bins were used as boundaries for selection prior to calculation of the averages. Data from the conductivity temperature and depth (CTD) probe were aligned to cruise segments, and only CTD casts within a given cruise segment were used for a given segment. Areas for biomass estimation We used the PP–backscatter relation to estimate the mesopelagic biomass from satellite-derived PP data. We determined the biomass from the sum of the biomasses estimated from satellite-derived PP estimates using only areas where the bottom depth was >1,000 m. We used the ETOPO1 data set ( http://www.ngdc.noaa.gov/mgg/global/global.html) to estimate the area. The bathymetry data set was translated down to a 10′ arc grid, and for every cell in the PP data set grid (~same spatial resolution, not identical grids), we assigned the depth from the closest grid-point in the bathymetry data set. Primary production grid-points/cells with bottom depths shallower than 1,000 m were then excluded from our biomass estimation, as were areas north and south of 40 degrees north and south. This resulted in an area of 222.3 million km2. Temperature data The temperature data to estimate the MFP/biomass ratio (P/B) were obtained from the Malaspina Expedition CTD profiles. Profiles inside the PP boxes, or those closest to the boxes were used. As more data-points were available for daytime, we used the temperature at the daytime weighted mean depth (WMD) of the acoustic data. In areas where day- and night-time acoustic WMDs were available, the average difference between the daytime WMD temperature and the average temperature between day- and night-time WMD was 1 could be explained by factors other than an overestimation of the biomass. In this region, the night-time vertical profiles were particularly close to the surface, suggesting that a large proportion of the biomass actually migrates to the near-surface dead-zone of the echosounder, leading to a lower total night-time backscatter57. In any case the data from that limited area do not influence the estimations of global backscatter using different methods (regressions or average). The spatial coherence of the GWR (GWR parameters and significance) is also a strong indication that resonance does not play a major role. Unless resonance was the same all along the transect (same community composition, sizes, depth distribution and migration patterns all around the world), resonance should eclipse the local relation between s A and PP in areas where resonance was relevant. The GWR does not show such variability and the relation remains significant all along the transect, except for the North East Pacific area with hypoxic deep layers that also show higher day/night ratios (Supplementary Fig. 4). Finally, the general estimate and the estimations along the transect agree with the Ecotroph model results using average parameters. The model is completely independent from acoustics estimates, based on PP and transfer efficiency. The agreement between the two independent approaches suggests that resonance is not a major source of bias in the acoustic data. We use a range of literature db/weight ratio values to estimate biomass. This is a simplification that precludes exact estimation at each single point, but is largely to generate reliable average estimates, as the inaccuracies go both ways. A large portion of the backscatter from an individual fish normally originates from its gas-filled swimbladder52, but in mesopelagic fish reduced swimbladders or fat-filled swimbladders are common58 and have a strong effect on the TS of the fish. For instance, adults of some species of the genus Cyclothone may have gas-filled swimbladders, whereas only juveniles in other Cyclothone species have gas-filled swimbladders. There are other species of the same genus that never have gas-filled swimbladders58. Ground truthing in each area is not possible, but the 25–75% quartile range used in the estimates provided here should encompass the average TS value for the oceanic mesopelagic fish populations. As our focus was mesopelagic fish, we did not include the backscatter from the upper 200 m layer (apart from the test on resonance). The additional fish biomass in the upper 200 m would contribute to a higher biomass estimate of total fish. The integrated s A for the upper layer is on average 27% of the integrated value from 200–1,000 m. However, if five areas with exceptionally high values were excluded, the average value would drop to 7% (Supplementary Fig. 11). Author contributions X.I., C.M.D. and S.K. conceived this project. U.M. and G.B. were in charge of calibration of the echosounders and data collection methodology. J.L.A., A.B., F.E., J.I.G.-G. and S.H.-L. ensured the quality of the data collection during the cruise. T.A.K. and A.R. analysed the data and contributed to interpretation and writing of the paper. S.A. contributed and interpreted the light attenuation data. X.I., S.K., T.A.K, A.R., D.L.A. and C.M.D. wrote and edited the paper. All authors discussed the results and commented on the manuscript. Additional information How to cite this article: Irigoien, X. et al. Large mesopelagic fishes biomass and trophic efficiency in the open ocean. Nat. Commun. 5:3271 doi: 10.1038/ncomms4271 (2014). Supplementary Material Supplementary Information Supplementary Figures 1-11, Supplementary Tables 1-3 and Supplementary References
            Bookmark
            • Record: found
            • Abstract: not found
            • Article: not found

            A review of the adaptive significance and ecosystem consequences of zooplankton diel vertical migrations

              Bookmark
              • Record: found
              • Abstract: not found
              • Article: not found

              Expansion of oxygen minimum zones may reduce available habitat for tropical pelagic fishes

                Bookmark

                Author and article information

                Journal
                Limnology and Oceanography
                Limnol. Oceanogr.
                Wiley
                00243590
                January 2016
                January 2016
                October 27 2015
                : 61
                : 1
                : 353-364
                Affiliations
                [1 ]Department of Atmospheric and Oceanic Sciences; University California, Los Angeles; Los Angeles California
                [2 ]School of Oceanography; University of Washington; Seattle, Washington
                Article
                10.1002/lno.10219
                ed9f9c44-e257-4c7a-8770-73dc87bf29d7
                © 2015

                http://doi.wiley.com/10.1002/tdm_license_1

                http://onlinelibrary.wiley.com/termsAndConditions

                History

                Comments

                Comment on this article