Harpacticoida, George & Veit-Köhler & Arbizu & Seifried & Rose & Willen & Bröhldick & Corgosinho & Drewes & Menzel & Moura & Schminke, 2014
|
publication ID |
https://doi.org/ 10.1007/s13127-013-0154-2 |
|
persistent identifier |
https://treatment.plazi.org/id/03A63301-AC22-A804-FCB7-6E0F384AFBAD |
|
treatment provided by |
Felipe |
|
scientific name |
Harpacticoida |
| status |
|
Species of Harpacticoida
Recently, Baguley et al. (2006) reported 696 harpacticoid species (distributed over 22 1 families) from the northern Gulf of Mexico. This impressively high number arises, however, from 43 stations covering an area of 1,130 x 273 km (see Fig. 1 View Fig in Baguley et al. 2006). Moreover, station depths range from 212 to 3,150 m, from the continental shelf down to the lower abyssal. Compared to those data, the 682 species collected from only two deep-sea stations in the Angola Basin during the DIVA 1 expedition is remarkable and unexpected. Although being separated by approximately 300 nautical miles, both stations were of approximately the same depth (5,400 m) and were in a generally uniform environment ( Fiege et al. 2010; Kröncke and Türkay 2003). Also, 87.8 % of the species were sampled at Station 346 alone: 600 Harpacticoida species at one deep-sea station is the highest ever record from any deep-sea locality so far.
The high percentage (99.3 %) of undescribed species is greater than any other known record (e.g. George 2005, 84.4 %; George and Schminke 2002, 96.4 %; Shimanaga et al. 2004, p. 1099, did not identify to species level). However, in shallower studies, from the sublittoral down to the abyssal (cf. Humes 1994; George 2005, Schminke 2007; Seifried 2004; Veit-Köhler et al. 2010), a high number of previously unknown Harpacticoida species are generally recorded. Thus, in a global context, most Harpacticoida species are doubtlessly unknown to science. Consequently, a significant increase in taxonomic research is needed to address this situation (e.g. Boero 2010; Brökeland and George 2009; Mallet and Willmott 2003; Wheeler 2004; Zhang 2008).
Despite our attempt to optimise the experimental sampling design by conducting eight MUC deployments per station and treating each deployment as a single replicate consisting of five (randomly chosen) cores, some restrictions to quantitative analysis still remain. For instance, no statement can be made regarding the biased occurrence of a species, particularly if it was recorded in low density. The absence of a taxon from one station is not necessarily evidence for its absence from that locality but may rather reflect an insufficient sampling effort (see above). Even if a species was quite frequent at one station but not reported from the other one (e.g. Argestes angolaensis ), a possible presence at the latter station could not be excluded. Bradya kurtschminkei and Emertonia andeep ( Veit-Köhler, 2004) , for example, were not recorded at Station 325, but B. kurtschminkei is known from the Guinea Basin, the Cape Basin, and the Porcupine Abyssal Plain ( Seifried and Martínez Arbizu 2008), and E. andeep has been reported from the Guinea Basin and the Weddell Sea ( Gheerardyn and Veit-Köhler 2009).
Awide distribution in the deep sea has been shown previously for species of the family Paramesochridae ( Gheerardyn and Veit-Köhler 2009) and the argestid genus Mesocletodes ( Menzel 2011a; Menzel et al. 2011). Also most of the five species already known show a wide distribution range: Mesocletodes robustus shows an Atlantic-wide distribution and was found also in Antarctic waters, the Mediterranean, and in the Eastern Pacific in depths ranging from 219 down to 5,000 m (see Menzel et al. 2011 for review); Microsetella norvegica is distributed worldwide, due to its planktonic living; however, its report from the deep sea is the first (cf. World Register of Marine Species, http:// www.marinespecies.org); Selenopsyllus dahmsi was found in the Antarcic Weddell Sea at 2,000 m depth ( Moura and Pottek 1998), and Styracothorax gladiator was collected northwest of Manila ( Philippines, Pacific Ocean) in 2,050 m depth ( Huys 1993). In contrast, Marsteinia parasimilis is the only species that had been reported previously from the Walvis Ridge in the Angola Basin ( Dinet 1974).
Such wide distribution of the mentioned species is thought possible because, among other factors, environmental conditions change only slightly even over great distances (cf. Ramírez-Llodra et al. 2010; Thistle 2003; Türkay 2006; Tyler 2003). Against this background, it seems somewhat astonishing that only 13.2 % of the species recorded here in the Angola Basin (90 out of 682 species) were present at both stations. The stations were not separated by geological barriers, had the same depth and a similar sediment composition (but lower productivity) ( Kröncke and Türkay 2003). At Station 346, a total of 600 species were recorded, of which 510 were exclusive to this site. Only 14.8 % of these 600 species (90 species) were also recorded at Station 325, where an additional 82 species were recorded. Thus from a total of 172 species, Station 325 shared 47.7 % species with Station 346. At Station 325, the low densities of species and individuals led to under-sampling compared to Station 346; this probably concealed much higher species numbers—possibly akin to Station 346. However, to sample these additional species, a much higher number of replicates than at Station 346 would have been necessary.
Community analysis
Qualitative comparisons between Stations 325 and 346 already indicated remarkable differences between their harpacticoid population structures. Species and specimen numbers per core per replicate (five cores per replicate) were much higher at Station 346 than Station 325. Multivariate similarity analysis confirmed this: the stations differed significantly regarding their community structure, as shown both by the MST-test. This is presumably caused by the lower absolute numbers of individuals in the replicates from Station 325. Similarly, differences in the number of individuals and species within Station 325 between cores also resulted in the higher relative differences between replicates in this low-abundance station, compared to the high-abundance Station 346. Hence, the greater dissimilarity of replicates from Station 325, as shown in the nMDS ordination, may be at least partly a mathematical artefact of the lower absolute abundances/ under-sampling at that station.
As demonstrated by J ’ and 1 -Λ values ( Fig. 5C View Fig ), both stations show a high evenness; however, the proportion of singletons was higher at Station 325 (65 %) than at Station 346 (57 %; Appendix A). This meant that at Station 325 replicates were more distinct from each other, than at Station 346, where densities were higher and under-sampling less severe. Data from Station 346 support the conclusions of Rose et al. (2005) regarding harpacticoid within-core diversity: the station showed low local-scale variability with respect to species composition and frequency, with all replicates being quite similar. In contrast, considerable local-scale differences were found at Station 325, and these occasionally exceeded differences at the regional scale (see Fig. 4 View Fig , replicates 325/3, 325/4). However, under-sampling is more severe when densities are low, causing strong “pseudo-turnover” between replicates (i.e. measured turnover due to overlooked species; Whittaker 1998). Pseudo-turnover appears to be a strong factor at Station 325, artificially rendering the replicates from this station more dissimilar than those from Station 346. Nonetheless, the exact proportion of pseudo-turnover affecting our results remains unknown. Thus, for the Harpacticoida we reject the null hypothesis: the harpacticoid assemblages of Stations 325 and 346 are statistically distinct in their structure (taxa composition) and diversity.
Average taxonomic diversity Δ and average taxonomic distinctness Δ*, were slightly higher at Station 325 compared to Station 346. However, due to the currently limited knowledge of deep-sea Harpacticoida this is difficult to interprete. The relatively high Δ and Δ * values at both stations (>70) underline the existence of phylogenetically heterogeneous assemblages that are not formed by different but closely related species deriving from few common ancestors. That would have been a hint for a rather isolated fauna that possibly conquered the environment by radiation. In contrast, it seems more likely that large-scale taxonomic exchange occurs, which has in the meantime been confirmed for Mesocletodes species by Menzel et al. (2011), resulting in quite heterogeneous, species-rich assemblages at Stations 325 and 346, whose diversity becomes manifest even in higher taxonomic levels.
Regarding classical diversity analyses, harpacticoid density (ind./ 10 cm 2) is about 4.5x higher at Station 346 than at Station 325 and, if comparing absolute S, Station 346 is much more diverse than Station 325. If including abundance, N, in an estimation of species diversity, H ’, this still holds true: H ’ ranges from 3.21 to 3.84 at Station 325, and from 4.30 to 4.94 at Station 346. However, when comparing species numbers accounting for abundance by rarefaction, diversity differences are not significant [e.g. for E (S 30)]. Even though in this study the Multicorer sampling effort per abyssal deep-sea station was greater than ever before, our results indicate potential under-sampling at Station 325; we might have recorded higher species richness, similar to Station 346, if a greater number of individuals had been collected.
Furthermore, a qualitative look at the data reveals that, on average, at Station 325 every second specimen resulted in an additional species (N / S =2.03), whereas at Station 346 three specimens were needed to add another species (N / S =3.01). This again qualifies the characteristics of higher diversity, as S, at the northern Station 346, and is supported also by evenness J ’, which is higher at Station 325 than at Station 346. It was estimated by Simpson’ s 1 -Λ whether the number of frequent species at stations 346 and 325 was high or low, the index calculating the probability that two specimens chosen randomly belong to the same species. As the index was almost identical at the two stations, the probability of finding two specimens of the same species is not significantly different between Stations 325 and 346. The values are high for both stations, emphasising the low number of common species .
As also shown by Rose et al. (2005) for within-core alphadiversity of harpacticoid copepods, a higher measured species richness was found at Station 346, since almost all cores from that station provided more species and specimens than those of Station 325. The almost perfect linearity seen in Fig. 7 View Fig indicates that the more specimens collected, the more species found, and that, even with the high sampling effort employed here, we are probably far from the asymptotic part of the collectors curve. Assuming severe under-sampling at least for Station 325, an increase in N might have led to a linear increase in S at that station for a considerable number of further individuals, possibly reaching similar species numbers as Station 346 with similar individual numbers. This leads to the conclusion that the difference in collected species numbers between the stations may be explained by species density rather than by species richness. However, under-sampling was probably also an issue for Station 346 as indicated by the linearity of the regression curve for this station also. This is confirmed by both the Rarefaction analysis as well as by extrapolative species estimation, as both methods suggest possible hyper-diversity but at the same time under-sampling.
Different productivity levels between the stations might explain the different densities recorded. This issue was discussed extensively by Rose et al. (2005) for both stations and can be summarised as follows: in contrast to Station 325, Station 346 was positioned in an area of upwelling with higher productivity. Under the assumption that deep-sea habitats generally show lower productivity compared to other marine or terrestrial habitats ( Grassle 1989; Thiel 1983; Tietjen 1992; Valiela 1995), productivity can be regarded as a major limiting factor in deep-sea environments under normal circumstances. A peak or subsequent descending, productivity–diversity relationship will probably not be reached within the abyssal productivity range (following Rosenzweig 1995, p. 351: “As productivity rises from very low to moderate levels, diversity also rises”). However, we have shown that, at family-level, the descending part of the unimodal productivity–diversity curve might already have been reached. Thus, large-scale heterogeneity in food availability could be an important factor in structuring harpacticoid communities in the abyss of the Angola Basin, and possibly also in other deep-sea regions ( Rose et al. 2005). Thus, it can be stated that Stations 325 and 346 differ noticeably in some aspects of species diversity, but less in others. By showing this, our investigation reflects the complexity of the term ‘diversity’. The study gives evidence for pronounced structural difference between the harpacticoid communities at Stations 325 and 346, which is expressed mainly by different abundance and species densities, probably caused by varying food availability in the Angola Basin.
No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.