Copepod species abundance from the Southern Ocean and other regions ( 1980 – 2005 ) – a legacy

This data collection originates from the efforts of Sigrid Schnack-Schiel (1946–2016), a zooplankton ecologist with great expertise in life cycle strategies of Antarctic calanoid copepods, who also investigated zooplankton communities in tropical and subtropical marine environments. Here, we present 33 data sets with abundances of planktonic copepods from 20 expeditions to the Southern Ocean (Weddell Sea, Scotia Sea, Amundsen Sea, Bellingshausen Sea, Antarctic Peninsula), one expedition to the Magellan region, one latitudinal transect in the eastern Atlantic Ocean, one expedition to the Great Meteor Bank, and one expedition to the northern Red Sea and Gulf of Aqaba as part of her scientific legacy. A total of 349 stations from 1980 to 2005 were archived. During most expeditions depth-stratified samples were taken with a Hydrobios multinet with five or nine nets, thus allowing inter-comparability between the different expeditions. A Nansen or a Bongo net was deployed only during four cruises. Maximum sampling depth varied greatly among stations due to different bottom depths. However, during 11 cruises to the Southern Ocean the maximum sampling depth was restricted to 1000 m, even at locations with greater bottom depths. In the eastern Atlantic Ocean (PS63) sampling depth was restricted to the upper 300 m. All data are now freely available at PANGAEA via the persistent identifier https://doi.org/10.1594/PANGAEA.884619. Abundance and distribution data for 284 calanoid copepod species and 28 taxa of other copepod orders are provided. For selected species the abundance distribution at all stations was explored, revealing for example that species within a genus may have contrasting distribution patterns (Ctenocalanus, Stephos). In combination with the corresponding metadata (sampling data and time, latitude, longitude, bottom depth, sampling depth interval) the analysis of the data sets may add to a better understanding how the environment (currents, temperature, depths, season) interacts with copepod abundance, distribution and diversity. For each calanoid copepod species, females, males and copepodites were counted separately, providing a unique resource for biodiversity and modelling studies. For selected species the five copepodite stages were also counted separately, thus also allowing the data to be used to study life cycle strategies of abundant or key species.

Copepoda (Crustacea) are probably the most successful metazoan group known, being more abundant than insects, 30 although far less diverse (Humes, 1994;Schminke, 2007). They occur in all aquatic ecosystems, from freshwater 31 to marine and hypersaline environments, and from polar waters to hot springs (Huys and Boxshall, 1991).

32
Although copepods are evolutionary of benthic origin (Bradford-Grieve, 2002), they have also successfully 33 colonised the pelagic marine environment where they can account for 80 -90% of the total zooplankton abundance 34 (Longhurst, 1985). In the Southern Ocean, copepods are next to Antarctic krill and salps the most important

Nansen net 95
During the expeditions PS04, DAE1979/80, and JB03 net sampling was carried out with a Nansen net (Table 1).

97
Thus, it is possible to sample discrete depth intervals to study the vertical distribution of zooplankton. The Nansen 98 net has an opening of 70 cm diameter and is usually 3 m long. Two different mesh sizes were used: 200 µm for 99 the cruises PS04 and JB03, and 250 µm for DAE1979/80. To conduct discrete depth intervals the net is lowered 100 to maximum depth and then hauled to a certain depth and closed via a drop weight. Then the net is hauled to the 101 surface and the sample is removed. This process of sampling depth intervals can be repeated until the surface layer 102 is reached. The volume of filtered water was calculated using the mouth area and depth interval due to the lack of 103 a flowmeter.

Multinet systems 106
Most presented data sets are based on plankton samples taken with a multinet system (MN) from Hydrobios (Table   107 1) a revised version (Weikert and John, 1981) of the net described by Be et al. (1959). The multinet is equipped 108 with five (midi) or nine (maxi) plankton nets, with a mouth area of 0.25 and 0.5 m 2 , respectively. These nets can 109 be opened and closed at depth on demand from the ship via a conductor cable. Thus, they allow sampling of 110 discrete water layers. The net system was hauled with a general speed of 0.5 m/s. Mesh sizes varied between the 111 data sets from 55 to 300 µm (

Metadata 184
Each data set has its own persistent identifier. The metadata are consistent among all data sets, thus ensuring the 185 comparability of the data sets and document their quality.

186
The following metadata can be found in each data set: 187 -"Related to:" includes the corresponding cruise report, related data sets and scientific articles that might 188 have used part of the data previously.      Fig. 2 a, b). The highest number of zooplankton samples 227 was taken in the 1980s (Fig. 2 b). In the 1980s the sampling effort was concentrated to the Antarctic Peninsula, 228 the Scotia Sea and the Weddell Sea (Fig. 2 a). Samples were taken in multiple years. In the 1990s until 2005 most 229 samples were taken in the Bellingshausen and Amundsen Sea, with fewer samples in the western and eastern 230 Weddell Sea. Two transects were sampled across the Weddell Sea in the 1990s in austral summer and autumn 231 ( Fig. 2 b). In general, most stations were sampled during summer (December to February), followed by autumn 232 (March to May) and spring (September to November), while winter samples are only available from 1986 in the 233 eastern Weddell Sea (Fig. 2 b, c). Summer and autumn samples are widely distributed from the Amundsen Sea to 234 the eastern Weddell Sea (Fig. 2 b), while spring and autumn samples are mostly present from the Scotia Sea and 235 Eastern Weddell Sea. Most samples were taken in January and February (Fig. 2 d). Samples are scattered 236 throughout the entire day (Fig. 3.).

Copepoda 247
In total, specimens from six copepod orders were recorded in the compiled data sets.  Table 2). All 262 calanoid species were counted separately as females, males and copepodites. For selected species also the five 263 copepodite stages were counted individually (Table 3) Microcalanus and Spinocalanus are considered as mesopelagic to bathypelagic. Thus, they are often not found at 288 shallow stations (< 300 m depth). In case of the sea ice-associated S. longipes, low sea-ice conditions and offshore 289 stations may have caused the restricted distribution. S. longipes occurred mainly in the upper water layers, but was 290 also recorded with low abundances in deeper layers (Fig. 4). This pattern may be due to its life cycle, shifting 291 seasonally from a sea-ice associated to a bentho-pelagic life cycle (Schnack-Schiel et al., 1995).

Other Copepoda 294
In total, 28 non-calanoid taxa were recorded. Four data sets provide only abundance and distribution data for non-295 calanoid copepod orders (PS06, PS10, PS29, PS35; Table 1), in particular on species of the order Cyclopoida from 296 the families Oithonidae (2 species) and Oncaeidae (6 species; Table 2). They were separated in female, male, 297 copepodite stages 1, 2, 3, 4, and 5. During VH1094 also Oithona species were identified (Table 2). In all other 298 data sets species of these two families were not separated. In all regions representatives of the family Lubbockiidae 299 were recorded. In the subtropical and tropical samples of PS63, M44/2 and M42/3 also abundances of species of 300 the families Corycaeidae and Sapphirinidae, and of the genus Pachos were recorded. Except for PS65, species of 301 the order Harpacticoida were not separated. In the latter five species were identified, mainly sea-ice associated 302 harpacticoids (

304
In most data sets, copepod nauplii are also recorded as one parameter. However, due to the small size of nauplii 305 they were not sampled quantitatively and should be discarded in further analysis.

Further remarks on the usage of the data compilation 308
Generally, the cruise reports have been linked to each data set. The cruise reports provide valuable information on 309 the itinerary, zooplankton sampling procedures and on other scientific activities on-board that could be useful for 310 the data analysis (e.g. CTD data). We have also added scientific article that are related to individual data sets.

311
Abundance data of selected species and data sets have been published previously in scientific articles. These 312 articles are linked to the respective data sets (under "Related to").

313
To use the data they can be downloaded individually as tab-delimited text files or altogether as a .zip file to allow 314 an import to other software e.g. in R (R core team, 2018) or Ocean Data View (Schlitzer, 2015) for further analysis.

315
Due to the consistent taxonomic nomenclature the individual files can be concatenated easily. It should be kept in 316 mind however, that not all data sets are directly comparable due to difference in net type and mesh sizes (see 317 section 2.2.4). In these cases we recommend to use only presences and absences of the species.  Table 1 Figure 4 24 Earth Syst. Sci. Data Discuss., https://doi.org/10.5194/essd-2018-36