Copepods of the genus Calanus are the key components of zooplankton. Understanding their response to a changing climate is crucial to predict the functioning of future warmer high-latitude ecosystems. Although specific Calanus species are morphologically very similar, they have different life strategies and roles in ecosystems. In this study, C. finmarchicus and C. glacialis were thoroughly studied with regard to their plasticity in morphology and ecology both in their preferred original water mass (Atlantic vs. Arctic side of the Polar Front) and in suboptimal conditions (due to, e.g., temperature, turbidity, and competition in Hornsund fjord). Our observations show that ‘at the same place and time’, both species can reach different sizes, take on different pigmentation, be in different states of population development, utilize different reproductive vs. lipid accumulation strategies, and thrive on different foods. Size was proven to be a very mutable morphological trait, especially with regard to reduced length of C. glacialis. Both species exhibited pronounced red pigmentation when inhabiting their preferred water mass. In other domains, C. finmarchicus individuals tended to be paler than C. glacialis individuals. Gonad maturation and population development indicated mixed reproductive strategies, although a surprisingly similar population age structure of the two co-occurring species in the fjord was observed. Lipid accumulation was high and not species-specific, and its variability was due to diet differences of the populations. According to the stable isotope composition, both species had a more herbivorous diatom-based diet in their original water masses. While the diet of C. glacialis was rather consistent among the domains studied, C. finmarchicus exhibited much higher variability in its feeding history (based on lipid composition). Our results show that the plasticity of both Calanus species is indeed impressive and may be regulated differently, depending on whether they live in their ‘comfort zone’ or beyond it.

The study was performed in the Polar Front region in the southern part of the West Spitsbergen shelf in July 2018. It was designed to collect Calanus either from the water mass they originate from (C. finmarchicus from the Atlantic Water (AT station) domain, from the West Spitsbergen Current that carries warm and saline Atlantic Water and C. glacialis from the colder and fresher Arctic-type Sørkapp Current (AR station)), or from the Hornsund fjord, where two species co-occur, both in the main fjord basin (F station) and in the glacial bay (G station). The station representing the Atlantic water domain (AT) was characterized by the highest water temperature, salinity and chlorophyll fluorescence (Table 1). The lowest water temperature, salinity and chlorophyll fluorescence were observed at the station located in the glacial bay (G). However, the glacial bay was characterized by extremely high concentrations of particles.

At each sampling station, two tows of a WP2 net (180 µm mesh size) were performed thorough the water column (70 m at G station, 200 m at F station, 170 m at AR station) or through the upper 200 m at the deep-water AT station. The sample from the first haul of the net was immediately fixed in a formaldehyde-borax solution for the analysis of Calanus abundance, developmental stage composition and gonad maturation stages. The samples from the next net hauls were put in buckets filled with seawater close to the in situ temperature. Then, the fifth copepodite stages of specific species (distinguished as small and large size modes) were photographed and preserved for further analyses.

Overall, 1 252 photos of Calanus were taken and analyzed to determine the prosome length, the area of the lipid sac and the pigmentation. Methods for specific morpho-ecological traits were as follows:

a. Size and species identification verification

The prosome length of 1252 photographed Calanus CV copepodites was measured in ImageJ/Fiji free software for image analyses from the tip of the cephalosome to the distal lateral end of the last thoracic somite. The medians and quartiles of prosome length measurements are presented as box plots. Additionally, 991 copepodites were measured from the formaldehyde-fixed samples, among which 372 were CV copepodites, for which kernel density size distribution curves were fitted.

Genetic species identification was used to test the accuracy of the applied morphological method of species discrimination for further analyses. Genetic identification was performed for each individual using its antennae and a set of six nuclear insertion–deletion markers (InDels) in a multiplex polymerase chain reaction (PCR) following the protocol described in Choquet et al. (2017). Accuracy of the morphological identification was calculated as the percent between the number of correct assessments in relation to the number of all assessments. Overall, the accuracy was 93%, but the accuracy was also calculated separately for each species and sampling location (Table 2). The species discrimination was  100% accurate at the fjord station (F) and almost fully accurate at the AR and AT stations. In the Atlantic domain, 93% accuracy was obtained. In the Arctic domain, one individual with a size of 2.93 mm, initially identified as C. glacialis, turned out to be C. finmarchicus. A high percent of the species misidentification (22%) occurred only in the glacial bay G station. All the C. glacialis were properly identified, but many of the small copepodites, assumed to represent C. finmarchicus, turned out to be small C. glacialis, with individuals as small as 2.39 mm.

a. Red pigmentation

The quantification of astaxanthin content was based on high performance  chromatography (HPLC) according to the methods described in Stoń-Egiert & Kosakowska 2005. Astaxanthin was isolated from previously lyophilized and weighed Calanus individuals by mechanical grinding in 90% acetone and sonication (2 min, 20 kHz, Cole Parmer, 4710 Series) for 2 h in darkness. Then, after clarification, the extract was subjected to chromatographic analysis. The HP1200  system (Agilent, Perlan Technologies) was equipped with a C18 LichroCART™LiChrospher™ 100 RP18e (Merck) analytical column (dimensions 250 x 4 mm, particle size 5 μm and pore size 100 Å). Pigments were separated in a gradient mixture of methanol, 1 M ammonium acetate and acetone. Calibration was conducted with commercially available standards (The International Agency for 14C Determination DHI Institute for Water and Environment in Denmark), which allowed for the qualitative assessment  of astaxanthin (based on retention time and similarity with the absorbance spectrum of the standards) and  quantitative assessment  (based on response factor values obtained during the calibration procedure).

Moreover, the pigmentation was analyzed by the visual examination of photos taken of live individuals by a color coding scheme, depending on whether there was a full (> 50%, red), mid (10-50%, zebra), slight (<10%, piece) or no (0%, transparent) coloration of antennae and prosome. The photos were also used to calculate the average pigmentation, where the rate of pigmentation was scored for each part of the body from 0 to 3 (e.g. antennae, prosome)  and 0 to 1 (swimming legs, urosome) and summed, considering that approximately 10% of the pigmentation may be contributed by the urosome, 10% from the legs, 20% from the prosome and 60% from the antennas.

b. Population development

The relative abundances of each of the Calanus copepodite stages were used to describe the population age structure. The abbreviations (CI-AF) refer to six successive copepodite stages of Calanus, i.e., CI, CII, CIII, CIV and CV refer to the first five copepodite stages, and AF to adult females. The species identification of particular developmental stages was based on the size discrimination assessed for the populations in the Hornsund fjord (Weydmann & Kwaśniewski, 2008). The age structure of specific species in the water domains studied was tested by Fisher’s exact test in R (fisher.test(data)), an independence test to determine if there is a significant relationship between two categorical variables.

The gonad maturation stages of Calanus females were determined as described by Niehoff (2007) and Niehoff & Runge (2003). In our case, we distinguished stage G3, characterizing females that are preparing for spawning (multiple layers of oocytes in both anterior and posterior diverticula); stage G3.5, indicating females that are ready to spawn (ventral layer of oocytes is already colored, but not yet fully brown); and G4, representing females that would spawn within hours (oocytes undergoing final maturation in the most ventral layer in the gonads). The spent stage (S) characterizes females that had already finished reproduction.

c. Lipid content

Lipid content was determined as the mass of lipids that was extracted for the fatty acid composition analyses as well as estimated by the measurements of the lipid sac area. The lipid sac area was manually measured by contouring the sac perimeter by hand in all the photographed copepods. Furthermore, the specific equations derived from the reliable calibration of the individual lipid contents of Arctic copepods (Vogedes et al., 2010) were applied to calculate the total lipid content. Then, the percentage of the lipid sac area (fulfilment) was calculated as a function of the total area of the prosome. To verify if the lipid content differed between the two species, the unpaired two-sample Mann-Whitney test was performed, and to verify if the lipid content differed among study locations and species, the non-parametric Kruskal-Wallis test was performed in R (Package “stats”). The output of the Kruskal-Wallis test indicated if there is a significant difference between groups, but to know which pairs of groups are different, the function pairwise.wilcox.test was used to calculate pairwise comparisons between groups.

d. Diet

To verify the source of stored lipids in of the two species occurring at the same time in diverse water masses and in fjord waters, where they coexist, a combination of stable isotope (δ¹³C, δ¹⁵N) and fatty acid composition analyses were performed.

Stable isotope analysis was performed following the protocol of Lebreton et al., (2012). Samples were analyzed using an elemental analyzer (Flash EA 1112, Thermo Scientific, Milan, Italy) coupled to an isotope ratio mass spectrometer (Delta V Advantage with a Conflo IV interface, Thermo Scientific, Bremen, Germany). The results are expressed in the δ unit notation as deviations from standards (Vienna Pee Dee Belemnite for δ13C and N2 in air for δ15N) following the formula: δ13C or δ15N = [(Rsample/Rstandard) - 1] x 103, where R is 13C/12C or 15N/14N, respectively. Calibration was performed using reference materials (USGS-24, IAEA-CH6, IAEA-600, USGS-61, USGS-62 for carbon; IAEA-N2, IAEA-NO-3, IAEA-600, USGS-61, USGS-62 for nitrogen). Analytical precision based on the analyses of acetanilide (Thermo Scientific) used as laboratory internal standard was \0.1 and \0.15 ‰ for carbon and nitrogen, respectively.

The identification and quantification of fatty acid methyl esters (FAMEs) was determined by gas chromatography/mass spectrometry (GC/MS) according to the method of Brown et al., 2011. An internal standard (10 μL; 1 mg mL^–1 nonadecanoic acid) was added to lyophilized and weighed Calanus samples. Samples were then saponified (20% KOH; 70°C; 60 min). Fatty acids were obtained by the addition of concentrated HCl (0.5 mL) to the saponified solutions followed by extraction into hexane (3x1 mL). Fatty acids were then methylated (1 mL; 1:9 HCl:MeOH; 80°C; 60 min) and re-extracted in hexane (3x1 mL) prior to analysis (Shimadzu QP2010 gas chromatograph coupled to a QP2020 quadrupole EI mass spectrometer; HP5ms). FAMEs were identified by comparison to authenticated standards (Supelco 37 Component FAME Mix), retention times and mass spectral library matches (>95% confidence). Instrumental abundances were normalized to the internal standard and sample mass for quantification.