Use of stable nitrogen (815N) isotopes in food web of the Adriatic Sea, Croatia
Uporaba stabilnega dušikovega izotopa (815N) v prehranjevalni verigi Jadranskega morja, Hrvaška
Petra Žvab1, *, Tadej Dolenec1, Sonja Lojen2, Goran Kniewald3, Josip vodopija4, Živana Lambaša Belak5, Matej Dolenec1
University of Ljubljana, Faculty of Natural Sciences and Engineering, Aškerčeva 12,
SI-1000 Ljubljana, Slovenia 2Jožef Stefan Institute, Jamova 39, SI-1000 Ljubljana, Slovenia 3Ruder Boškovic Institute, Bjienička 54, 10000 Zagreb, Croatia 4Dalmar d.o.o., Obala kralja Petra Krešimira IV 64, 23211 Pakoštane, Croatia 5Šibenik-Knin County Administration, Vladimira Nazora 1, 22000 Šibenik, Croatia
Corresponding author. E-mail: petra.zvab@guest.arnes.si Received: November 6, 2009	Accepted: December 11, 2009
Abstract: This study presents the first attempt to classify organisms in the food web of the Adriatic Sea into trophic levels using a nitrogen (S15N) isotope tracer. Three main trophic levels were identified with significantly different nitrogen values, ranging from primary producers to higher consumers. TL-1 represents plankton and POM samples, TL-2 mostly benthic organisms and TL-3 fishes. The results also indicate the influence of anthropogenic pollution, which significantly increases nitrogen values.
Izvleček: Raziskava je prvi poskus klasifikacije organizmov v prehranjevalni verigi Jadranskega morja v trofične nivoje na podlagi izotopske sestave dušika S15N. Organizmi so bili generalno razvrščeni v tri trofične nivoje, ki vključujejo primerke od primarnih producentov do višjih konzumentov. TL-1 pomeni plankton in POM, TL-2 večinoma bentoške organizme in TL-3 ribe. Na izotopsko sestavo dušika pomembno vpliva antrop-ogeno onesnaževanje, neobdelane komunalne in industrijske odplake, kot tudi marikulturne dejavnosti, ki občutno povišajo vrednosti dušikovega izotopa S15N.
Key words: food web, trophic levels, nitrogen (S15N) isotope composition, Adriatic Sea. Ključne besede: prehranjevalna veriga, trofični nivoji, izotopska sestava dušika, Jadransko morje
Introduction
In ecological studies, stable isotopes, mostly carbon and nitrogen, are commonly used to trace food webs and distinguish between natural and anthropogenic sources. Using stable isotopes the sources and manner of feeding in marine ecosystems can be determined. The isotopic distribution in animals reflects the isotopic composition of their food and their position in the food web. The nitrogen (515N) values in animal bodies are usually more positive than those in their food (DeNiro & Epen-stein, 1981; Minagawa & Wada, 1984; Fry, 1988). Heavy isotope enrichment for nitrogen is about 1.3-5.3 %% per trophic level and is more useful for studying the relations between trophic levels than that for carbon, for which enrichment is estimated to be about 1 % per trophic level (Minagawa & Wada, 1984 Peterson & Fry, 1987; Fry, 1988). On the basis of stable isotope (ô15N, 513C) composition, many researchers have tried to classify organisms into groups (trophic levels) in different aquatic and terrestrial ecosystems (Corbisier et al., 2006; Fry, 1988; Hobson et al., 2001; Iken et al., 2005; Wada et al., 1993).
The stable nitrogen isotopes also reflect unusual anthropogenic food sources that can increase or decrease nitrogen values (Fry, 1988). For this reason the interpretation should consider natural and also anthropogenic factors. Stable nitrogen isotopes can be used to track untreated communal and industrial sewage as well as animal excrement. Many papers have shown the negative influences of untreated municipal and industrial sewage (Constanzo et al., 2001; Dolenec & Vokal, 2002; Dolenec et al., 2005, 2006b, 2006c, 2007; Rogan et al., 2007) and fish farm activities on marine ecosystems (Dolenec et al., 2006a, 2007; La Rosa et al., 2001; Mirto et al., 2002; Pergent et al., 1999; Rogan et al., 2007; Ruiz et al., 2001; Sara et al., 2004). In addition, variations in stable nitrogen (515N) within the same species could be the result of seasonal effects (Constanzo et al., 2001), size and age effects (Minagawa & Wada, 1984; Wada et al., 1993), or depth effects (Saino & Hatory, 1980; Muscatine & Kaplan, 1994).
The aim of this paper is to present the relationships between stable nitrogen (515N) values in organisms in the
Adriatic Sea and their nutrition and position in the food web. For this purpose selected organisms, from primary consumers to higher consumers, were studied, and on the basis of nitrogen (S15N) values were classified into different trophic levels.
Materials and methods Study area
The study area included mostly coastal areas of the Northern and Central Adriatic Sea in Croatia. Samples were collected at several locations along the coast of Piran bay and Istra Peninsula (from Savudrija to Pula), and inshore and offshore areas of Mutter Sea, the semi-enclosed Pirovac Bay, and sites around the Kornati Islands and Korcula Island (Figure 1). Some sampling locations were highly polluted by human sewage and industrial effluents, especially coastal areas in the vicinity of larger cities, marinas and ports, as well as by fish farming activities that produce N-rich waste of fish excrement and uneaten food. Sampling was also carried out at relatively unaffected sites to obtain information about the natural variability of stable nitrogen (S15N) isotope composition in the Adriatic Sea.
Sample collection and analysis
Several marine organisms as well as particulate organic matter (POM) were sampled for isotopic analyses. Samples
were collected from spring to autumn (May to September) from 2000 to 2008, but the seasonal and annual variations are not the topic of this study.
Water for particulate organic matter (POM) was collected at about 1 m depth at different sampling sites with different amounts of anthropogenic pollution. 5 L to 10 L of water was filtered through glass fibre microfilters (GF/C, Whatman). Filters were freeze-dried in liofilizator and prepared for isotopic analyses.
Plankton samples were collected with a plankton net at 50-0 m depth and other organisms (anemones, sponges, molluscs, sea urchins, crustaceans, fishes) by scuba diving from about 25 m depth. For this study average values of soft parts of tissues were presented. All samples were stored in plastic capsules or bags and refrigerated immediately after collection. In the laboratory, samples were rinsed with deionized water and freeze-dried for at least 72 h. Dried samples were homogenized and crushed to a fine powder by grinding in an agate mortar. Powdered samples were packed into tin capsules and were preserved in desiccators at room temperature until the analyses were carried out.
Nitrogen (515N) isotopic composition was measured using a Europa 20-20 mass spectrometer with an ANCA
Figure 1. Map of the study area in the Adriatic Sea (Croatia) and locations of groups A, B and C with different amounts of anthropogenic pollution
SL preparation module (PDZ Europa Ltd., U. K.). The results were expressed in the standard 515N notation as parts per mille (%o) relative to the nitrogen standard (atmospheric nitrogen, 515N = 0 %o). The analytical precision (1 standard deviation) of triplicate analyses of IAEA N-1 and N-2 standards was better than ±0.16 %o. The precision (1 standard deviation) of
duplicate isotopic analyses of samples was within ±0.2 %o.
Statistical analysis
Statistical analysis of presented data was performed by use of the statistical software program Statistica 6.0. The differences between the groups A, B, C of TL-1 and TL-2 and groups I, II, III of TL-3 were tested with t-test (Table 5).
Results and discussion
In order to understand the energy flow through the complex marine food web in the Adriatic Sea different organisms, from primary producers to upper trophic level consumers, were collected. Sampling was carried out in three areas with different amounts of anthropogenic pollution: (A) no or low anthropogenic impact around the Kornati Islands and Korcula Island, (B) inputs of fish faeces and uneaten food around fish farms near Vrgada Island and (C) municipal, industrial and agricultural inputs in coastal areas around Murter Island, the semi-enclosed Pirovac Bay and the Istra Peninsula (Figure 1).
The lowest nitrogen (S15N) values were generally observed in plankton and POM samples. Plankton values in group A were between +2.5 %o and +4.4 %o, and in the more polluted group C values increased up to +8.0 %o. The mean 515N value of all plankton samples was 5.16 % (Table 1, Figure 2). Similar values but with slightly higher variation were measured in POM, which generally contains a mixture of detri-tic material, phyto- and zooplankton, bacteria and particles from different sources. The nitrogen values varied between -1.27 % and +13.79 % with a mean value of 4.84 %o (Table 2, Figure 3). Extreme values were detected in areas that were highly influenced by human activity (group C) in the high
tourist season, and low values were found in unaffected areas of group A. The relatively large variations in nitrogen (S15N) values of POM and plankton samples could also be related to the rapid movement of floating material owing to the wind, current circulation and tides. Although POM and plankton samples showed notable variations in nitrogen values, they still exhibited the lowest values and were thus classified as trophic level 1 (TL-1).
The second trophic level (TL-2) contained mostly benthic organisms, including anemones (Anemonia sulcata), sponges (Aplysina aerophoba), barnacles (Balanus perforatus), molluscs (Mytilus galloprovincialis, Arca neae, Patella sp., snails), sea urchins (Abracia lixula) and crustaceans (Squilla mantis, Nephrops norvegi-cus). These organisms are sessile or have a very limited habitat and reflect local environmental conditions. The food for these organisms is generally from the lower trophic level TL-1 at the base of the food web. Nitrogen (515N) values varied between -1.30 %o up to +11.87 %o with a mean value of +4.74 % (Table 3, Figure 4). Division into three groups regarding the amount of pollution was also evident in this group. Low values were observed in the non-affected areas (A), medium values around fish farms (B) and the highest values near the larger sources of pollution (C).
Table 1. Basic statistics for nitrogen (515N) isotopic composition of plankton in the Adriatic Sea
Group	N	Min	Max	Median	Means	SD
A	12	2.50	4.40	3.12	3.35	0.61
B	14	3.50	7.30	4.45	4.60	0.85
C	16	4.40	8.00	7.20	7.00	0.87
All groups	42	2.50	8.00	4.55	5.16	1.73
Table 2. Basic statistics for nitrogen (515N) isotopic composition of particulate organic mater (POM) in the Adriatic Sea
Group	N	Min	Max	Median	Means	SD
A	32	1.50	7.80	3.15	3.55	1.50
B	101	1.40	6.80	4.00	3.96	1.03
C	166	-1.27	13.79	5.70	5.61	1.91
All groups	299	-1.27	13.79	4.80	4.84	1.84
Table 3. Basic statistics for nitrogen (515N) isotopic composition of TL-2 in the Adriatic Sea
Group	N	Min	Max	Median	Means	SD
A	100	-1.30	10.60	2.40	2.60	2.13
B	146	1.10	8.80	4.60	4.63	1.43
C	115	0.50	11.87	6.70	6.73	2.39
All groups	361	-1.30	11.87	4.70	4.74	2.53
Table 4. Basic statistics for nitrogen (515N) isotopic composition of fishes in the Adriatic Sea
Groups	N	Min	Max	Median	Means	SD
I	8	6.50	13.45	8.12	9.16	2.48
n	14	6.90	12.70	9.67	9.93	1.79
m	31	7.80	14.70	11.20	11.22	1.90
All groups	53	6.50	14.70	10.60	10.57	2.09
Figure 2. Whisker plots of nitrogen (S15N) values of plankton in locations with different amounts of pollution
Figure 3. Whisker plots of nitrogen (S15N) values of particulate organic matter (POM) in locations with different amounts of pollution
Figure 4. Whisker plots of nitrogen (S15N) values of TL-2 in locations with different amounts of pollution
Figure 5. Whisker plots of nitrogen (S15N) values of TL-3 fish tissues with different modes of nutrition
Table 5. Significance levels of the t-test to evaluate the differences between groups with different amounts of anthropogenic pollution A, B, C of TL-1 and TL-2 and groups I, II and III of TL-3 (ns = not significant, * = p < 0.05; ** = p < 0.01)
TL1	A	B	C
A	-	**	**
B	**	-	**
C	**	**	-
TL2	A	B	C
A	-	**	**
B	**	-	**
C	**	**	-
TL3	I	II	III
I	-	ns	*
II	ns	-	*
III	*	*	-
TL-2 is the most heterogeneous, including a large number of different species. It could probably be further divided into sublevels but the 'state of the art' data did not allow this. Sampling was carried out in different seasons and the number of individual species varied notably (from 5 up to 150 samples). Surprisingly, some species of TL-2 showed significantly lower nitrogen values than those in TL-1, such as Aplysina aerophoba, which has previously been observed and studied in detail by Rogan et al. (2007). Depleted nitrogen values have been interpreted by fractionation of nitrogen isotopes during uptake and assimilation by symbiotic bacteria (Dolenec et al., 2007). Otherwise, the nitrogen (S15N) values of TL-2 organisms were generally enriched compared with those of TL-1.
The highest nitrogen (515N) values were measured in fishes, which comprised the highest trophic level (TL-3) of the samples studied. Within TL-3, fish samples were divided into three sublevels according to their mode of nutrition; I. planktivores and detriti-vores (Liza spp., Boops boops, etc.), II. herbivores (Spondyliosoma cantharus, Diplodus vulgaris, etc.), and III. piscivores (Merluccius merluccius, Ci-tharus linguatula, Phycis, phycis, Mu-raena helena, Dicentrarchus labrax, Seriola dumerili, Conger conger, etc.). The mean nitrogen value increased from group I to III (Figure 5); +9.16 %o for group I, +9.93 % for group II and +11.22 % for group III (Table 4, Figure 5). The isotopic composition of fish tissues depends on their 'squeamishness' and varies within the same group. Ni-
trogen (515N) values are also the result of anthropogenic inputs, which lead to fish migration and/or adjustment to polluted water (Liza spp.).
The results of the t-test of groups within trophic levels are presented in Table 5. T-tests between groups with different amount of pollution A, B, C (Figure 1) of trophic levels TL-1 and TL-2 all showed significant differences at confidence level 99 % (p < 0.01). Slightly lower significant differences were observed between I and III as well as II and III group of TL-3 (confidence level 95 %; p < 0.05), while t-tests between I and II group showed no significant differences.
Conclusion
The nitrogen (S15N) isotope distribution of organisms in the Adriatic Sea can be used to follow the energy flow through the food web of the ecosystem and to classify individual species within trophic levels. Enrichment of nitrogen (S15N) within the whole food web was more than +10 %o, which enabled the assessment and interpretation of feeding habits and relationships between particular organisms. Organisms in the Adriatic Sea were classified into three trophic levels: TL-1 (plankton and POM), TL-2 (mostly benthic organisms; anemones, sponges, barnacles, molluscs, sea urchins and crustaceans) and TL-3 (fishes).
Apart from demonstrating the trophic level of different organisms, this study also showed significant variation in nitrogen (S15N) values within individual species. These fluctuations were related to pollution of the sampling areas. Compared to the natural background, nitrogen (S15N) enrichment was observed around cities, ports and tourist facilities in the coastal regions of Murter Sea, Pirovac Bay and Istra Peninsula. These areas are exposed to untreated municipal, industrial and agricultural pollution that increases the amount of nitrogen (S15N) input into the marine ecosystem. Furthermore, nitrogen (S15N) enrichment could also reflect intensive aquaculture with increased input from fish faeces and uneaten food.
Acknowledgements
This research was financially supported by the Ministry of Higher Education, Science and Technology, Republic of Slovenia (Bilateral projects between Croatia and Slovenia 2001-2009) and Geoexp, d. o. o., Tržič, Slovenia.
References
Constanzo, s. d., Donohue, m. j., Den-nison, w. c., Loneragan, n. r. & Thomas, m. (2001): A new approach for detecting and map-
ping sewage impacts. Marine Pollution Bulletin, Vol. 42, pp. 149-156.
Corbisier, t. n., Soares, l. s. h., Petti, m. a. v., Muto, e. y., Silya, m. h. c., McClelland, j. & Ya-liela, i. (2006): Use of isotopic signatures to assess the food web in a tropical shallow marine ecosystem of Southeastern Brazil. Aquatic Ecology, Vol. 40, pp.381-390.
Deniro, m. i. & Epstein, s. (1981): Influence of diet on the distribution of carbon isotopes in animals. Geochimica et Cosmochimica Acta, Vol. 45, pp. 341-351.
Dolenec, t. & Yokal, b. (2002): Isoto-pic composition of nitrogen and carbon in Anemonia sulcata as a possible indicator of anthropogenic impacts on marine costal ecosystems of the Adriatic. Materials and Geoenvironment, Vol. 49, No. 4, pp. 449-457.
Dolenec, t., Yokal, b. & Dolenec, m.
(2005): Nitrogen - 15 signals of anthropogenic nutrient loading in Anemonia sulcata as a possible indicator of human sewage impacts on marine coastal ecosystem s: a case study of Pirovac Bay and the Murter Sea (Central Adriatic). Croat. Chem. Acta, Vol. 78, No. 4, pp. 593-600.
Dolenec, t., Lojen, s., Lambasa, z. & Dolenec, m. (2006a): Effects of fish farm loading on sea grass Posidonia oceanica at Vargada Island (Central Adriatic): a nitrogen stable isotope study. Isot.
Environ. Health Studies, Vol. 42, No. 1, pp. 77-85.
dOLENEC, t., lOJEN , s., dOLENEC, m., lAM-basa, z., Dobnikar, m. & ro-gan, n. (2006b): 15N and 13C enrichment in Balanus perforatus : tracers of municipal particulate waste in tha Murter Sea (Central Adriatic, Croatia). Acta Chim. Slav., Vol. 53, pp. 469-476.
Dolenec, t., Lojen, s., Kniewald, g., Dolenec, m. & Rogan, n. (2006c): S15N of particulate organic matter and Anemonia sulcata as a tracer of sewage effluent transport in the marine costal ecosystem of Pirovac Bay and the Murter Sea (Central Adriatic). Materials and Geoenviron-ment, Vol. 53, No. 1, pp. 1-13.
Dolenec, t., Lojen, s., Kniewald, g., Dolenec, m. & Rogan, n. (2007): Nitrogen stable isotope composition as a tracer of fish farming in invertebrates Aply-sina aerophoba, Balanus perforatus and Anemonia sulcata in central Adriatic. Aquaculture, Vol. 262, pp. 237-249.
Fry, b. (1988): Food web structure on Georges Bank from stable C, N, and S isotopic compositions. Limnology and oceanography, Vol. 33, pp. 1182-1190.
Hobson, k. a., Fisk, a., Karnovsky, n., Holst, m., Gagnon, j.-m. & Fortier, m. (2002): A stable isotope (S13C, S15N) model for the North Water food web: implications for evaluating trophody-namics and the flow of energy
and contaminants. Deep-Sea Research II, Vol. 49, pp. 51315150.
Iken, k., Bluhm, B.A. & Gradinger, r.
(2005): Food web structure in the high Arctic Canada Basin: evidence from S13C and S15N analysis. Polar Biol., Vol. 28, pp.238-249.
La Rosa, t., Mirto, s., Mazzola, a. & Danavaro, r. (2001): Differential responses of bentic microbes and meiofauna to fish-farm disturbance in coastal sediments. Environmental Pollution, Vol. 112,pp.427-434.
Miniwaga, m. & Wada, e. (1984): Step-wise enrichment of 15N along food chains: Further evidence and the relation between S 15N and animal age. Geochimica et Cosmochimica Acta, Vol. 48, pp.1135-1140.
Mirto, s., La Rosa, t., Gambi, c., Danavaro, r. & Mazzola, a. (2002): Nematode community response to fish-farm impact in the west-em Mediterranean. Environmental Pollution, Vol. 116, pp. 203-214.
Muscatine, l. & Kaplan, i. r. (1994): Resource partitioning by reef corals as determined from stable isotope composition II S15N of zooxanthellae and animal tissue versus depth. Pac. Sci., Vol.48, pp.304-312.
Pergent, g., Mendez, s., Pergent-Martini, c. & Pasquali, v. (1999): Preliminary data on the impact
of fish farming facilities on
Posidonia oceanica meadows in the Mediterranean. Oceanologi-ca Acta, Vol. 22, pp. 95-107.
Peterson, b. ù7ooJ. & Fry, b. (1987): Stable isotopes in ecosystem studies. Annual review ecological systems, Vol. 18, pp. 293-320.
Rogan, n., Dolenec, t., Lojen, s., Lambasa, z.& Dolenec, m. (2007): 15N signal of Aplysina aerophoba as a tracer of anthropogenic nitrogen in the Murter Sea and Pirovac bay (central Adriatic). Materials and Geoen-vironment, Vol. 54, No. 1, pp. 63-76.
ruiz, m., Perez, m. & Romero, j. (2001): Effects of fish farm loadings seagrass (Posidonia oceanica) distribution, growth and photosynthesis. Marine Pollution Bulletin, Vol. 42, pp. 749-760.
Saino, t. & Hatory, a. (1980): 15N natural abundance in oceanic suspended particulate matter. Nature, Vol. 283, pp.752-754.
Sara, g., Scilipoti, d., Mazzola, a. & Modica, a. (2004): Effects of fish farming waste to sedimentary and particulate organic matter in a southern Mediterranean area (Gulf of Castellammare, Sicily): a multiple stable isotope study S13C and S15N. Aquaculture,, Vol. 234, pp. 199-213.
Wada, E., Kabaya, Y. & Kurihara, Y.
(1993): Stable isotopic structure of aquatic ecosystems. J. Bios-ci.,, Vol. 18, No. 4, pp. 483-499.