Original scientific article	UDC 574.587:581.524.12(262.3)
Received: 2009-04-24
EVIDENCE FOR SEAGRASS COMPETITION IN A CENTRAL CROATIAN
ADRIATIC LAGOON
Claudia KRUSCHEL, Stewart T. SCHULTZ & Melita BACALJA University of Zadar, Department of Maritime Sciences, HR-23000 Zadar, M. Pavlinovica bb, Croatia E-mail: claudia@claudiakruschel.com
Sven DAHLKE
University of Greifswald, Biologische Station Hiddensee, D-18565 Kloster / Hiddensee, Biologenweg 15, Ge rmany
ABSTRACT
Evidence for competition between two co-occurring seagrass species, Zostera noltii and Z. marina, was found along an isobath transect in the Novigrad Sea, Central Croatian Adriatic. The transect included developed (marina) and undeveloped regions of shoreline. Each species' coverage and presence was evaluated using DGPS-tracked underwater videography. Seagrass shoot density was estimated from SCUBA samples. The two Zostera species exhibited opposite significant changes correlated with the environmental gradient: Z. marina cover and shoot density decreased while Z. noltii cover and shoot density increased with distance from marina. Significantly more of the spatial variation was explained by species interaction than by environmental variables. We conclude that competition is a major process structuring the distribution of Zostera species in the Novigrad Sea.
Key words: Zostera, seagrass competition, environmental gradient, DGPS-supported videography, Adriatic
EVIDENZA DI COMPETIZIONE TRA FANEROGAME MARINE NELLA LAGUNA DELL'ADRIATICO CENTRALE (CROAZIA)
SINTESI
L'articolo tratta la competizione esistente fra due specie di fanerogame marine, Zostera noltii e Z. marina, che coabitano nel mare di Novigrad (Adriatico centrale). Il transetto, effettuato lungo un'isobata, ha compreso zone svi-luppate (marina) e non sviluppate della costa. La presenza e la copertura di ogni specie sono state valutate con l'ai-uto della videografia subacquea sostenuta da DGPS. La densita dei ciuffi e stata calcolata dai campioni raccolti durante l'immersione. Per le due specie di Zostera sono state riscontrate variazioni significant opposte correlate con il gradiente ambientale. La copertura e la densita dei ciuffi di Z. marina diminuiscono, mentre la copertura e la densita dei ciuffi di Z. noltii aumentano con la distanza dalla marina. La variazione nello spazio delle specie e stata significativamente chiarita meglio considerando l'interazione fra le specie che le variabili ambientali. Gli autori sostengono che la competizione sia il fattore principale che influenza la distribuzione delle specie di Zostera nel mare di Novi-
Parole chiave: Zostera, competizione tra fanerogame marine, gradiente ambientale, videografia sostenuta da DGPS,
Adriatico
INTRODUCTION
Much information is available on the spatial distribution of seagrasses as a result of the interplay between seagrass and the physical environment, such as light attenuation (reviewed by Leoni et a/., 2008), nutrient concentrations (reviewed by Touchette & Burkholder,
2000),	sediment grain size and organic content (reviewed by de Boer, 2007), salinity (reviewed by Touchette, 2007), temperature (reviewed by Lee et a/.,
2007),	wave exposure (reviewed by Cabago et a/.,
2008),	air exposure during low tides, benthic slope, and depth. These responses are interpreted as an outcome of physiological tradeoffs during the process of adaptation to these physical conditions and resources. Fewer studies have incorporated both physical and biological processes, such as competition and their effect on spatial distribution and abundance (Fourqurean et a/., 1995; Laugier et a/., 1999; Tanaka & Kayanne, 2007), and seemingly none have done so within a single analysis.
Our study focuses on two co-occurring Zostera species, Z. no/tii and Z. marina, along a 3-m isobath transect spanning a developed area (residential and tourist housing, light commercial business, seawall, marina, agricultural fields), to an undeveloped shoreline. The restriction to the 3-m isobath ensured that (i) neither species was investigated at its lower or upper depth limit (depth distribution for both species ranges from 0.25-4.5 m in the study area; Schultz et a/., 2009), that (ii) depth related variation in physical environmental variables was ignored, and (iii) that the two species were co-occurring within most of the study transect and associated plots. This gave us the opportunity to investigate spatial variation of the two species in relation both to physical environmental gradients along the transect and to each other's presence.
Much is known about the two Zostera species, although the majority of studies have been carried out outside the Adriatic and even the Mediterranean. In the Adriatic, Z. no/tii is described as a widely distributed species mainly found in the shallow benthos of a range of habitats from sheltered low energy to higher energy environments with more wave and current activity (Widdows et a/., 2008). Z. marina in the Adriatic is described as the species with the more restricted and irregular distribution, requiring sheltered places and freshwater input (Guidetti, 2000). Although Z. no/tii is considered the more marine species, both species are known to have great euryhalinity tolerance (den Hartog, 1970) and both have been found to benefit from decreased salinities (Hootsmans et a/., 1987; Vermaat et a/., 2000; Charpentier et a/., 2005). Z. no/tii appears to prefer sandy sediments, whereas Z. marina tends to be found on muddier ground with a higher organic content (Duarte & Kalff, 1988; Caniglia et a/., 1992; Koch,
2001).	Z. marina as the larger species is expected to be
more tolerant of low light conditions, e.g. as a consequence of sediment re-suspension, as it has greater light-absorbing surface area and higher resource-storing volume. Both species have been shown to respond positively to addition of ammonium and nitrate in N-limited environments (Marba et a!., 1996), a situation that can be expected near residential areas and freshwater inflow.
The placement of the transect within the study area offered (i) significant variation in physical environmental variables, including salinity, wave exposure, sediment characteristics, distance from anthropogenic influence and freshwater due to the change in geographical location (without depth interference), and (ii) an opportunity to investigate the two Zostera species within a continuous mixed bed. This allowed us to compare the relative importance of the physical environment and species interactions for each species' spatial distribution.
MATERIALS AND METHODS Study area
The Novigrad Sea, Croatia (44°12'N, 15°30'E), is a protected estuarine embayment of 29 km2 (approximately 8 x 5 km at longest axes) in the eastern Adriatic Sea (Fig. 1). It is connected to the Velebit Channel in the north by a narrow strait (the Maslenica Channel), and is receiving freshwater inflow from the Zrmanja River in the northeast, by underground springs, a few small seasonal creeks, and a canal draining water from the agricultural area near the town of Posedarje (Sinovcic et a!., 2004; Figs. 1 and 2). Benthic habitats include dense macroal-gae/rock, unconsolidated bare sediments (gravel, sand, and mud) and sparse to dense seagrass belonging to three species Zostera noltii, Z. marina, and Cymodocea nodosa. The present study was carried out in spring 2007 (salinity measurements in fall and winter) in the low gradient westernmost portion of the Novigrad Sea where the seagrass cover is a continuous meadow extending from the developed north side of the bay (town of Posedarje) to the undeveloped south side (Fig. 2). Here the water is shallower than 5 m and the bottom is muddy to sandy.
DGPS/videography and video analysis
Along a constant depth (3-m isobath) transect of 1167 m length, a video sensor (Sony, 480 colour TVL) continuously recorded the sea bottom. Simultaneously overlaid on the video image was the satellite time recorded every two seconds. Depth was monitored by a 200 KHz, 11°, single-beam transducer. Horizontal DGPS coordinates were taken with real-time submeter accuracy from radio beacon transmissions to a GPS antenna held by kayak operator directly above the video sensor visible from the surface (Norris et al., 1997;
IaJ CROATIA
POSEDA RJE
NOVIGRAD SEA
3m isobath transect
Fig. 1: Approximate location of the 3 m isobath transect at the Novigrad Sea, Croatia. SI. 1: Približna lokacija trimetrskega izobatnega transekta v Novigrajskem morju, Hrvaška.
Dauwalter et al., 2006; Schultz, 2008). Subsequent analysis of every second of the recorded video allowed for calculation of the linear coverage for all seagrass species present.
Seagrass and sediment sampling and processing
Following over the DGPS tracks of the video transect by submeter real-time navigation, a SCUBA diver was led to 32 stations evenly spaced along the previously followed video transect. At each station, the diver harvested all seagrass from three replicate circular plots of 0.085 m2 each, as well as 3 sediment cores (100 ml syringes with stopper) of 3 cm diameter inserted into the bottom to a depth of approximately 8-10 cm. Seagrass and sediment samples were kept refrigerated. Seagrass was processed within 48 hours, sediment within 6 hrs. The uppermost 5 cm of sediment were placed in plastic containers, homogenized by stirring and kept air tight and refrigerated until further processing (within 72 hours), when a portion of the sediment (approximately 30 g) was wet-weighed and then dried at 60°C until constant dry weight. Water content of the sediment was calculated as the weight loss in percent. Loss-on-ignition (LOI) analysis was used to measure the sediment organic
matter content. Small samples of sediment (mean weight 9 g) were dried overnight at 105°C until weight constancy and then combusted at 530°C for 2 h. Organic content was calculated as the weight loss in percent. In seagrass samples from each of the 3 plots at each of the 32 stations the shoots of all species present (Z. marina, Z. noltii, Cymodocea nodosa) were counted. Ten shoots (or less if less available in the sample) from each species were randomly chosen and the length (ruler) and width (dissecting scope) of the longest intact leaf of each shoot was measured and the leaf area calculated.
Salinity, bottom slope, and wave exposure
Water samples were taken from the surface and salinity immediately measured with a refractometer. All samples were taken from a kayak and approximately every 10 m while tracking the 3-m isobath transect with submeter accuracy. Bottom slope was calculated from the shortest distance between the 2 m isobath line and the 4 m isobath line passing through each of the 32 sampling stations on the 3-m isobath line. All isobath lines were identified by kayak-based DGPS tracking with the depth continuously monitored as described above. Depth-dependent wave exposure (REId) was calculated
Fig. 2: Detailed view of the 1,167 m long 3 m isobath transect in the westernmost section of the Novigrad Sea, Croatia. Colour code identifies seagrass species and relative visual presence for individual units of approximately 0.3 m. Legend: C = Cymodocea nodosa, N = Zostera noltii, Z = Zostera marina, NZ = mixed Zostera with Z. marina observed within a 0.3 m unit but not in every video frame, ZN = mixed bed with Z. marina present in every video frame, NSG = no seagrass (unconsolidated sediments).
Sl. 2: Podroben prikaz 1.167 m dolgega in 3 m širokega izobatnega transekta v najzahodnejšem delu Novigra-jskega morja na Hrvaškem. Vrsti sta označeni z različnima barvama v posameznih približno 0,3 m velikih enotah. Legenda: C = Cymodocea nodosa, N = Zostera noltii, Z = Zostera marina, NZ = obe vrsti, pri čemer se Z. marina pojavlja v 0,3 m veliki enoti, vendar ne v vsakem video-okvirju. NSG = brez morskih trav (nestrjene usedline).
using the formulas given in Krause-Jensen et al. (2003). Necessary data on wind direction, velocity and frequency were provided by the Croatian Meteorological and Hydrological Service. Fetch distances were meas-
ured as the distance between a sampling point and the nearest shoreline in each of eight compass directions (N, NW, W, SW, S, SE, E, NE) as measured on satellite photos with resolution 2.5 m per pixel (Google Earth 5.0, 2009).
Data analysis
Effects of physical variables on shoot densities were tested with single- and multiple-factor analysis of variance (ANOVA) on untransformed response variables, which were not significantly different from those normally distributed. Significance of individual coefficients in the ANOVA multiple-factor was evaluated with t-tests. F-test was used to compare the results of two nested analyses of variance that differed only in the presence or absence of seagrass shoot density. Relationships among variables were visualized with a standard biplot of the results of principal components analysis based on correlations among all variables.
RESULTS
Seagrass visual cover along the transect
Along the 1,167 m transect, three seagrass species were observed (Fig. 2). Z. noltii was present over 974 m (83.5% of the transect), Z. marina over 877 m (75.1%) and C. nodosa over 17 m (1.5%). C. nodosa was restricted to one small section of the transect. The Zostera species mostly co-occurred in mixed communities (781 m linear coverage, 66.9% of the transect). Near the marina development, Z. marina had greater visual presence and Z. marina blades were frequent and observed in every video frame (Fig. 2: "ZN", coded dark blue). Further away from the marina, fewer blades of Z. marina were observed within the Z. noltii-dominated seagrass bed (Z. marina present within each base unit of approximately 0.3 m which equals one second of video, but not in every video frame (Fig. 2: "NZ", coded dark green). Monospecific patches of Z. marina (Fig. 2: "Z", coded light blue, 96 m accumulative coverage, 8.2% of the transect) were generally rare and occurred either near the marina or along a heavily frequented beach area (Fig. 2). Monospecific patches of Z. noltii (Fig. 2: "N", coded light green, total coverage 193 m, 16.5%) were twice as common as those of Z. marina, but basically restricted to the natural side, where they were most expansive along the sand bank (Fig. 2) located in the southwest of the study area. Z. marina visual presence increased again (Fig. 2 shift from "NZ" to "ZN") in the area directly influenced by a seasonal creek at the very south end of the transect (Fig. 2).
Tab. 2: Summary of variables representing the changes in physical environment along the transect. For references with geographical locations refer to figure.2.
Tab. 2: Povzetek parametrov, ki ponazarjajo spremembe v fizičnem okolju vzdolž transketa. Glede podrobnosti o geografskih lokacijah na preučevanem območju glej sliko 2.
Tab. 1: Description of the general environment along the transect. For references with geographical locations refer to figure 2.
Tab. 1: Opis okolja vzdolž transekta. Glede podrobnosti o geografskih lokacijah na preučevanem območju glej sliko 2.
Geographical location	Description of the environment
North	Transect located within the marina, a concreted area that experiences regular small boat traffic and mooring, as well as freshwater drainage and wastewater input.
Northwest	Near transect enters a freshwater canal, which receives drainage from agricultural fields and some residential waste water. The mouth of the canal is characterized by dense growth of marsh plants, such as Phrajmites spp.
West	Transect near a low-gradient shallow beach area experiencing frequent bathing and boating/anchoring traffic from May to October.
Southwest	Transect follows the edge of a shallow sand bank, an area where the benthic slope at 3 m is highest.
South	Transect ends near the mouth of a seasonal creek off an undeveloped rocky shore (freshwater entering mainly in early spring).
Physical environmental variable	Summary of results based on measurements
Benthic slope	Generally low, ranges from 0.8 to 1.8 %, lowest values in the north and highest in the southwest.
Sediment organic-matter content and water content	Organic-matter content ranged from 3.0 to 10.9%, water content ranged between 32 and 65%, with highest values in and near the marina and lowest values in the south and along the sand bank (southwest). Sediment organic-matter and water content were highly positively correlated (p = 2.2x10-16, R2 = 0.94).
Salinity	Ranged between 9 to 14 ppt in the winter and 32 and 34 ppt in the fall. In the fall salinity was significantly highest in the center of the transect, in winter salinity was significantly highest near the natural side (south).
Depth-dependent wave exposure index	Ranged between 6.90 and 6.99 and significantly increased with distance from the north and the south ends, but was slightly higher within the marina (6.95) than at the South end (6.90).
Seagrass responses to the physical environment
Mean shoot density of Z. noltii was 656 shoots/m2 and significantly higher than that of Z. marina (111 shoots/m2, p = 5.9 x 10"8). Z. noltii shoot density significantly increased with distance from marina (p = 0.035, R2 = 0.11, Fig. 3) and was positively affected by bottom slope (p = 0.45, R2 = 0.093; Tab. 3). Sediment characteristics, like organic and water content, had a marginally negative effect, while distance to a sandbank had a marginally positive effect (Tab. 3). On the contrary, Z. marina shoot density decreased with distance from marina (p = 0.0003, R2 = 0.34; Fig. 3), was lowest near
(alongside) the sand bank (p = 0.0074, R2 = 0.30), and was negatively affected by benthic slope (p = 0.033, R2 = 0.11; Tab. 4). Z. marina shoot density correlated positively with sediment organic content (p = 0.0056, R2 = 0.20) and water content (p = 0.0037, R2 = 0.22; Tab. 4).
The physical environment along the transect
The 3-m isobath transect reaches north to south from the town of Posedarje to the opposite natural shore (Fig. 2). Tables 1 and 2 summarize and describe changes of the overall environment and specific aspects of the physical environment along the transect.
R2 = 0.11. P = 0.035
Ra = 0.33, p = 0.00031
o •
°0 * - S -:- »oVV#
O Organic matter	0 ° - n
• Water content	Q
Fig. 3: Scatterplots, regression lines, and ANOVA results (distance from marina) for Z. noltii and Z. marina shoot densities, proportion of seagrass that was Z. marina, and sediment organic matter and water content. Sl. 3: Razpršeni diagrami, regresije in rezultati ANOVE (razdalja od marine) za gostoto šopov vrst Z. noltii in Z. marina, delež morske trave, ki je bila Z. marina, organske usedline in vsebnost vode.
The proportion of seagrass showed that Z. marina was significantly correlated with the same predictor variables and in the same direction as Z. marina shoot density. Total shoot density showed less significant correlations with any of the predictor variables, but the directions of all responses were equal to those for Z. noltii shoot density. In a biplot (Fig. 4b), the first two principle components were the distance from the marina and the distance from the freshwater canal outlet (see in Fig. 2). Again, Z. marina shoot density is negatively correlated with the distance from marina and the freshwater canal, while Z. noltii is positively correlated with these two variables. In addition, Z. marina shows positive correlation with wave exposure and negative correlation with the slope, while the opposite is true for Z. noltii.
Tab. 3: ANOVA results for Zostera noltii shoot density; Df = 1.1. Statistically significant results are highlighted (bold).
Tab. 3: Rezultati analize variance (ANOVA) za gostoto šopov vrste Zostera noltii; Df = 1,1. Statistično značilni rezultati so poudarjeni (mastne črke).
Predictor variables	Zostera noltii shoot density		
	F	P	R2
distance from marina	4.86	0.035	0.11
distance from sand bank	3.1	0.088	0.064
distance from beach area	1.01	0.32	0.00037
distance from canal	3.32	0.078	0.060
wave exposure	3.72	0.063	0.080
salinity fall	0.11	0.74	-0.029
salinity winter	0.25	0.62	-0.025
bottom slope	4.18	0.05	0.093
sediment organic matter %	3.74	0.063	0.081
sediment water content %	3.03	0.092	0.061
Seagrass interactions
Z. marina shoot density was highly negatively and significantly correlated with Z. noltii shoot density (p = 9 x 10-11, R2 = 0.35; Fig. 4a). When all physical variables were combined in a multiple regression analysis (p = 0.0019, F = 3.14, Df = 10, 85), only 18% of the variation in Z. noltii shoot density was explained and only fall salinity had an independent significant effect. A subsequent multiple regression analysis with Z. marina shoot density, added as one of the predictor variables (p = 1.0 x 10-7, F = 6.51, Df = 11, 84), significantly improved the model (p = 5.14 x 10-7, F = 29.06, Df = 84, 85 in comparison to the nested ANOVAs; Tab. 5) and explained additional 20.5% of the variation. The only physical variables having an independent significant effect on Z. noltii shoot density were fall salinity (positive), wave exposure (negative), sediment organic content (negative), and distance from a sand bank (negative).
Tab. 4: ANOVA results for Zostera marina shoot density; Df = 1.1. Statistically significant results are highlighted (bold).
Tab. 4: Rezultati ANOVE za gostoto šopov vrste Zostera marina; Df = 1,1. Statistično značilni rezultati so poudarjeni (mastne črke).
Predictor variables	Zostera marina shoot density		
	F	P	R2
distance from marina	16.6	0.00031	0.33
distance from sand bank	14.12	0.00074	0.3
distance from beach area	8.4	0.007	0.19
distance from canal	1.29	0.26	0.0093
wave exposure	3.4	0.07S	0.072
salinity fall	0.19	0.67	-0.027
salinity winter	2.S1	0.12	0.047
bottom slope	S	0.033	0.11
sediment organic matter %	8.89	0.0056	0.2
sediment water content %	9.91	0.0037	0.22
Fig. 4: (a) Scatterplot, regression line, and ANOVA results (Z. marina shoot density) for Z. no/tii shoot density. (b) Standard biplot of the results of a principal components analysis (PCA) based on correlation among all variables. Legend: "dist" = distance, "sal" = salinity, "wavex" = wave exposure index, zoma = Z. marina, zono = Z. no/tii, dens = shoot density, water = sediment water content, organic = sediment organic content.
Sl. 4: (a) Razpršeni diagram, regresija in rezultati ANOVE (gostota šopov vrste Z. marina) za gostoto šopov vrste Z. no/tii. (b) Standard biplot-grafikon rezultatov analize glavnih komponent (PCA) sloni na korelaciji spremenljivk. Legenda: "dist" = razdalja, "sal" = slanost, "wavex" = indeks izpostavljenosti valovom, zoma = Z. marina, zono = Z. no/tii, dens = gostota šopov, water = vsebnost vode v sedimentu, organic = vsebnost organske mase.
DISCUSSION
Both Zostera species were present and mostly co-occurred along the entire transect. There was no obvious spatial segregation, although Z. noltii was the species present over a larger linear distance than Z. marina and at a higher mean shoot density. The shoot densities of Z. noltii (range 0-1600 shoots/m2) compare well with those found in mixed beds by Laugier et al. (1999) in a French Mediterranean lagoon (range 50-2,500 shoots/m2), but were an order of magnitude lower than those found in a monospecific Z. noltii bed in a lagoon in southern Portugal (6,000-8,000 shoots/m2; Cabaço et. al., 2008). Similarly, Z. marina shoot density in the Novigrad Sea (range 0-333 shoots/m2) was lower than described by Guidetti (2000) from monospecific Z. marina beds in the northwestern Adriatic (range 280-775 shoots/m2), but compared well with shoot densities of Z. marina in mixed Zostera beds in a coastal lagoon in France (100180 shoots/m2; Laugier et al., 1999).
Tab. 5: Results of a nested analyses of variance (Df = 11.84) of Zostera noltii shoot density with Z. marina shoot density included as one of the (otherwise physical environmental) predictor variables. Statistically significant results are highlighted (bold). Tab. 5: Rezultati vgnezdene analize variance (Df = 11,84) gostote šopov vrste Zostera noltii, kjer je gostota šopov vrste Z. marina vključena kot ena izmed (sicer fizično okoljskih) prediktorskih parametrov. Statistično značilni rezultati so poudarjeni (mastne črke).
In agreement with the two species' description worldwide, Z. marina was the larger species, with longer and broader leaves, and thus a larger mean leaf area (Z. noltii : 480 mm2, Z. marina: 1385 mm2, p = 8.2 x 10"14),
also indicating a larger biomass per shoot. Individually, physical variables such as salinity, wave exposure, and bottom slope explained none or only very little of the variation in shoot density of Z. marina (0-11%, Tab. 4) and Z. noltii (0-8%, Tab. 3), while sediment characteristics did so to a larger extent (Z. marina: 20-22%, Z. noltii: 6-8%, Tabs. 3 and 4). Z. marina was positively affected by finer sediments with higher organic and water content, while the opposite was true for Z. noltii. Caniglia et al. (1992) showed a similar distribution of Z. marina and Z. noltii in the Venice lagoon based on grain size; Z. noltii occurred on coarse textured sediment (sand), while Z. marina inhabited finer substrate with higher organic content.
Z. marina reached highest shoot density, and highest relative abundance near the marina. This is consistent with the hypothesis that the marina environment increased the competitive ability of Z. marina within the mixed Zostera bed. The taller Z. marina may have been the better competitor in a low light environment associated with frequent sediment re-suspension (muddier sediment, more frequent disturbance), conditions that could have resulted in what we observed: both species grew tallest shoots within and near the marina (data not shown). In contrast to Z. marina, Z. noltii shoot density was highest outside the marina where Z. marina was absent or very sparse. The strong negative correlation of the two Zostera species' shoot densities could be evidence for competition; they explained a higher percentage of variation than any other predictor variable tested, and had an effect with far higher statistical significance than any of the other physical variables in the multiple regression (Tab. 5). We conclude that Z. noltii beds of the Novigrad Sea are likely to be successfully invaded by Z. marina in areas where sediments have a higher organic and water content and/or experience more mechanical disturbance (wave exposure and more frequent anthropogenic traffic). Here Z. marina may be better suited to tolerate mechanical stress by means of deeper and larger rhizomes that are more likely to remain after a severe disturbance (e.g. anchoring) allowing for quick re-growth of new shoots. The occurrences of small monospecific Z. marina patches may be the direct result of such disturbances. Z. noltii appears to be the better competitor on coarser and less disturbed sediments, where its high shoot density may prevent the establishment of Z. marina shoots. Among species with a seasonal growth pattern, the smaller species are known to show faster increase in shoot density than larger species (Marba et al., 1996). Our findings corroborate the description of Z. noltii as the species tolerating high and low energy environments and a broad range of salinities in the Adriatic (den Hartog, 1970). Our results, however, somewhat contradict the finding that Z. marina requires protected areas (Guidetti, 2000). Most relatively wave protected areas in Adriatic bays are actually under an-
Predictor variables	Z. noltii shoot density		
	estimate	t	P
Z. marina shoot density	-2.54	-5.44	5.14x10"7
distance from marina	11.88	1.38	0.17
distance from sand bank	31.14	1.68	0.096
distance from beach area	-29.84	-1.11	0.27
distance from canal	6.28	0.33	0.74
wave exposure	-13817.4	-1.68	0.096
salinity 1	-101.45	-2.34	0.022
salinity 2	-166.78	-1.54	0.13
bottom slope	-20029.9	-0.71	0.48
sediment organic matter %	-161.31	-2.15	0.034
sediment water content %	3465.96	1.54	0.13
thropogenic use, and experience boat traffic and residential drainage carrying organic particles and sediment, which combined result in lowered light attenuation that could then be the more direct cause of Z. marina being able to invade or even replace Z. noltii. Future research should address the mechanisms allowing coexistence of the two species on the scale of centimetres across the wide range of physical conditions observed in this study.
ACKNOWLEDGEMENTS
We thank the University of Zadar, Department of Maritime Science, and the Ernst-Moritz-Arndt Universität
Greifswald, Biologische Station Hiddensee, for their support. This project was financially supported by a research grant from the Croatian Ministry of Science for Project 269-0362975-3174 to S.S., Monitoring and ecology of benthic communities of the Croatian Adriatic. We also thank Tatjana Bakran-Petricioli and Zeljka Vi-dakovic-Cifrek, University of Zagreb, for inviting us to use equipment in their laboratories. Furthermore we thank the Meteorological and Hydrological Service of the Republic of Croatia for providing wind velocity and frequency data, and Masa Suric, University of Zadar, Department of Geography, for helping us applying these data.
DOKAZI O TEKMOVALNOSTI MED MORSKIMI TRAVAMI V LAGUNI SREDNJEGA
JADRANA (HRVAŠKA)
Claudia KRUSCHEL, Stewart T. SCHULTZ & Melita BACALjA University of Zadar, Department of Maritime Sciences, HR-23000 Zadar, M. Pavlinovica bb, Croatia E-mail: claudia@claudiakruschel.com
Sven DAHLKE
University of Greifswald, Biologische Station Hiddensee, D-18565 Kloster / Hiddensee, Biologenweg 15, Germany
POVZETEK
Avtorji članka so na izobatnem transektu v Novigrajskem morju (Hrvaška, srednji jadran) našli dokaze o tekmovalnosti med sobivajočima morskima travama Zostera noltii in Z. marina. Transekt je vključeval razvita (marina) in nerazvita območja obrežnega pasu. Pokritost s posameznima vrstama morske trave je bila ugotovljena s podvodno videografijo DGPS, medtem ko je bila gostota poganjkov ocenjena na osnovi primerkov, ki so jih nabrali potapljači. Izkazalo se je, da se vrsti na okolje odzivata povsem drugače: medtem ko sta se pokrovnost in gostota šopov vrste Z. marina zmanjševali z razdaljo od marine, sta se pokrovnost in gostota šopov vrste Z noltii povečevali. Prostorsko razširjenost vrst je bilo mogoče v precej večji meri pojasniti z vzajemnim delovanjem in vplivanjem dveh vrst kot pa z okoljski parametri. Sklep avtorjev je bil, da je tekmovalnost med dvema vrstama poglavitni proces, ki določa razširjenost obeh vrst iz rodu Zostera v Novigrajskem morju.
Ključne besede: Zostera, tekmovalnost med morskimi travami, okoljski parametri, DGPS-videografija, Jadran
REFERENCES
Cabalo, S., R. Santos & C. M. Duarte (2008): The impact of sediment burial and erosion on seagrass: A review. Est. Coast. Shelf Sci., 79(3), 354-366. Caniglia, G., S. Borella & B. Carpene (1992): Distribution of Sea-grasses Zostera marina L., Zostera noltii Hor-nem., Cymodocea nodosa Ucrea Asch. in Venice lagoon. Lav. Soc. Veneciana Sci. Nat., 17, 137-150.
Charpentier, A., P. Grillas, F. Lescuyer, E. Coulet & I. Auby (2005): Spatio-temporal dynamics of a Zostera noltii dominated community over a period of fluctuating salinity in a shallow lagoon, southern France. Est. Coast. Shelf Sci., 64, 307-315.
Dauwalter, D. C., W. L. Fisher & K. C. Belt (2006):
Mapping Stream Habitats with a Global Positioning System: Accuracy, Precision, and Comparison with Traditional Methods. Environ. Manage., 37 (2), 271-280.
de Boer, W. F. (2007): Seagrass - sediment interactions, positive feedbacks and critical thresholds for occurrence: a review. Hydrobiologia, 591, 5-24. den Hartog, C. (1970): The Seagrasses of the World. North-Holland, Amsterdam, 275 p. Duarte, C. M. & J. Kalff (1988): Influence of lake morphology on the response of submerged macrophytes to sediment fertilization. Can. J. Fish. Aquat. Sci., 45, 216221.
Fourqurean, J. W., G. V. N. Powell, W. J. Keworthy & C. Zieman (1995): The Effect of Long-Term Manipulations of Nutrient Supply on Competition between the Seagrasses Thallassia testudium and Halodule wrightii in Florida Bay. Oikos, 72(3), 349-358. Google Earth 5.0 (2009) http://earth.google.com Guidetti, P. (2000): Temporal dynamics of Zostera marina L. off the Lagoon of Grado (Northern Adriatic Sea, Italy). Bot. Mar., 43(6), 541-546.
Hootsmans, M. J. M., J. E. Vermaat & W. van Vierssen (1987): Seed-bank development, germination and early seedling survival of two seagrass species from the Netherlands: Zostera marina L. and Zostera noltii Hornem. Aquat. Bot., 28, 275-285.
Koch, E. W. (2001): Beyond light: physical, geological and geochemical parameters as possible submersed aquatic vegetation habitat requirements. Estuaries, 24, 1-17.
Krause-Jensen, D., M. F. Pedersen & K. Jensen (2003):
Regulation of eelgrass (Zostera marina) cover along depth gradients in Danish coastal waters. Estuaries, 26, 866-877.
Laugier, T., V. Rigollet & M. L. de Casabianca (1999):
Seasonal dynamics in mixed eelgrass beds, Zostera marina L. and Z. noltii Hornem., in a Mediterranean coastal lagoon (Thau lagoon, France). Aquat. Bot., 63, 51-69. Lee, K., S. R. Park & Y. K. Kim (2007): Effects of irradi-ance, temperature, and nutrients on growth dynamics of seagrass: A review. J. Exp. Mar. Biol. Ecol., 350, 144175.
Leoni, V., A. Vea, V. Pasquaini, C. Pergent-Martini & G. Pergent (2008): Effects of experimental reduction of light
and nutrient enrichment (N and P) on seagrass: a review. Aquat. Conserv: Mar. Freshw. Ecosyst., 18, 202-220. Marba, N., J. Cebrian, S. Enriquez & C. M. Duarte (1996): Growth patterns of Western Mediterranean sea-grasses: species-specific responses to seasonal forcing. Mar. Ecol. Prog. Ser., 133, 203-215. Norris, J. G., S. Wyllie-Echevierra, T. Mumford, A. Bailey & T. Turner (1997): Estimating basal area coverage of subtidal seagrass beds using underwater videography. Aquat. Bot., 58, 269-287.
Schultz, S. T. (2008): Seagrass monitoring by underwater videography: Disturbance regimes, sampling design, and statistical power. Aquat. Bot., 33, 228-238. Schultz, S. T., C. Kruschel & T. Bakran-Petricioli (2009):
Influence of seagrass meadows on predator-prey habitat segregation in an Adriatic lagoon. Mar. Ecol. Prog. Ser., 374, 85-99.
Sinovcic, G., M. Franicevic & V. Cikes Kec (2004): Unusual occurrence and some aspects of biology of juvenile gilt sardine (Sardinella aurita Velenciennes, 1847) in the Zrmanja River estuary (eastern Adriatic). Appl. Ichthyol., 20, 53-57.
Tanaka, Y. & H. Kayanne (2007): Relationship of species composition of tropical seagrass meadows to multiple physical environmental factors. Ecol. Res., 22, 87-96. Touchette, B. W. (2007): Seagrass - salinity interactions: Physiological mechanisms used by submersed marina angiosperms for a life at sea. J. Exp. Mar. Biol. Ecol., 350, 194-215.
Touchette, B. W. & J. M. Burkholder (2000): Review of nitrogen and phosphorus metabolism in seagrasses. J. Exp. Mar. Biol. Ecol., 250, 133-167. Vermaat, J. E., C. A. Verhagen & D. Lindenburg (2000):
Contrasting responses in two population of Zostera noltii Hornem. to experimental photoperiod manipulation at two salinities. Aquat. Bot., 76, 179-189. Widdows, J., N. D. Pope, M. D. Brinsley, H. Asmus & R. M. Asmus (2008): Direct effects of seagrass beds (Zostera noltii and Z. marina) on near-bed hydrodynamics and sediment resuspension. Mar. Ecol. Prog. Ser., 358, 125-136.