Temporal Dynamics and Biomass Partitioning in Three Adriatic Seagrass Species: Posidonia oceanica, Cymodocea nodosa, Zostera marina

P.S.Z.N.: Marine Ecology, 23 (1): 51±67 (2002) ã 2002 Blackwell Verlag, Berlin ISSN 0173-9565

Accepted: August 8, 2001

Temporal Dynamics and Biomass Partitioning in Three Adriatic Seagrass Species: Posidonia oceanica, Cymodocea nodosa, Zostera marina Paolo Guidetti1, 2, Maurizio Lorenti1, Maria Cristina Buia1, * & Lucia Mazzella ² 1 2

Laboratorio di Ecologia del Benthos, Stazione Zoologica ªA. Dohrnº di Napoli, Punta S. Pietro, 80077 Ischia-Porto, Naples, Italy. Present address: Laboratorio di Zoologia e Biologia Marina, Dipartimento di Scienze e Tecnologie Biologiche ed Ambientali, UniversitaÁ di Lecce, Via Provinciale Monteroni, 73100 Lecce, Italy.

With 11 figures

Keywords: Seagrasses, temporal dynamics, biomass partitioning, Adriatic Sea, Mediterranean Sea. Abstract. The temporal dynamics of three seagrasses, Posidonia oceanica, Cymodocea nodosa and Zostera marina, was studied in different areas of the Adriatic Sea by analysing phenological parameters and biomass trends in different compartments of seagrass systems. For this purpose, samplings were conducted in 1997 once per season at each station, Otranto (southern Adriatic Sea) and Grado (northern Adriatic Sea). Structural parameters and biomass of plant compartments differed among seagrasses both in absolute values and in seasonal variability. P. oceanica was the largest plant, showing the highest number of leaves per shoot, highest leaf surface, Leaf Area Index and shoot weight. Z. marina was intermediate in size and had the longest leaves, whereas C. nodosa was the smallest seagrass. P. oceanica accounted for the highest total biomass (mean ± SE: 1895.9 ± 180.2 g DW ´ m±2; CV = coefficient of variation: 19.0 %), considerably more than C. nodosa (mean ± SE: 410.4 ± 88.4 g DW´m±2; CV: 43.1 %) and Z. marina (mean ± SE: 312.1 ± 75.1 g DW ´ m±2; CV: 48.1 %), although the two latter species displayed a higher seasonal variability. Similarly, other features, such as shoot density, leaf surface, LAI, shoot weight and relative contributions of above- and below-ground compartments, were less variable across seasons in P. oceanica than in the two other seagrasses, while leaf length showed the highest seasonal fluctuation in P. oceanica. As for biomass partitioning, C. nodosa showed a higher proportion of the below-ground relative to above-ground biomass (up to 90 %), with a distinct seasonality, whereas in P. oceanica the proportion of below-ground biomass (around 80 %) was fairly constant during the year. We infer that in P. oceanica the seasonal forcing is probably buffered * Author to whom correspondence should be addressed. E-mail: [email protected] U. S. Copyright Clearance Center Code Statement: 0173-9565/02/2301 ± 0051$15.00/0

52

Guidetti, Lorenti, Buia & Mazzella

by the availability of internal resources stored permanently during the year in the below-ground. In C. nodosa and Z. marina, instead, growth processes seem to be amplified by a greater influence of environmental factors.

Problem The important multifunctional role that seagrasses exert in contributing to the productivity of the coastal zone and in providing habitats and resources to rich invertebrate and fish communities is well known for several seas (Ott, 1980; Young, 1981; Bell & Pollard, 1989; Larkum et al., 1989; Mazzella et al., 1992; Edgar & Shaw, 1995; Francour, 1997). In the Mediterranean basin five species of seagrasses are recorded. Posidonia oceanica (L.) Delile and Cymodocea nodosa (Ucria) Ascherson are the most important and widespread species, the latter mainly present in sheltered sites. Zostera noltii Hornemann mainly colonises protected bays and the mid-littoral zone in coastal lagoons, and Zostera marina L. is restricted to shallow stands in brackish lagoons and sheltered environments (den Hartog, 1970; Mazzella et al., 1993; Buia & Marzocchi, 1995; MarbaÁ et al., 1996). Halophila stipulacea (Forssk.) Ascherson is a Lessepsian migrator introduced in the eastern Mediterranean via the Suez Canal and recently recorded also in the western basin (Acunto et al., 1995). Interestingly, all five species mentioned above are present in the Adriatic Sea (Mazzella et al., 1998 and references therein). P. oceanica is mainly present along the open coasts of the southern and north-eastern Adriatic (Gamulin-Brida, 1967; Damiani et al., 1988); C. nodosa and Z. noltii are more evenly distributed at shallow stands in marine environments (Gamulin-Brida, 1967) and even more in the coastal lagoons in the northern Adriatic Sea (Caressa et al., 1995; Curiel et al., 1997; Rismondo et al., 1997a, 1997b). Z. marina, as well, is mainly confined to the lagoons in the northern sector of the basin (Curiel et al., 1997; Mazzella et al., 1998) and in front of the Conero promontory (Ancona; A. Solazzi, pers. comm.), while Halophila stipulacea has been recorded in recent years along the Albanian coasts (Kashta, 1992). In the framework of a wider programme (PRISMA II) aimed at identifying issues for the management and preservation of marine natural systems, research was undertaken that focused on studying the functioning of seagrass ecosystems in the Adriatic basin. In this respect, P. oceanica along open coasts, and C. nodosa and Z. marina in lagoons, are species which play a key role, in other regions of the Mediterranean Sea as well; all three can be seriously endangered. In particular, Z. marina, which is very rare in the Mediterranean, forms extended meadows in the northern Adriatic (Curiel et al., 1997; Mazzella et al., 1998). The distribution patterns of different seagrass species are related to their morphofunctional characteristics. The three above-mentioned seagrasses exhibit different plant size, architecture and reproductive strategies, which are translated into species-specific responses to environmental forcing (Buia & Mazzella, 1991; MarbaÁ et al., 1996; Guidetti et al., 2000, 2001). Although several studies have been performed on these species, little attention has been paid to the below-ground compartment and to assessing its functional role in the seagrass systems studied.

Temporal dynamics of Adriatic seagrasses

53

The paper is aimed at examining the temporal dynamics of P. oceanica, C. nodosa and Z. marina in selected areas along the Adriatic coasts where seagrasses experience optimal conditions. By analysing phenological features and biomass partitioning of different plant compartments, we sought to characterise the different strategies exhibited by the three species in maintaining their structure and in allocating biomass.

Material and Methods 1. Study sites Seagrass was sampled in February, April, August and November 1997 by SCUBA diving in two different areas located in the southern and northern Adriatic Seas, namely Otranto (Apulia: 40°10' N, 18°29' E) for P. oceanica and Grado (Friuli: 45°41' N, 13°28' E) for C. nodosa and Z. marina (Fig. 1). At Otranto, samplings were performed north of the local harbour, at 6.5 m depth where P. oceanica grows on organogenic sand and rock. The meadow extends from about 5 to 25 m depth on a gentle slope and shows a patchy distribution. Ripple marks provide evidence of strong water movements induced by wave action. C. nodosa samples were collected at about 1.5 m depth off the sea entrance of the Grado lagoon, where the seagrass grows on a sandy bottom. Z. marina, instead, settles on a pelitic sand substrate at about 0.5 ± 1 m depth near the sea entrance of the lagoon of Grado. Water temperature at the three sites was recorded during each sampling by means of a reversing thermometer. At Otranto, values ranged from 11.8 °C (winter) to 24.8 °C (summer). Winter minima were recorded also for C. nodosa (8.9 °C) and Z. marina (8.7 °C) sites, whereas maximum temperatures occurred in summer (24.3 and 24.3 °C, respectively). Salinity was around 37±38 at Otranto in all four seasons, while at Grado values ranged from 28 to 34 (C. nodosa) and from 25 to 33 (Z. marina), with minima recorded in summer. The attenuation coefficient of PAR averaged 0.15 (± 0.02 SD) at Otranto and 0.53 (± 0.35 SD) at Grado.

Fig. 1.

Location of the sampling areas.

54

Guidetti, Lorenti, Buia & Mazzella

2. Sampling methods and sample analysis Shoot density was measured in situ within 40 ´ 40 cm quadrats for P. oceanica (for a total of 20 replicates) and Z. marina (10 replicates per season), and within 20 ´ 20 cm quadrats for C. nodosa (10 replicates per season). In order to compare the morphological data of the three species, 20 shoots per season of each seagrass were used for the phenological analysis, which was carried out according to the standardised procedures utilised for P. oceanica (Giraud, 1977). The following parameters were measured: total number of leaves per shoot; number, length and width per leaf age class (juvenile, intermediate and adult leaves). Leaf Area Index (LAI: m2 ´ m±2) was determined by multiplying the mean leaf surface per shoot (cm2 ´ l ±1) by the meadow shoot density. Biomass was sampled by removing clods within quadrats 30 ´ 30 cm for P. oceanica (3 replicates per season) and by means of a corer of 23.6 cm in diameter (5 replicates per season) for C. nodosa and Z. marina. The plant material was rinsed after sieving to remove sediment. The above-ground portion was subdivided into leaf blades, leaf bases and brown tissue; it was separated from the below-ground portion which comprises dead sheaths (old leaf bases persisting on rhizomes, a feature shown only in P. oceanica), living and dead rhizomes as well as living and dead roots. The heterogeneous plant detritus collected within clods and comprising both allochthonous and autochthonous material, even of terrigenous origin, was also considered. Subsequently, in order to determine the dry weight (g DW), the various compartments were dried at 60 °C to constant weight.

3. Data analysis The coefficient of variation (CV) was used to represent the extent of the variability among seasons of the different parameters. Means were compared by using the t-test or one-way ANOVA and Tukey's test for multiple comparisons a posteriori. In order to comply with the assumptions of parametric procedures, data were tested for normality by the Kolmogorov-Smirnov test, for homogeneity of variances by Cochran's C test; they were opportunely transformed whenever necessary (Underwood, 1997).

Results 1. Meadow density and plant phenology In P. oceanica, the mean shoot (sh) density was 707.9 ± 65.2 sh ´ m±2. No statistical differences were observed between the shoot density in spring and autumn (t-test, P = 0.07). In contrast, in C. nodosa the shoot density varied from 977.5 ± 158.3 sh ´ m±2 (winter) to 1657.9 ± 85.3 sh ´ m±2 (summer) (ANOVA, P < 0.001) and in Z. marina from 277.8 ± 44.6 sh ´ m±2 (winter) to 773.7 ± 90.5 sh ´ m±2 (summer) (ANOVA, P < 0.001). The coefficient of variation was highest in Z. marina (39.2 %), followed by C. nodosa (21.7 %) and P. oceanica (9.2 %). The mean number of leaves per shoot was highest in all seasons in P. oceanica, followed by Z. marina and C. nodosa (Fig. 2). In P. oceanica and Z. marina, this parameter did not vary significantly with season (ANOVA, P > 0.05), while in C. nodosa the spring value was significantly higher than that measured in summer (ANOVA, P = 0.015; Tukey's test, P < 0.05). Coefficients of variation, on the whole, were low. However, in comparing the three seagrasses, CV was highest for C. nodosa and lowest for Z. marina (Fig. 2).

Temporal dynamics of Adriatic seagrasses

55

Fig. 2. Seasonal pattern in the mean number of leaves per shoot. Bars indicate standard errors. The magnitude of seasonal fluctuations in the leaf number of the three seagrasses is reported in parentheses (CV: coefficient of variation).

Fig. 3. Seasonal pattern in the number of different age classes of leaves in the three investigated seagrasses. Bars indicate standard errors.

56

Guidetti, Lorenti, Buia & Mazzella

The mean number of leaves per shoot belonging to different age classes differed significantly among seasons (ANOVA, P < 0.05) in all the three species, showing different temporal patterns reflecting the species-specific leaf growth. In P. oceanica, the number of juvenile and adult leaves increased in summer and autumn, whereas intermediate leaves decreased in summer (Fig. 3a). In C. nodosa, the number of juvenile leaves was highest in the autumn-winter period, unlike the intermediate leaves, while adult leaves were significantly more abundant in spring and summer (Fig. 3b). In Z. marina, juve-

Fig. 4. Seasonal pattern of leaf length of the different leaf age classes in the three investigated seagrasses. Bars indicate standard errors.

Temporal dynamics of Adriatic seagrasses

Fig. 5.

57

Seasonal pattern of leaf surface per shoot in the three species studied. Bars indicate standard errors.

nile and intermediate leaves decreased in spring and spring-summer, respectively, while adult leaves showed an opposite trend (Fig. 3c). In all three species adult leaves usually accounted for the highest number, with the exception of P. oceanica in winter and C. nodosa in winter and autumn. The leaf lengths of different age classes in P. oceanica provided evidence of marked seasonal dynamics (Fig. 4a). Thus, analysis of variance showed highly significant differences in the mean leaf length for each age class (P < 0.001). Juvenile leaves were shortest in winter, while intermediate and adult leaves were longest in spring and summer, respectively (Tukey's test, P < 0.05). In C. nodosa, which bore the shortest leaves, significant differences in mean length among seasons were observed in juvenile (P = 0.005) and adult leaves (P < 0.001), differently from the intermediate ones (P = 0.127). Juvenile leaves were longest in spring, while adult leaves were longest in summer (Tukey's test, P < 0.05; Fig. 4b). In Z. marina, juvenile leaf length differed significantly over the year (P = 0.034), as did the other two age classes (P < 0.001). Intermediate and adult leaves were longest in spring-summer (Tukey's test, P < 0.05; Fig. 4c). On the whole, P. oceanica showed the highest seasonal variability in the leaf length whatever the age class considered (Fig. 4d). The shoot surface area was highest in P. oceanica, followed by Z. marina and C. nodosa, with a striking difference among species (Fig. 5). This parameter was significantly different among seasons for all three species (ANOVA, P < 0.001). In P. oceanica the leaf surface area was significantly highest in autumn, in C. nodosa in summer and in Z. marina during the spring-summer period (Tukey's test, P < 0.05; Fig. 5). The highest seasonal variability was observed in C. nodosa (CV = 56.6 %) versus the other two species (CV = 30.1 % and 31.7 % for P. oceanica and Z. marina, respectively). The green surface area in P. oceanica ranged from 56.7 cm2 ´ sh±1 in winter to 146.9 cm2 ´ sh±1 in spring, in C. nodosa from 4.4 cm2 ´ sh±1 in winter to 21.6 cm2 ´ sh±1 in summer and in Z. marina from 29.1 cm2 ´ sh±1 in autumn to about 70 cm2 ´ sh±1 during spring-summer. Brown tissue in P. oceanica was mainly present in summer

58

Fig. 6.

Guidetti, Lorenti, Buia & Mazzella

Seasonal pattern of Leaf Area Index in the three species studied.

(accounting for 3.4 % of the total leaf surface) and autumn (8.8 %), in C. nodosa in winter (29.1 %), whereas it was negligible whatever the season in Z. marina. In all seasons the leaf area index was highest in P. oceanica, followed by Z. marina and C. nodosa. In P. oceanica this parameter was lowest in winter and similar during the other seasons. In the other two seagrasses, however, the highest values were recorded during summer (Fig. 6).

2. Biomass partitioning Posidonia oceanica bore the largest shoot, followed by Z. marina and C. nodosa (Fig. 7). For all three species, the mean shoot weight varied significantly among seasons (ANOVA, P < 0.01). Seasonal fluctuation were highest in Z. marina, followed by C. nodosa and P. oceanica (Fig. 7). P. oceanica had the highest total biomass (mean ± SE: 1895.9 ± 180.2 g DW ´ m±2; CV: 19.0 %), with a striking difference in comparison to C. nodosa (mean ± SE: 410.4 ± 88.4 g DW ´ m±2; CV: 43.1 %) and Z. marina (mean ± SE: 312.1 ± 75.1 g DW ´ m±2; CV: 48.1 %). Nonetheless, the CV values of the latter two species were characterised by a higher seasonal variability than P. oceanica. The relative contribution of the above- and below-ground compartments during the different seasons varied very slightly in P. oceanica. The below-ground, for example, accounted for a fraction ranging from 78 % in winter to 83 % in autumn (Fig. 8a). In C. nodosa, the seasonal variability was higher than in P. oceanica, the below-ground accounting from 81 % in summer to about 90 % in the remaining seasons (Fig. 8b). The highest seasonal variability was observed, instead, in Z. marina: below-ground biomass varied from 60 % in summer to 93 % in autumn. Note that in the latter species, the high-

Temporal dynamics of Adriatic seagrasses

59

Fig. 7. Seasonal pattern of the shoot weight in the three seagrasses. The magnitude of seasonal fluctuations in the shoot weight is reported in parentheses (CV: coefficient of variation).

est relative contribution of the above-ground was observed, although during summer only (Fig. 8a, b, c). In P. oceanica, the above-ground compartment, represented by leaf blades and leaf bases, remained relatively constant whatever the season. Brown tissue contributed slightly in summer and autumn (Fig. 9a). As far as the below-ground is concerned, old leaf sheaths accounted for more than 50 % of the biomass of this compartment, followed by living rhizomes. Living and dead roots, and dead rhizomes, were less represented in all seasons (Fig. 9b). The detritus, mainly composed by fragments of leaves and sheaths of varying size, prevailed in spring and was absent in summer and autumn (Fig. 9c). Significant differences among seasons were observed in leaf blades (ANOVA, P = 0.04), leaf bases (P < 0.001) and brown tissue (P < 0.001), in contrast to all the below-ground compartments (living rhizomes, P = 0.11; dead rhizomes, P = 0.51; living roots, P = 0.5; dead roots, P = 0.14) except for detritus (P = 0.02). Detritus, brown tissue, leaf bases and dead roots showed the highest annual fluctuation, whereas the other compartments were less variable over the year (Fig. 9d). In C. nodosa, as far as the above-ground (leaf blades and bases) compartment is concerned, note the high leaf standing crop in summer versus the other seasons (Fig. 10a). Leaf blades and leaf bases, moreover, showed a significantly higher biomass in summer than during the other seasons (ANOVA, P < 0.001; Tukey's test, P < 0.05) and comparatively more brown tissues were observed in summer and autumn (ANOVA, P < 0.001; Tukey's test, P < 0.05). Living rhizomes and living roots accounted for a high fraction of the below-ground in all seasons, while dead rhizomes and roots were slightly better represented in summer-autumn (Fig. 10b). In this plant, all below-ground compartments differed significantly with season. Living rhizomes and roots were better represented in summer (ANOVA, P < 0.001; Tukey's test, P < 0.05 for both) and dead

60

Fig. 8.

Guidetti, Lorenti, Buia & Mazzella

Percent contribution of the above- and below-ground compartments in the three species studied.

rhizomes and roots in summer and autumn (ANOVA, P = 0.01 and P = 0.02, respectively; Tukey's test, P < 0.05). Detritus ± mainly leaf fragments but also terrigenous plant material ± was better represented in autumn and spring (Fig. 10c; ANOVA, P < 0.001; Tukey's test, P < 0.05). In this plant, the coefficient of variation showed the greatest seasonal fluctuation for all compartments (Fig. 10d). The biomass of leaf blades and dead rhizomes widely changed with seasons, as did the biomass of the remaining compartments, which showed CVs around 50 %. Fig. 9. Seasonal pattern of biomass of different compartments in the Posidonia oceanica system. a: aboveground; b: below-ground; c: detritus; d: magnitude of seasonal biomass fluctuation of the compartments. Fig. 10. Seasonal pattern of biomass of different compartments in the Cymodocea nodosa system. a: aboveground; b: below-ground; c: detritus; d: magnitude of seasonal biomass fluctuation of the compartments.

Temporal dynamics of Adriatic seagrasses

Fig. 9

Fig. 10

61

62

Guidetti, Lorenti, Buia & Mazzella

Fig. 11. Seasonal pattern of biomass of different compartments in the Zostera marina system. a: above-ground; b: below-ground; c: detritus; d: magnitude of seasonal biomass fluctuation of the compartments.

In Z. marina, leaf blades prevailed on leaf bases and both compartments accounted for a significantly higher biomass in spring-summer (Fig. 11a; ANOVA, P < 0.001; Tukey's test, P < 0.05) than during the remaining seasons. The biomass of the brown tissue was higher in summer (Fig. 11a; ANOVA, P = 0.004; Tukey's test, P < 0.05). Dead rhizomes accounted for a large fraction of the below-ground, followed by living rhizomes (Fig. 11b). All below-ground compartments showed highly significant differences related to season. Living rhizomes showed a biomass peak during summer (ANOVA, P < 0.001; Tukey's test, P < 0.05), whereas dead rhizomes were less variable in all seasons but autumn, when the lowest biomass was measured (ANOVA, P < 0.001; Tukey's test, P < 0.05). Living roots were less represented in winter (ANOVA, P < 0.001; Tukey's test, P < 0.05), whereas dead roots prevailed in spring (ANOVA, P = 0.004; Tukey's test, P < 0.05). In this system, the detritus was well represented in all seasons, but was more abundant in spring and autumn (Fig. 11c; ANOVA, P < 0.001; Tukey 's test, P < 0.05). Z. marina showed higher seasonal fluctuations than the other two species (Fig. 11d). Brown tissue biomass reached CV values as high as 100 %. Leaf blades, leaf bases, living rhizomes, living roots and dead roots were highly variable in biomass (CV between 50 and 80 %), but the seasonal fluctuations of dead rhizomes and detritus were also not negligible (CV = 36 and 47 %, respectively).

Temporal dynamics of Adriatic seagrasses

63

Discussion The three seagrasses studied in the Adriatic Sea showed species-specific temporal dynamics. Differences were recorded not only in the absolute values, but also in the amplitude of the within-species seasonal variability of structural parameters and biomass of different plant compartments; this probably involves different strategies of storage capability, as underlined by the trends of seasonal dynamics observed in the present paper and by the patterns of nutrient contents and allocation (Guidetti et al., 2001). Posidonia oceanica was the largest plant, showing the highest values of leaves per shoot, leaf surface, LAI, shoot weight and total biomass, followed by Z. marina and then C. nodosa. On the other hand, the P. oceanica system seemed to be the less variable in time. Thus, important structural features such as shoot density, leaf area index, shoot weight and biomass of the below-ground were less variable among seasons in P. oceanica than in the other two species. As far as P. oceanica is concerned, our results describing the annual trends of several structural parameters (number of leaves per shoot, leaf surface area, leaf area index, shoot weight) are consistent with the data reported by Mazzella & Ott (1984) and Buia et al. (1992). The above authors provided evidence that the leaf number per shoot varied little over the year, while shoot surface, LAI and shoot weight commonly reached their maxima in May-June. Moreover, Buia et al. (1992) observed annual patterns in the number and mean length of leaves of different age classes in P. oceanica at shallow stands, which agree with the trends observed at Otranto. Note the high contribution of sheaths to the total biomass of the system. Sheaths and rhizomes are reported as the main compartments contributing to the net loss of biogenic elements by burial into the sediments (Pergent et al., 1994; Pergent et al., 1997), a phenomenon exclusive of P. oceanica systems. With regard to C. nodosa, the seasonal patterns of the below-ground and aboveground found in the present study broadly agree with data from other Mediterranean locations (Terrados & Ros, 1992; Mazzella et al., 1993; PeÂrez & Romero, 1994; Rismondo et al., 1997a; Sfriso & Ghetti, 1998; Rigollet et al., 1998), although absolute values may vary substantially, probably in relation to different local environmental factors. In all locations studied by the above authors, in any case, shoot density, aboveground biomass and LAI were highest in summer and lowest in winter. By contrast, clear seasonal patterns in the below-ground biomass were not observed by some of these authors (Terrados & Ros, 1992; Sfriso & Ghetti, 1998). The seasonal variation of Z. marina standing biomass per unit area at Grado was due to both fluctuations in shoot density and in the shoot size. In contrast, MarbaÁ et al. (1996) mainly attributed seasonal variations in the above-ground to changes in shoot weight. Our data on seasonal fluctuations in Z. marina are mostly consistent with the patterns observed in the Venice lagoon (Italy) by Rismondo et al. (1995) and Sfriso & Ghetti (1998), and in the Thau lagoon (France) by Laugier et al. (1999), who reported biomass maxima in late spring and summer. The above authors, however, carried out monthly samplings, allowing the identification of biomass peaks with a better time resolution than in our investigation (four months). Outside the Mediterranean Sea and, more precisely, in north-European waters, a number of authors (Sand-Jensen, 1975; Pedersen & Borum, 1993; Olesen & Sand-Jensen, 1994; van Lent & Verschuure, 1994) observed seasonal fluctuations in several structural parameters of Z. marina meadows

64

Guidetti, Lorenti, Buia & Mazzella

and trends of both above- and below-ground compartments consistent with our observations from Grado. On the whole, differences among species in the pattern of certain phenological parameters (i. e. number and leaf length of different age classes) probably account for differences in the species-specific leaf growth dynamics. In C. nodosa and Z. marina, leaves were longest in summer, as opposed to P. oceanica, which shows maximum growth in spring and autumn (Mazzella & Ott, 1984; Buia et al., 1992). In addition, oldest leaves of C. nodosa and Z. marina are the most numerous and longest in all seasons, differently from P. oceanica. This latter species, for example, has the most intermediate leaves in winter and the longest leaves of this age class in winter and spring, the usual picture for P. oceanica. This can be explained by the different growth dynamics of P. oceanica versus the other two species. Morphology (which mainly includes leaf length and width) accounts for differences between species in terms of leaf surface per shoot and per unit area (LAI). Our data suggest that seasonal changes may affect the growth cycle of small-sized seagrass species; large plants with long-lived and large-sized rhizomes (Duarte, 1991), like P. oceanica, are able to store and reallocate resources, allowing the plant to support growth patterns comparatively independent of environmental conditions (MarbaÁ et al., 1996). Based on elemental analysis conducted on the same seagrass populations from Otranto and Grado considered in this work, Guidetti et al. (2001) found the highest relative nitrogen content in the rhizome tissue of P. oceanica in all seasons; they also provided evidence of seasonal variations attributable to resource allocation. By contrast, in Cymodocea and Zostera the nitrogen content in the above- (mainly leaves) and below-ground (mainly living rhizomes) was more balanced or higher in the leaves. These patterns are consistent with those observed by other authors working in and outside the Mediterranean Sea (Pirc & Wollenweber, 1988; Romero et al., 1992; Pedersen & Borum, 1993). Nevertheless, the relative contribution of the two compartments in terms of total biomass is similar in all three species. This suggests that the below-ground compartments exert an important structural role such as mechanically anchoring the plants to the substrate, which accounts for the stability and persistence of seagrasses. In all the three seagrasses studied, the more or less strong influence of seasonal forcing on the growth dynamics of rhizomes, which occurs by apposition of reiterative modules (internodes), has important implications for the biomass development of different plant compartments. Fluctuations in rhizome growth, moreover, are closely related to seasonal fluctuations in shoot density; this directly influences seasonal changes in the meadow physiognomy and structural architecture. Thus, C. nodosa and Z. marina showed higher seasonal variations in shoot density than P. oceanica, in which this parameter is commonly considered to be stable on an annual scale (Mazzella & Buia, 1989). Another aspect which differentiates the life strategies and growth dynamics of the three seagrasses is the living/dead biomass ratio of the below-ground, which is at the basis of the biomass turnover of this compartment. Thus, in Z. marina, dead plant biomass tended to prevail in winter-spring at Grado, as opposed to the data reported by Laugier et al. (1999), who measured slightly higher values in spring-summer in the Thau lagoon. However, these authors studied mixed stands of Z. marina and Z. noltii and did not separate dead parts of the two seagrasses. Living rhizomes of Z. marina from Grado were comparatively short, with terminal, oldest portions showing evident

Temporal dynamics of Adriatic seagrasses

65

signs of decay. This suggests that rhizomes undergo a continuous decay process over the year, even during the periods of maximum elongation. Differently from the other two species, dead parts in the below-ground of C. nodosa accounted for a negligible proportion of this compartment's biomass in all seasons. This could involve a higher longevity of buried structures of C. nodosa versus Z. marina, thus indicating a more efficient strategy of the former to survive unfavourable periods. In the P. oceanica bed of Otranto, dead parts, mostly represented by old sheaths still attached to the plant, account for more than 50 % of the below-ground biomass in all seasons; this is similar to other areas (Romero et al., 1992; Mateo & Romero, 1997). The detritus qualitatively differed at the three sites in relation to the environment colonised by the three species. Its amount reflects the general dynamics of the sites (currents, wave action or seasonality of terrigenous input) on a spatial mesoscale. This is supported by the fact that the patterns observed in C. nodosa and Z. marina meadows are substantially identical, although the quality of the detritus differed. In P. oceanica systems, however, the amount and the seasonal dynamics of the detritus are mainly affected by the bed density, depth, and the rate of leaf fall and export; most of these phenomena depend on local hydrodynamic conditions (Romero et al., 1992). In conclusion, our results provided evidence of different seasonal growth patterns of P. oceanica, C. nodosa and Z. marina from the Adriatic Sea. The responses of P. oceanica to seasonality appear to be partially independent of environmental factors in comparison with smaller-sized species. An indication of this is the pronounced seasonality of key resource allocation, whereas growth shows little seasonality (Wittmann, 1984) compared, for instance, with C. nodosa (Terrados & Ros, 1992; Mazzella et al., 1993). Thus, we can assume that the three seagrasses show different life strategies closely linked to their ability to use (allocate and store) internal resources and to buffer or amplify the external seasonal forcing. Consequently, because the patterns found by us are common to extra-Adriatic populations of the same species, an issue which warrants further investigations is the extent to which these species-specific strategies explain large-scale distribution at the basin level.

Summary The analysis of phenological parameters and biomass trends in different compartments of three seagrass species from the Adriatic Sea, namely Posidonia oceanica, Cymodocea nodosa and Zostera marina, are presented in this paper. Seasonal samplings were carried out during 1997 at Otranto (southern Adriatic Sea) and Grado (northern Adriatic Sea). Differences among the three seagrass species consisted in absolute values and in the amplitude of the within-species seasonal variability of both structural parameters and the biomass of plant compartments. Based on our findings, we infer that in P. oceanica the seasonal forcing is probably buffered by the availability of internal resources stored permanently in the below-ground during the year. In C. nodosa and Z. marina, instead, growth processes seem to be amplified by a greater influence of environmental factors.

66

Guidetti, Lorenti, Buia & Mazzella

Acknowledgements This study has been funded by the national research programme PRISMA II, and Dr. P. Guidetti benefited from a C.N.R. (National Research Council) grant in the framework of this programme. Many thanks are due to A. Rismondo, D. Curiel, S. Caressa and P. Scaffidi for the help provided during sampling activities.

References Acunto, S., F. Maltagliati, F. Rindi, F. Rossi & F. Cinelli, 1995: Osservazioni su una prateria di Halophila stipulacea (Forssk.) Aschers. (Hydrocharitaceae) nel Mar Tirreno meridionale. Atti Soc. Toscana Sci. Nat., Mem., Serie B, 102: 19±22. Bell, J. D. & D. A. Pollard, 1989: Ecology of fish assemblages and fisheries associated with seagrasses. In: W. D. Larkum, A. J. Mc Comb & S. A. Shepherd (Eds.), Biology of Seagrasses. Elsevier, Amsterdam: 565±609. Buia, M. C. & M. Marzocchi, 1995: Dinamica dei sistemi a Cymodocea nodosa, Zostera marina e Zostera noltii nel Mediterraneo. G. Bot. Ital., 129: 319±336. Buia, M. C. & L. Mazzella, 1991: Reproductive phenology of the Mediterranean seagrasses Posidonia oceanica (L.) Delile, Cymodocea nodosa (Ucria) Aschers. and Zostera noltii Hornem. Aquat. Bot., 40: 343±362. Buia, M. C., V. Zupo & L. Mazzella, 1992: Primary production and growth dynamics in Posidonia oceanica. P.S.Z.N. I: Marine Ecology, 13(1): 2±16. Caressa, S., C. Ceschia, G. Orel & R. Treleani, 1995: Popolamenti attuali e pregressi nel Golfo di Trieste da Punta Salvore a Punte Tagliamento (Alto Adriatico). Rivista Marittima suppl.: 160±187 (references: 255±271). Curiel, D., A. Rismondo, F. Scarton & M. Marzocchi, 1997: Flowering of Zostera marina in the Lagoon of Venice (North Adriatic, Italy). Bot. Mar., 40: 101±105. Damiani, V., C. N. Bianchi, O. Ferretti, D. Bedulli & C. Morri, 1988: Risultati di una ricerca ecologica sul sistema marino costiero pugliese. Thalassia Salentina, 18: 153±169. Duarte, C. M., 1991: Allometric scaling of seagrass form and productivity. Mar. Ecol. Prog. Ser., 77: 289±300. Edgar, G. J. & C. Shaw, 1995: The production and trophic ecology of shallow-water fish assemblages in southern Australia. I. Species richness, size-structure and production of fishes in Western Port, Victoria. J. Exp. Mar. Biol. Ecol., 194: 53±81. Francour, P., 1997: Fish assemblages of Posidonia oceanica beds at Port-Cros (France, NW Mediterranean): assessment of composition and long-term fluctuations by visual census. P.S.Z.N.: Marine Ecology, 18(2): 157±173. Gamulin-Brida, H., 1967: The benthic fauna of the Adriatic Sea. Oceanogr. Mar. Biol. Annu. Rev., 5: 535±568. Giraud, G., 1977: Contribution aÁ la description et aÁ la phenologie quantitative des herbiers de Posidonia oceanica (feuilles et eÂpiphytes) dans la baie de Port-Cros. PhD thesis, Universite de Aix-Marseille III; 150 pp. Guidetti, P., M. C. Buia, M. Lorenti, M. B. Scipione, V. Zupo & L. Mazzella, 2001: Seasonal trends in Adriatic seagrass communities of Posidonia oceanica (L.) Delile, Cymodocea nodosa (Ucria) Ascherson, Zostera marina L.: plant phenology, biomass partitioning, elemental composition and faunal features. In: F. M. Faranda, L. Guglielmo & G. Spezie (Eds.), Mediterranean Ecosystems: Structures and Processes. Springer-Verlag Italia, 37: 289±295. Guidetti, P., M. C. Buia & L. Mazzella, 2000: The use of lepidochronology as a tool of analysis of dynamic features in the seagrass Posidonia oceanica of the Adriatic Sea. Bot. Mar., 43: 1±9. Hartog, C. den, 1970: The seagrasses of the world. North Holland Publication, Amsterdam; 275 pp. Kashta, L., 1992: Data on the Albanian marine flora. Rapp. Comm. Int. Mer MeÂdit., 33: 349. Larkum, A. W. D., A. J. Mc Comb & S. A. Shepherd, 1989: The Biology of Seagrasses: ATreatise on the Biology of Seagrasses with Special Reference to the Australian Region. Elsevier, Amsterdam; 841 pp. 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. Lent, F. van & J. M. Verschuure, 1994: Intraspecific variability of Zostera marina L. (eelgrass) in the estuaries and lagoons of the southwestern Netherlands. I. Population dynamics. Aquat. Bot., 48: 31±58. MarbaÁ, N., J. Cebrian, S. Enriquez & C. M. Duarte, 1996: Growth patterns of western Mediterranean seagrasses: species-specific responses to seasonal forcing. Mar. Ecol. Prog. Ser., 133: 203±215.

Temporal dynamics of Adriatic seagrasses

67

Mateo, M. A. & J. Romero, 1997: Detritus dynamics in the seagrass Posidonia oceanica: elements for an ecosystem carbon and nutrient budget. Mar. Ecol. Prog. Ser., 151: 43±53. Mazzella, L. & M. C. Buia, 1989: Variazioni a lungo termine di alcuni parametri strutturali di una prateria a Posidonia oceanica. Nova Thalassia, 10(1): 533±542. Mazzella, L., M. C. Buia, M. C. Gambi, M. Lorenti, G. F. Russo, M. B. Scipione & V. Zupo, 1992: Plant-animal trophic relationship in the Posidonia oceanica ecosystem of the Mediterranean Sea: a review. In: J. D. M. Hawkins S. J. & J. H. Price (Eds.), Plant-animal Interaction in the Marine Benthos. Systematic Association, spec. vol., 46: 165±187. Mazzella, L., P. Guidetti, M. Lorenti, M. C. Buia, V. Zupo, M. B. Scipione, A. Rismondo & D. Curiel, 1998: Biomass partitioning in Adriatic seagrass ecosystems (Posidonia oceanica, Cymodocea nodosa and Zostera marina). Rapp. Comm. Int. Mer MeÂdit., 35: 562±563. Mazzella, L. & J. A. Ott, 1984: Seasonal changes in some features of Posidonia oceanica (L.) Delile leaves and epiphytes at different depths. In: Ch.-F. Boudouresque, A. Jeudy De Grissac & J. Olivier (Eds.), International Workshop on Posidonia oceanica beds. GIS Posidonie Publ., Marseille, France, 1: 119±127. Mazzella, L., M. B. Scipione, M. C. Gambi, M. C. Buia, M. Lorenti, V. Zupo & G. Cancemi, 1993: The Mediterranean seagrasses Posidonia oceanica and Cymodocea nodosa. A comparative overview. In: E. Ozhan (Ed.), The first International Conference on Mediterranean Coastal Environment. MEDCOAST 93, Antalya, Turkey, 2±5 November 1993: 103±116. Olesen, B. & K. Sand-Jensen, 1994: Demography of shallow eelgrass (Zostera marina) populations ± shoot dynamics and biomass development. J. Ecol., 82: 379±390. Ott, J. A., 1980: Growth and production in Posidonia oceanica (L.) Delile. P.S.Z.N. I: Marine Ecology, 1: 47±64. Pedersen, M. F. & J. Borum, 1993: An annual nitrogen budget for a seagrass Zostera marina population. Mar. Ecol. Prog. Ser., 101: 169±177. PeÂrez, M. & J. Romero, 1994: Growth dynamics, production, and nutrient status of the seagrass Cymodocea nodosa in a Mediterranean semi-estuarine environment. P.S.Z.N. I: Marine Ecology, 15(1): 51±64. Pergent, G., V. Rico-Raimondino & C. Pergent-Martini, 1997: Fate of primary production in Posidonia oceanica meadows of the Mediterranean. Aquat. Bot., 59: 307±321. Pergent, G., J. Romero, C. Pergent-Martini, M. A. Mateo & C. F. Boudouresque, 1994: Primary production, stocks and fluxes in the Mediterranean seagrass Posidonia oceanica. Mar. Ecol. Prog. Ser., 106: 139±146. Pirc, H. & B. Wollenweber, 1988: Seasonal changes in nitrogen, free amino acids and C/N ratio in Mediterranean seagrasses. P.S.Z.N. I: Marine Ecology, 9(2): 167±179. Rigollet, V., T. Laugier, M. L. De Casablanca, A. Sfriso & A. Marcomini, 1998: Seasonal biomass and nutrient dynamics of Zostera marina L. in two Mediterranean lagoons: Thau (France) and Venice (Italy). Bot. Mar., 41: 167±179. Rismondo, A., D. Curiel, M. Marzocchi & C. Micheli, 1995: Autoecology and production of Zostera marina in Venice lagoon. Rapp. Comm. Int. Mer MeÂdit., 34: 42. Rismondo, A., D. Curiel, M. Marzocchi & M. Scattolin, 1997a: Seasonal pattern of Cymodocea nodosa biomass and production in the lagoon of Venice. Aquat. Bot., 58: 55±64. Rismondo, A., P. Guidetti & D. Curiel, 1997b: Presenza delle fanerogame marine nel Golfo di Venezia: un aggiornamento. Boll. Mus. Civ. Stor. Nat. Venezia, 47: 317±328. Romero, J., G. Pergent, C. Pergent-Martini, M. A. Mateo & C. Regnier, 1992: The detritic compartment in a Posidonia oceanica meadow: litter features, decomposition rates and mineral stocks. P.S.Z.N. I: Marine Ecology, 13(1): 69±83. Sand-Jensen, K., 1975: Biomass, net production and growth dynamics in an eelgrass (Zostera marina L.) population in Vellerup Vig, Denmark. Ophelia, 14: 185±201. Sfriso, A. & P. F. Ghetti, 1998: Seasonal variation in biomass, morphometric parameters and production of seagrasses in the lagoon of Venice. Aquat. Bot., 61: 207±223. Terrados, J. & J. D. Ros, 1992: Growth and primary production of Cymodocea nodosa (Ucria) Ascherson in a Mediterranean coastal lagoon: the Mar Menor (SE Spain). Aquat. Bot., 43: 63±74. Underwood, A. J., 1997: Experiments in Ecology: their logic design and interpretation using analysis of variance. University Press, Cambridge; 504 pp. Wittmann, K. J., 1984: Temporal and morphological variations of growth in a natural stand of Posidonia oceanica (L.) Delile. P.S.Z.N. I: Marine Ecology, 5: 301±316. Young, P. C., 1981: Temporal changes in the vagile epibenthic fauna of two seagrass meadows (Zostera capricorni and Posidonia australis). Mar. Ecol. Prog. Ser., 5: 91±102.