Biomass of Ruppia cirrhosa and Potamogeton pectinatus in a

eschweizerbartxxx

Macrophyte biomass in a Mediterranean lagoon 243

Introduction

In many brackish shallow waters on Mediterranean coasts, primary production is dominated by large dense stands of submerged macrophytes that are important feeding and rearing habitats for waterfowl, fi sh and in-vertebrates. The phytomass in the Ichkeul ecosystem is dominated by two submerged macrophyte species, depending on the freshwater infl ow and resulting sa-linity. Almost half (40 %) of the lagoon area (BCEOM et al. 1995) is covered with monospecifi c beds of Pota-mogeton pectinatus L. and to a much lesser extent by a meadow of Ruppia cirrhosa (Petagna) Grande. The biomass of the phytoplankton is reported to be negli-gible (8–11 µg Chlorophyll-a/l) compared to the high amount of particulate matter suspended in the water column (20–60 mg/l, Ben Rejeb-Jenhani 1989).

R. cirrhosa is very common in medium and large permanent mixosaline estuarine and marine wetlands with an annual mean salinity between 2 and 35 ‰ where fetches are not large (Verhoeven 1980). In Eu-rope, Ruppia taxa are never found in coastal areas under the tidal regime. R. cirrhosa has an annual life cycle, hibernating as short quiescent leaf-bearing sto-lons.

P. pectinatus is a submerged freshwater angiosperm of worldwide distribution that includes many different habitat types, i.e. fresh and brackish waters, standing and running waters and waters of different trophic sta-tus. Among the Potamogetonaceae, only P. pectinatus tolerates high salinity and heavily polluted sites. Tur-bidity is the factor that most frequently limits P. pecti-natus growth (Kantrud 1990). The species reproduces by many different means such as seeds, tubers (subter-

DOI: 10.1127/1863-9135/2007/0168-0243 1863-9135/07/0168-0243 $ 3.25 © 2007 E. Schweizerbartsche Verlagsbuchhandlung, D-70176 Stuttgart

1 Authors’ addresses: UMR 6540 CNRS Dimar «Diversité, Evolution et Ecologie fonctionnelle marine», Centre d’Océa-nologie de Marseille, Université de la Méditerranée, Campus de Luminy, case 901, 13288 Marseille Cedex 9, France. E-mail: [email protected] Department of Biology, University of Freiburg, Germany

Biomass of Ruppia cirrhosa and Potamogeton pectinatus in a Mediterranean brackish lagoon, Lake Ichkeul, Tunisia

Caterina Casagranda1, 2 and Charles François Boudouresque1

With 8 fi gures and 3 tables

Abstract: The biomass of the macrophytes Potamogeton pectinatus L. and Ruppia cirrhosa (Petagna) Grande and their energy input that is at the basis of the functioning of the Ichkeul wetland, northern Tunisia, which harbours a conspicuous population of wintering waterfowl, were investigated. The mean (above- and belowground) biomass of P. pectinatus from July 1993 – April 1994 was at a maximum in September and ranged between 340.9 ± 38.4 and 480.6 ± 91.8 g dry mass (DM)/m² depending on the area. Tubers represented 10–30 % of this biomass. In R. cirrhosa maximum biomass of 368.7 ± 68.1 g DM/m² was recorded in October. In both species about 40 % of the total biomass was allocated to the roots and rhizomes. The production estimates in the Ichkeul lagoon are within the same range as estimates for other Mediterranean lagoons. Other energy sources in the Ichkeul lagoon were litter (60.1 g DM/m²), the overwintered roots and rhizomes (61.4 g DM/m²) and the suspended particulate organic matter (7.64 g DM/m²). Total mean energy income from the macrophytes in the Ichkeul ecosystem during the study period is estimated at 5306 kJ/m².

Key words: Potamogeton pectinatus, Ruppia cirrhosa, biomass, brackish lagoon, Mediterranean.

Fundamental and Applied LimnologyArchiv für HydrobiologieVol. 168/3: 243–255, March 2007© E. Schweizerbart’sche Verlagsbuchhandlung 2007

eschweizerbartxxx

244 Caterina Casagranda and Charles François Bououresque

ranean turions) and rhizomes, that can all be found as regeneration and dispersal mechanisms depending on habitat and environmental stress. The morphological plasticity and growth characteristics vary considerably in different habitats. At Ichkeul, conditions of high salinity in autumn and low temperature make P. pec-tinatus behave as an annual by decomposing and re-generating from tubers when salinity decreases (Hollis 1986).

The great value of P. pectinatus as food for mi-grant and staging waterfowl has long been recognized (Hollis 1986, Tamisier et al. 1987, Tamisier & Bou-douresque 1994, Tamisier et al. 2000), but the implica-tions for the macroinvertebrate consumers are not well understood. One of the most notable features of the macrophyte beds is the high faunal biomass relative to those in adjacent, unvegetated habitats (BCEOM et al. 1995). In the Ichkeul lagoon, a large proportion of the carbon fi xed by primary production enters the detritical pool during the autumn. This biomass is an important reservoir of nutrients and their removal may help to prevent eutrophication and dystrophic crisis. The impact of macrophytes on the benthic foodweb depends on the relative production, decomposition and digestibility. An estimate of the macrophyte produc-tion is needed in order to obtain a quantitative measure of its trophic potential in the functioning of the Ichkeul ecosystem for supporting resident and migratory con-sumer populations (e.g., waterfowl, macroinverte-brates), which utilize the macrophyte beds as feeding areas and refugia.

Production investigations of P. pectinatus have been undertaken by Schiemer & Prosser (1976), Howard-Williams (1978), Howard-Williams (1980), Kautsky (1987), Van Wijk (1988), Menéndez & Comín (1989), Hart & Lovvorn (2000) and Menéndez et al. (2002). Much attention has been focused on R. cirrho-sa (Howard-Williams 1980, Kiørboe 1980, Verhoeven 1980, Ballester 1985, Pérez & Camp 1986, Menéndez & Comín 1989, Calado & Duarte 2000, Menéndez 2002, Menéndez et al. 2002). Little is known about R. maritima L. (Nixon & Oviatt 1973, Edwards 1978, Verhoeven 1980).

The present study is part of a more extensive re-search programme on the functioning of Lake Ichkeul (Tunisia), the purpose of which is to identify the eco-logical and physical characteristics of the ecosystem in order to draw up a predictive forecasting model with a view to devising a conservation management programme taking into account the social and eco-nomic development of the region. On the basis of the phytoplankton study by Ben Rejeb-Jenhani (1989),

Tamisier & Boudouresque (1994) considered the Ichkeul lagoon as a rare example of an oligotrophic coastal lagoon in the Mediterranean basin since algal blooms or dystrophic crises have never been reported from the Ichkeul lagoon (Hollis 1986, Ben Rejeb-Jen-hani 1989). This study describes seasonal changes in above- and belowground phytomass, tuber biomass, litter and suspended particulate organic matter and estimates their energy input that is at the basis of the functioning of this temperate brackish lagoon (Casa-granda & Boudouresque 2005, Casagranda et al. 2005, Casagranda et al. 2006).

Material and methods

Study site

The study was carried out at Lake Ichkeul, an inland brack-ish lagoon of 9,000 ha surrounded by 3,000 ha of temporary marshes on the northern coast of Tunisia (Fig. 1). It is linked by a narrow channel (Tinja channel) to the Lagoon of Bizerte which in turn has an outlet to the Mediterranean Sea. The wet-land is shallow with a mean depth of 2–3 m in winter and 1 m in summer. It is fi lled up with freshwater from autumn and winter rainfall (from 7 wadis, i.e. temporary rivers) that overfl ows into the Lagoon of Bizerte. In summer, high evaporation lowers the water level and allows seawater to enter into the lake. The sa-linity displays considerable seasonal changes from 3 ‰ in the innermost parts in spring to 38 ‰ at the mouth of the Tinja channel in autumn.

Sampling

The lagoon was divided into 4 study areas on the basis of the macrophytal covering (Fig. 1). The western (henceforth re-ferred to as ‘Sejnene’) and the southern (Joumine) areas, sup-plied with freshwater from the wadis, are covered by extensive beds of P. pectinatus. The eastern area (Tinja) close to the Tinja channel and supplied with seawater is covered by a meadow of R. cirrhosa. The central area (Centre) of the Ichkeul lagoon is completely vegetation free. Three replicate samples were taken monthly at a total of 21 sites (Fig. 1) from July 1993 to April 1994. Aboveground phytomass (stems and leaves) was sampled using a sampler which is essentially a section of a metal venti-lation pipe 30 cm in diameter. Belowground phytomass (roots and rhizomes, P. pectinatus tubers) and litter was sampled using a 15 cm inner diameter cylindrical Plexiglas coring tube col-lecting 20 cm of substrate. The sediment cores were rinsed in a 300 µm gauze hand-net and preserved in 75 % ethanol until sorting. The data of Chl-a (fi ltration over a glass fi ber fi lter GF 92, [µg/l]), the suspended particulate organic matter (SPOM) (fi ltration over a 0.45 µm fi lter, [mg/l]), salinity, Secchi depth [m] and photosynthetically available radiation (PAR) (LiCor-meter at depth increments of 5 cm, [µmol m–2 s–1]) were derived from time series in BCEOM et al. (1995).

Biomass and CHN content

The biomass of stems and leaves, roots and rhizomes, litter and P. pectinatus tubers was determined by measuring dry mass

eschweizerbartxxx


(DM) at 60 °C after 48 h. In order to obtain a mean estimate, the biomass estimates for each sampling area were weighted by their contribution (Sejnene, 34 %; Centre, 55 %; Tinja, 3 %; Joumine, 8 %) to the total surface area of the wetland. The av-erage CHN content was measured with a LECO 800 analyser as described by Casagranda & Boudouresque (2002). The en-ergy content of each component was estimated from literature. A t-test was performed to compare the P. pectinatus biomass between the Sejnene and Joumine areas. The goodness of nor-mality fi t was tested by a Kolmogorov-Smirnov statistic, the homogeneity of variances by a F-statistic.

Results

Aboveground phytomass

During the study period from July 1993 to April 1994, only negligible autumn and winter rainfall was reg-istered. Infl ow of seawater into the lagoon continued from summer to winter and was reversed only in Feb-ruary and March (2 months instead of 8 in average years). At the end of spring 1994, unusually high av-erage salinity of about 28 ‰ (Fig. 2) and low water level were observed. The aboveground phytomass is represented in Fig. 2. The meadows of P. pectinatus at Sejnene and Joumine disappeared from October on-ward, and thick layers of dead vegetation piled up all

along the southern shores. Only the R. cirrhosa mead-ow at Tinja remained in place during the study period. Maximum mean aboveground biomass of the P. pecti-natus meadows was reached in September with 250.8 ± 60.2 g DM/m² at Sejnene and 99.4 ± 23.9 g DM/m² at Joumine. The t-test indicated that the biomass at Se-jnene was signifi cantly higher (p < 0.05) than that at Joumine. At Tinja, the peak aboveground R. cirrhosa biomass attained 229.9 ± 55.2 g DM/m² in October. From November to February, a period of winter qui-escence was observed in the R. cirrhosa meadow. In April, 39.1 ± 17.6 g DM/m² of R. cirrhosa were meas-ured. The weighted mean aboveground phytomass over the whole lagoon surface amounted to 101.6 g DM/m². The mean C- and N-content was measured to be 33.6 % and 1.48 %, respectively, yielding a C- and N-production of 34.1 g C/m² and 1.5 g N/m² in the lagoon. The calorifi c value of submerged phanero-gams of the family Potamogetonaceae has a mean of 13.57 kJ/gDM (Nixon & Oviatt 1973), rendering the aboveground biomass as 1378 kJ/m².

Belowground phytomass

The biomasses of the P. pectinatus tubers at Sejnene and at Joumine are shown in Fig. 3. The maximum

Fig. 1. The Ichkeul lagoon, with location of the sampling sites (21). Location of 7 wadis (i.e. temporary rivers), the connection to the sea via the Lagoon of Bizerte, the meadows of Potamogeton pectinatus and Ruppia cirrhosa are indicated.

eschweizerbartxxx


Aboveground phytomass

0

50

100

150

200

250

300

350

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr

[g D

M/m

²]

0

5

10

15

20

25

30

35

40

45

Sal

inity

Sejnene Tinja Joumine Mean salinity Undisturbed

Fig. 2. Seasonal changes in mean aboveground phytomass [g DM/m²] at each study area. Bars = standard error. Mean salinity ac-cording to BCEOM et al. (1995). Dashed line = salinity of the undisturbed ecosystem (Hollis 1986).

P. pectinatus tuber biomass

0

20

40

60

80

100

120

140

160


[g D

M/m

²]

Sejnene Joumine

Fig. 3. Seasonal changes in mean P. pectinatus tuber biomass [g DM/m²] at each study area. Bars = standard error.

eschweizerbartxxx


mean tuber biomass was reached in September with 51.9 ± 20.8 g DM/m² at Sejnene and 105.3 ± 42.1 g DM/m² at Joumine. The tuber biomass at Joumine was signifi cantly higher (p < 0.05) until October and sig-nifi cantly lower (p < 0.05) until April than at Sejnene

although the t-test over the whole period indicated no signifi cant difference between Joumine and Sejnene. The weighted mean over the total wetland amounted to 25.7 g DM/m². The mean C- and N-content was measured to be 42.10 % and 0.92 %, respectively,

Table 1. Mean root and rhizome biomass ± standard error [g DM/m²] at each study area. m.d. = missing data. Study area Sejnene Centre Tinja Joumine weighted meanSpecies P. pectinatus vegetation-free R. cirrhosa P. pectinatus lagoonSurface area [km²] 30.5 49.6 3.0 6.9 90.0 July 114.1 ± 34.2 0 314.9 ± 94.5 409.0 ± 122.7 80.5October 177.8 ± 53.4 0 138.8 ± 41.6 136.2 ± 40.9 75.3December 193.6 ± 77.5 0 m.d. 47.2 ± 18.9 69.2April 146.1 ± 58.5 0 78.7 ± 31.5 120.7 ± 48.3 61.4

Table 2. Maximum phytomass ± standard error [g DM/m²] of the different plant parts viz. aboveground phytomass, roots and rhi-zomes and P. pectinatus tubers at each study area. (–) = R. cirrhosa does not have tubers.

P. pectinatus R. cirrhosa P. pectinatusMacrophyte different parts Sejnene % Tinja % Joumine % Aboveground phytomass 250.8 ± 60.2 52.2 229.9 ± 55.2 62.4 99.4 ± 23.9 29.2Roots and rhizomes 177.8 ± 53.4 37.0 138.8 ± 41.6 37.6 136.2 ± 40.9 40.0Tubers 51.9 ± 20.8 10.8 – 105.3 ± 42.1 30.9Total biomass 480.6 ± 91.8 100 368.7 ± 68.1 100 340.9 ± 38.4 100

Litter biomass

0

20

40

60

80

100

120

140

160

180

200


[g D

M/m

²]

Sejnene Centre Tinja Joumine

Fig. 4. Seasonal changes in mean litter biomass [g DM/m²] at each study area. Bars = standard error.

eschweizerbartxxx


yielding a C- and N-production of 10.81 g C/m² and 0.24 g N/m² in the lagoon. P. pectinatus tubers have a mean calorifi c value of 13.57 kJ/gDM (Hurter 1979) rendering the tuber production as 344 kJ/m².

The mean seasonal dry mass of the roots and rhi-zomes for each area is given in Table 1. On the ba-sis of the October value, the macrophytes would have produced 75.3 g DM/m² of roots and rhizomes in the lagoon. At all areas, about 40 % of the total standing crop by the end of the growing period was allocated to the roots and rhizomes (Table 2). The t-test indicated signifi cant difference (p < 0.05) between Sejnene and Joumine. The roots and rhizome biomass at Joumine was signifi cantly higher (p < 0.001) during summer and autumn and signifi cantly lower (p < 0.01) during winter and spring than at Sejnene. The mean C- and N-content was measured as 36.03 % and 1.50 % respec-tively rendering the C- and N-production as 27.1 g C/m² and 1.1 g N/m². Using the calorifi c value according to Hurter (1979), the roots and rhizomes have a mean energy input of 1009 kJ/m².

The biomass of the overwintered roots and rhi-zomes in the sediment in April (Table 1) was higher in the P. pectinatus than in the R. cirrhosa meadow. At Sejnene the overwintered roots and rhizome biomass amounted to 146.1 ± 58.5 g DM/m², at Tinja 78.7 ± 31.5 g DM/m² and at Joumine 120.7 ± 48.3 g DM/m2 yielding a weighted mean of 61.4 g DM/m². The C- and N-production amounted to 22.1 g C/m² and 0.9 g N/m² respectively. Applying the calorifi c value according to

Meyer (1991) the energy input from the overwintered roots and rhizome biomass yielded 1230 kJ/m².

Litter

The litter biomass is represented in Fig. 4. The maxi-mum mean biomass was reached at Sejnene in Febru-ary with 138.2 ± 48.4 g DM/m², at Tinja in October with 136.9 ± 54.8 g DM/m² and at Joumine in March with 91.5 ± 27.4 g DM/m². In the Centre, the litter mass was always very low, with a maximum biomass of 3.2 ± 1.3 g DM/m² in October. The weighted mean of the whole lagoon amounted to 60.1 g DM/m². The mean C- and N-content was measured as 25.43 % and 1.50 %, respectively, yielding a C- and N-biomass of 15.3 g C/m² and 0.9 g N/m² in the lagoon. The calorif-ic value of 20.03 kJ/gDM (Meyer 1991) was used for the energetic content of the litter rendering the energy income as 1205 kJ/m².

Chlorophyll-a and suspended matter content of the water column

Complete time series of Chl-a were only available for 4 sample sites from BCEOM et al. (1995), i.e. sites 4 and 5 in the Sejnene area and sites 6 and 8 in the central area (Fig. 5). No correlation between Chl-a and transparency (Secchi depth [m], BCEOM et al. 1995) could be found. The mean during the study pe-riod amounted to 4.0 ± 1.4 µg/l indicating a negligi-ble phytoplankton biomass compared to the amount

Chlorophyll-a

0

5

10

15

20

25

30

35

40


[µg/l]

4

5

6

8

Fig. 5. Seasonal changes in Chlorophyll-a [µg/l] at Sejnene (sample sites 4 and 5) and Centre (sample sites 6 and 8). Data according to BCEOM et al. (1995).

eschweizerbartxxx


0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Centrevege--tion%

Dep

th [m

]

00.050.1

0.150.2

0.250.3

0.350.4

0.450.5

SejneneP. pectinatus

00.050.1

0.150.2

0.250.3

0.350.4

0.450.5

TinjaR . cirrhosa

00.050.1

0.150.2

0.250.3

0.350.4

0.450.5

JoumineP. pectinatus

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Centre vegetation-free

Fig. 6. Average PAR profi les [% of inci-dent], total depth [m] and canopy depth [m] at study areas of Lake Ichkeul in August 1993. Attenuation coeffi cients were for P. pectinatus at Sejnene: k = 2.30/m and at Joumine: k = 19.01/m, for R. cirrhosa at Tinja: k = 2.71/m, and for the vegetation-free Centre: k = 4.86/m. Grey areas are below canopies.

iculat

0

50

100

150

200

250

300

350

400


[mg/

l]

Sec

chi d

epth

[m

]

mean Transparency Joumine

[mg/

l]

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

Secc

hi d

epth

[m]

mean Transparency Sejnene Centre Tinja Joumine

Suspended particulate organic matter

Fig. 7. Seasonal changes in mean suspended particulate organic matter (SPOM) [mg/l] at each study area. Data according to BCEOM et al. (1995). Bars = standard error. In the Joumine area, only one sample site (16) was measured.

eschweizerbartxxx


of SPOM in the water column. On the basis of PAR profi les measured at the 21 samples sites mid-August (Fig. 6), the attenuation coeffi cient was calculated for each area. The transparency is signifi cantly correlated with the attenuation coeffi cient (r = –0.98, p < 0.001). In the vegetation-free Centre, the attenuation coeffi -cient (BCEOM et al. 1995) was 4.86/m (Fig. 6). Light was less attenuated at Sejnene (k = 2.30/m) and Tinja (k = 2.71/m), i.e. macrophyte colonization ceased at depths where, on average, 26–32 % of PAR incident on the water surface is received during the growing season. At Joumine, however, the attenuation coeffi -cient was 19.01/m (Fig. 6).

The SPOM occurring in the water column is pre-sented in Fig. 7 from August to April at each area. The transparency is signifi cantly correlated with the SPOM in the water column (r = –0.75, p < 0.001). For the estimation of the energy income from the SPOM, the average value for the whole study period was used, which is 62.0 ± 15.2 mg DM/l at Sejnene, 78.7 ± 30.3 mg DM/l in the Centre, 44.8 ± 14.1 mg DM/l at Tinja and 138.3 ± 55.9 mg DM/l at Joumine, yielding a weighted mean of 76.4 mg DM/l. Using the calorifi c value for detritus according to Cummins et al. (1966) the energy income from the SPOM was estimated as 141 kJ/m².

Discussion

Factors structuring macrophyte colonization

Both species are poorly tolerant of wave action (Ver-hoeven 1980). In shallow Ichkeul, the wave-mixed zone coincides with the macrophyte zone, but the beginning of the growing season coincides with the greatest increases in water level from spring runoff. Once seedlings are established, growing macrophyte stands greatly dampen wave action from winds. The extent of the wave-mixed zone depends on slope, shore aspect in relation to prevailing winds and on lake size, principally fetch. Dense macrophyte stands are therefore able to develop only in the sheltered bays of Sejnene, Joumine and Tinja where the water column above coarser-textured bottoms is less subject to wind-induced turbidity.

Salinity determines species composition of the mac-rophyte community in the Ichkeul lagoon. Although R. cirrhosa is also vital in waters with mean salinity be-low 4 ‰, the vitality of P. pectinatus is at its optimum and it can outcompete R. cirrhosa completely. This is attributed to the larger number of better adapted hi-

bernating organs (tubers, rhizomes, seeds) of P. pecti-natus and the rapidity with which it fi lls both the sub-strate and water column (Verhoeven 1980). The way in which P. pectinatus adapts to salinity, and the actual salt tolerating mechanism involved, remains unclear. The University College London (UCL) greenhouse studies (Hollis 1986), conducted with material from the lagoon, showed that at salinity below 3 ‰, the vi-tality of P. pectinatus is at its optimum level and that any increase in salinity resulted in a reduction in veg-etative growth. Salinity higher than 20 ‰ was found to be fatal to many plants. The strength of the effect depends on the age of the plant when it is subjected to increased salinity (Hollis 1986). The greatest effect will come if the plants are very young (< 9 weeks) when the high salinity occurs. Once the plants have started to fl ower, normally in June when salinity in the lagoon is between 8–10 ‰, and have set seed, the impact of the salinity increase is not as great. Dense stands of P. pectinatus are therefore able to develop only in the sheltered bays of Sejnene and Joumine close to the freshwater inputs where sediments have a relative high proportion of silt and organic matter. In Ichkeul lagoon, P. pectinatus occurs in a regular annu-al cycle where salinities of 40 ‰ occur by the end of summer as the sea fl ows in. Winter rains then freshen the lake for spring growth of P. pectinatus. The land-ward extension of P. pectinatus is determined by the water level in late spring. Areas less than 30 cm deep when P. pectinatus is growing will result in desicca-tion of the areal shoots. P. pectinatus appears to be at its optimum within the lagoon in late spring at water levels between 80 and 120 cm in sheltered areas.

The lakeward extension of macrophytes will be controlled by the depth of the wave-mixed zone. In shallow wetlands where fetches are large, the wave-mixed zone deepens and the effects of wave action on sediment-induced turbidity prevent growth of macro-phytes (Anderson 1978). In the central area of Ichkeul lagoon, macrophytes are absent, possibly because of very large fetch and the substrate is high in easily sus-pendible fi ne clays deposited there. The attenuation coeffi cient in the Centre (Fig. 6) is less than 10 % of PAR incident on the water surface received at depths of macrophyte-covered areas. There, the canopy of P. pectinatus and R. cirrhosa formed at the water sur-face dampen light attenuation caused both by man-made and natural suspended particles by reducing wa-ter movement (Schiemer & Prosser 1976). However, macrophyte colonization ceased at depths where, on the average, 26–32 % of PAR incident on the water surface is received during the growing season (Fig. 6).

eschweizerbartxxx


This indicates that photosynthesis is rapidly light satu-rated and that this could be a competitive disadvan-tage in turbid waters. However, at Joumine, the Sec-chi depth was reduced to 9 cm at Joumine during the growing season (BCEOM et al. 1995), the SPOM was maximum (BCEOM et al. 1995). According to Kantrud (1990), Secchi transparency less than 20 cm indicates waters that will not support P. pectinatus. The water column at Joumine is subject to high pollution-in-duced turbidity from municipal sewage from the town of Mateur through the Joumine wadi, essentially solu-ble reactive phosphorous (P-PO4, 101.7 ± 25.3 µmol/l) and ammonium (N-NH4, 1130.7 ± 470.5 µmol/l), largely exceeding CEE edibility limits (BCEOM et al. 1995). However, P-PO4 and N-NH4 levels in the Joumine water column were low (0.2 ± 0.1 µmol/l and 5.8 ± 4.8 µmol/l, respectively) (BCEOM et al. 1995). According to Hart & Lovvorn (2000), cycles of phy-toplankton-macrophyte dominance can be a function of water depth, the competitive advantage being with the macrophytes in shallow waters. The Chl-a values available from BCEOM et al. (1995) in fact show only one value exceeding the low general mean (Fig. 5). Kantrud (1990) postulated that P. pectinatus cannot compete in waters low in P and shows more affi nity for waters high in P but suffers from turbidity. Accord-ing to Van Wijk (1988), P. pectinatus has adaptations to unfavourable light climates common to eutrophic or brackish waters. These adaptations include increased relative turion production (Fig. 3) and increased shoot length, fewer and coarser leaves and stems and a light-er green colour (pers. obs.). Plants can reach the water surface at an earlier stage and concentrate foliage in the surface layer before phytoplankton or macroalgae have the advantage. In the past, phytoplankton blooms have never been reported from Ichkeul lagoon (Hollis 1986, Ben Rejeb-Jenhani 1989, BCEOM et al.1994). Low P-PO4 and N-NH4 levels in the water column indi-cate macrophyte uptake; however, most P-PO4 and N-NH4 is supposed to be concentrated in the sediments. Ben Rejeb-Jenhani (1989) showed that the Ichkeul sediments have high P-PO4 adsorption capacity. As the dissolved oxygen is in the range of 83–117 % of saturation, the BOD5 low (1.5–2.8 mg/l) and the redox potential high (370–400 mV) (BCEOM et al. 1995), N fi xation rather than denitrifi cation would be more likely. Under aerobic conditions and lack of assimi-lable N and P in the water column, sediments are the major nutrient source for P. pectinatus. If P. pectinatus does not leak P as some common submerged plants are thought to do (Madsen 1986), the species could be considered rather a P reservoir instead of a pump.

Under such circumstances, we postulate that P. pec-tinatus can control P availability and suppress light-limiting phytoplankton blooms.

Gerloff & Krombholz (1966) employed the tissue content of total nitrogen and phosphorous as an index of element availability in water. For this approach, the minimum tissue concentration that is necessary for maximum growth must be determined. In our study, the tissue content of phosphorous has not been ana-

(A) Water column

0

20

40

60

80

0 1 2 3 4 5 6 7

Ptot [µmol/l]

Nto

t[ µ

mo

l/l]

(B) Interstitial water

0

20

40

60

80

100

120

140

160

180

0 1 2 3 4 5 6 7 8 9 10 11 12 13

Ptot [µmol/l]

Nto

t[ µ

mo

l/l]

(C) Sediment, root layer

0

20

40

60

80

100

120

140

160

180

200

220

240

260

280

300

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

Ptot [mmol/kg]

Nto

t [m

mo

l/kg

]

Fig. 8. The N/P ratio in the water column (A), interstitial wa-ter (B) and sediments, root layer (C) of Lake Ichkeul mid-September 1993. The line gives the theoretical distribution of [P]=10[N] (Chiaudani & Vighi 1976).

eschweizerbartxxx


lysed but the nitrogen content (1.48 %) was close to the minimum content of 1.3 % published by Gerloff & Krombholz (1966). This supports the viewpoint that in fertile lakes nitrogen is more likely to limit macrophyte growth than would phosphorous. An al-ternative approach of relating nutrient concentrations in the water to potential for supporting plant growth has been developed by Chiaudani & Vighi (1976). In this approach, the N/P ratio of 10/1 is considered to be an optimum assimilation proportion where neither phophorous nor nitrogen could be considered as lim-iting factor. In Ichkeul lagoon, distribution of N and P (BCEOM et al. 1995) differed markedly from this ratio in both water column and sediment (Fig. 8). In the sediment, the N/P ratio was less than 10/1. In the macrophyte-free sediments it was 10/2. However, in the water column, the N/P ratio was much greater than 10/1. BCEOM et al. (1995) reported high organic N values (155.6 ± 65.2 µmol/l) in the water column during autumn which indicates inputs from meadow decomposition. Howard-Williams & Davies (1979) showed that P was lost to the water much more rapidly than N from decomposing P. pectinatus. At Ichkeul, rapid elimination of P-PO4 from the water column by sediments has been demonstrated (Ben Rejeb-Jenhani 1989). Consequently, if there are no meadows to as-similate the adsorbed P, one might envisage the sedi-ments becoming progressively poorer in nitrogen and richer in phosphorous.

Macrophyte biomass and productionThe measurement of submerged angiosperm produc-tion can be undertaken by means of harvesting tech-niques, incubation experiments or marking methods (Schwoerbel 1994). Incubation experiments are lim-ited to short term investigations and marking methods cannot be used for R. cirrhosa and P. pectinatus as the leaves are too small and narrow and their verti-cal stems are too strongly branched. If plants show a seasonal development pattern, the annual cumulative production can be determined by measuring plant bio-mass at the moment of maximum standing crop. The P. pectinatus and the R. cirrhosa stands show complete annual growth, the aboveground parts produced in the previous season and lost in winter are completely de-composed at the time of maximum biomass and no corrections are necessary. The disadvantage of this method is the repeated destructive sampling necessary for the determination of the moment of maximum bio-mass.

There was 30–50 % more income from below-ground biomass than in both submersed macrophytes

and SPOM combined (Table 2). The belowground biomass was higher in the P. pectinatus than in the R. cirrhosa meadow. In the literature, the belowground biomass can vary from 4 % to 78 % of total plant weight (Howard-Williams 1980, Kautsky 1987, Van Wijk 1988, Menéndez & Comín 1989). The signifi -cant differences between the P. pectinatus meadows at Sejnene and Joumine leads to the conclusion that “lighter” plants from polluted sites allocated more biomass to underground parts than “heavier” plants from unpolluted sites. Ozimek et al. (1986) found that P. pectinatus from polluted sites grew faster reaching peak weight in July and began dying earlier than plants from unpolluted sites that continued to gain weight un-til October.

The seed material collected from Ichkeul has a high temperature requirement for germination (20 °C, Hol-lis 1986). These temperatures only tend to be found in the lake during summer when salinity is high, re-ducing the success of germination. According to Van Wijk (1988), seeds are mainly a means of long term survival. Consequently, it is thought that the major re-establishment of the P. pectinatus meadow in Ichkeul during the following summer derives from tuber re-cruitment rather than from seed recruitment. The esti-mated tuber production is between 10 and 30 % of the P. pectinatus production (Table 2). During the winter of the study period, the largest population of Common Pochards (Aythya ferina) for the last ten years was counted (26,000 between November 1993 and Febru-ary 1994, BCEOM et al. 1995). Tubers contain much more carbon than foliage. The high carbohydrate ac-cumulation by tubers explains their exhaustive use by migrant and wintering waterfowl. This predation has been favoured during the study period by the lack of macrophyte cover (Fig. 2) and low water level.

In autumn, tons of P. pectinatus decomposed on shore as stems and leaves wash up after becoming col-lapsed and detached. At Joumine, litter formation was delayed by 2 months while at Tinja it was advanced (Fig. 4) although the meadow remained in place long-est (Fig. 2). This indicates a redistribution of the litter mass from the south east to the north east favoured by the dominant SW winds during late summer (BCEOM et al. 1995). Litter bag experiments by Verhoeven (1980) showed how important shredding and grazing invertebrates were for macrophyte decomposition. He found that half of the material had decomposed within two months and that after a year practically no plant material was left. In Ichkeul, the pool was supplied with material from the packs on the lee side shores moved around by wave action.

eschweizerbartxxx


Maximum macrophyte biomass was assumed to equal the lagoonwide primary production. Total mean energy income in the Ichkeul ecosystem during the study period is estimated as 5306 kJ/m².

Wetland production in relation to latitude and trophic implications

In Table 3 the maximum biomass data from this study are compared to values given in the literature. The production estimates in the Ichkeul lagoon are with-

in the same range as estimates for other Mediterra-nean lagoons in the Ebro delta (Pérez & Camp 1986, Menéndez & Comín 1989, Menéndez 2002, Menén-dez et al. 2002) and in the Camargue (Verhoeven 1980, Van Wijk 1988). Greatest P. pectinatus biomasses (> 1500 g DW/m²) were found in Africa (Howard-Wil-liams 1978). This is exceptionally high for any sub-merged hydrophyte. Congdon & McComb (1981) found a maximum dry-mass standing crop of 503 g DM/m² for Ruppia sp. in the Blackwood River estuary, south-west Australia. Much lower values are reported

Table 3. Comparison of published data on maximum biomass (g DM/m²) sorted by distance from the equator (latitude). a = original values in g AFDM/m², converted to dry mass according to author’s indication. Study area B Authors Ruppia maritimaTvärminne (Finland) 106a Verhoeven (1980)Texel (The Netherlands) 50a Verhoeven (1980)Camargue (France) 129a Verhoeven (1980)Bissel Cove (R.I. USA) 180–1460 Nixon & Oviatt (1973)Huizache-Caimanero lagoon complex (Mexico) 866 Edwards (1978) Ruppia cirrhosaTvärminne (Finland) 156a Verhoeven (1980)Ringkøbing Fjord (Denmark) 19 Kiørboe (1980)Texel (The Netherlands) 146a Verhoeven (1980)Zeeland (The Netherlands) 141a Verhoeven (1980)Camargue (France) 136a Verhoeven (1980)Fangar bay (Spain) 407 Pérez & Camp (1986)Buda lagoon (Spain) 457 Menéndez et al. (2002)Tancada lagoon (Spain) 660a Menéndez & Comín (1989)Tancada lagoon (Spain) 709a Menéndez (2002)Alfaques bay (Spain) 263 Pérez & Camp (1986)Santo André lagoon (Portugal) 295 Calado & Duarte (2000)Mar Menor lagoon (Spain) 111 Ballester (1985)Ichkeul lagoon (Tunisia) 369 this studyWilderness lakes system (South Africa) 83 Howard-Williams (1980) Potamogeton pectinatusTvärminne (Finland) 29a Verhoeven (1980)Tvärminne (Finland) 96 Van Wijk (1988)Askö (Sweden) 13 Kautsky (1987)Lauwersmeer (The Netherlands) 55 VanWijk (1988)Texel (The Netherlands) 52a Verhoeven (1980)Veluwemeer (The Netherlands) 55 Van Wijk (1988)Schouwen (The Netherlands) 22a Verhoeven (1980)Zeeland (The Netherlands) 723 Van Wijk (1988)Limburg brooks (The Netherlands) 202 Van Wijk (1988)Neusiedlersee (Austria) 7 Schiemer & Prosser (1976)Camargue (France) 70a Verhoeven (1980)Camargue (France) 505 Van Wijk (1988)Creighton wetland (Wy. USA) 160 Hart & Lovvorn (2000)Twin Buttes wetland (Wy. USA) 520 Hart & Lovvorn (2000)Buda lagoon (Spain) 609 Menéndez et al. (2002)Tancada lagoon (Spain) 468a Menéndez & Comín (1989)Ichkeul lagoon (Tunisia) 411 this studyWilderness lakes system (South Africa) 415 Howard-Williams (1980)Swartvlai (South Africa) 2717a Howard-Williams (1978)

eschweizerbartxxx


from Neusiedlersee (Schiemer & Prosser 1976), the Ringkøbing Fjord (Kiørboe 1980) and the Askö area (Kautsky 1987) with maximum biomasses of 3–28 g DM/m². According to Van Wijk (1988) this is less due to strongly branching leafy stems colonizing the water column and never forming a dense canopy (Kautsky 1987) than to less favourable climatic conditions.

In marine and estuarine wetlands, the extent and effects of invertebrate herbivory on submersed macro-phytes is poorly understood. But the study of the ben-thic primary consumers in Ichkeul reveals that a few species can signifi cantly decrease macrophyte biomass (Casagranda et al. 2006). If waterfowl feeding within a wetland depend on food availability, one might expect that, in an area where food is abundant, abundance of waterfowl should be high. The macrophyte biomass is negatively and signifi cantly correlated with latitude (r = –0.73, p < 0.001), decreasing with increasing distance from the equator. The correlation between latitude and phytomass suggests that mild weather conditions and absence of extreme temperatures, especially cold tem-peratures, might allow prolonged macrophyte produc-tion and a continuous supply of food for higher trophic levels. Tamisier et al. (2000) demonstrated that densi-ties of macrophyte feeding waterbirds within a limited geographic area were linked to food availability. If this pattern holds true throughout the non-breeding range of palearctic breeding migrant waterbirds, and water-bird density is linked to wetland productivity, it can be predicted that warm temperate and tropical wetlands will support higher waterbird densities than cold tem-perate wetlands. High densities of waterbirds at the Ichkeul lagoon (180,000–230,000 birds) (Tamisier & Boudouresque 1994), and other warm temperate sites on the western Mediterranean coast (Tamisier 1987), support this prediction.

On the assumption of the replacement of R. cirrhosa by P. pectinatus, the phytomass which would be avail-able to the hibernating waterfowl at Ichkeul, would be a great deal less than that actually available from P. pectinatus. How the rest of the wetland community re-fl ects differences in primary production between these states depends on how production fl ows through food-webs. In the P. pectinatus meadow macro-invertebrate consumers are mostly shredder and epibenthic depos-it-feeders (gastropods and amphipods) (Casagranda et al. 2005, Casagranda et al. 2006). In the R. cirrhosa meadow, consumers are also fi lter-feeders and endo-benthic deposit-feeders (Casagranda & Boudouresque 2005). State shifts in vegetation structure and resulting forms of primary production might have a profound impact on the structure and function of higher trophic

levels. Elucidating these trophic linkages will require a variety of research approaches, such as analysis of diet and secondary production, stable isotope analysis and integrative models.

Acknowledgements

This study was carried out as part of the international pro-gramme « Etude pour la sauvegarde du Parc National de l’Ichkeul » fi nanced by the Kreditanstalt für Wiederaufbau (KfW) under the aegis of the Tunisian authorities, in particular the Agence Nationale pour la Protection de l’Environnement. Thanks are due to members of the Groupement d’Intérêt Sci-entifi que (GIS) Posidonie for fi eld assistance, to colleagues at the UMR 6540 CNRS Dimar (Diversité, Evolution et Ecolo-gie fonctionnelle marine) for advice and work facilities. The present research was supported by a Ph. D. grant from the Landesgraduiertenförderungsgesetzes (LGFG) Germany. The study would not have been possible without the support of Prof. Jürgen Schwoerbel, mentor and friend. Finally, the authors are grateful to two anonymous referees for very valuable comments and to Michael Paul for improving the English text.

References

Anderson, M. G., 1978: Distribution and production of sago pondweed (Potamogeton pectinatus L.) on a northern prairie marsh. – Ecology 59: 154–160.

Ballester, R., 1985: Biomasa, estacionalidad y distribuión de tres macrófi tos: Ruppia cirrhosa, Cymodocea nodosa y Caulerpa prolifera en el Mar Menor (Murcia, SE de España). – Ann. Biol. (Biología Ambiental 1, Universidad de Murcia, Spain) 4: 31–36.

BCEOM (Bureau Central d’Etudes pour les Equipements d’Outre-Mer), Fresinus Consult, CE Salzgitter & STUDI, 1994: Etude pour la sauvegarde du Parc National de l’Ich-keul. Rapport de 1ère partie: Situation actuelle de la zone d’étude et état actuel de l’écosystème / Ministère de l’En-vironnement et de l’Aménagement du Territoire, ANPE (Agence Nationale de Protection de l’Environnement) Tunis (Tunisia). – BCEOM publ., Tunis (Tunisia), pp. 1–284.

– 1995: Etude pour la sauvegarde du Parc National de l’Ich-keul. Rapport de 3ème partie: Mesures et études spécifi ques / Ministère de l’Environnement et de l’Aménagement du Ter-ritoire, ANPE (Agence Nationale de Protection de l’Environ-nement) Tunis (Tunisia). – BCEOM publ., Tunis (Tunisia), pp. 1–421.

Ben Rejeb-Jenhani, A., 1989: Le lac Ichkeul: conditions de mi-lieu, peuplements et biomasses phytoplanctoniques. – Ph. D. Thesis Univ. Tunis (Tunisia), pp. 1–52.

Calado, G. & Duarte, P., 2000: Modelling growth of Ruppia cirrhosa. – Aquat. Bot. 68 : 29–44.

Casagranda, C. & Boudouresque, C. F., 2002: A sieving meth-od for rapid determination of size-frequency distribution of small gastropods. Example of the mud snail Hydrobia ven-trosa (Gastropoda: Prosobranchia). – Hydrobiologia 485: 143–152.

– – 2005: Abundance, Population Structure and Production of Scrobicularia plana and Abra tenuis (Bivalvia: Scrobicu-laridae) in a Mediterranean Brackish Lagoon, Lake Ichkeul, Tunisia. – Internat. Rev. Hydrobiol. 90: 376–391.

eschweizerbartxxx


Casagranda, C., Boudouresque, C. F. & Francour, P., 2005: Abundance, Population structure and Production of Hydro-bia ventrosa (Gastropoda: Prosobranchia) in a Mediterranean brackish lagoon, Lake Ichkeul, Tunisia. – Arch. Hydrobiol. 164: 411–428.

Casagranda, C., Dridi, M. S. & Boudouresque, C. F., 2006: Abundance, population structure and production of macro-invertebrate shredders in a Mediterranean brackish lagoon, Lake Ichkeul, Tunisia. – Estuar. Coast. Shelf Sci. 66: 437–446.

Chiaudani, G. & Vighi, M., 1976: Comparison if different tech-niques for detecting limiting or surplus nitrogen in batch cultures of Selen astrum capricornutum. – Wat. Res. 10: 725–729.

Congdon, R. A. & McComb, A. J., 1981: The vegetation of the Blackwood River estuary, south-west Australia. – J. Ecol. 69: 1–16.

Cummins, K. W., Coffmann, W. P. & Roff, P.A., 1966: Trophic relationships in a small woodland stream. – Verh. Internat. Verein. Limnol. 16: 627–638.

Edwards, R. R. C., 1978: Ecology of a coastal lagoon complex in Mexico. – Estuar. Coast. Mar. Sci. 6: 75–92.

Gerloff, G. C. & Krombholz, P. H., 1966: Tissue analysis as a measure of nutrient availability for the growth of angiosperm aquatic plants. – Limnol. Oceanogr. 11: 529–537.

Hart, E. A. & Lovvorn, J. R., 2000: Vegetation dynamics and primary production in saline, lacustrine wetlands of a Rocky Mountain basin. – Aquat. Bot. 66: 21–39.

Hollis, G. E., 1986: Modelling and management of the inter-nationally important wetland at Garaet El Ichkeul, Tunisia. Final Report on Contract ENV-676-UK (H) of the Third Environment Research Programme of the Commission of the European Communities. – IWRB (International Water-fowl Research Bureau) Slimbridge (U.K.) Spec. publ. 4, pp. 1–121.

Howard-Williams, C., 1978: Growth and production of aquatic macrophytes in a south temperate saline lake. – Verh. Inter-nat. Verein. Limnol. 20: 1153–1158.

– 1980: Aquatic macrophyte communities of the Wilderness lakes: community structure and associated environmental conditions. – J. Limnol. Soc. South. Afr. 6: 85–92.

Howard-Williams, C. & Davies, B. R., 1979: The rates of dry matter and nutrient loss from decomposing Potamogeton pec-tinatus in a brackish south-temperate coastal lake. – Fresh-wat. Biol. 9: 13–21.

Hurter, H. U., 1979: Die Nahrungsökologie des Blässhuhns Fulica atra an den Überwinterungsgewässern im nördlichen Alpenvorland. – Ornithol. Beob. 76: 257–288.

Kantrud, H. A., 1990: Sago pondweed (Potamogeton pectina-tus L.): A literature review. – U.S. Fish and Wildlife Service, Fish and Wildlife Resource Publication 176. Jamestown ND: Northern Prairie Wildlife Research Center, pp. 1–98.

Kautsky, L., 1987: Life-cycles of three populations of Pota-mogeton pectinatus L. at different degrees of wave exposure in the Askö area, Northern Baltic proper. – Aquat. Bot. 27: 177–186.

Kiørboe, T., 1980: Production of Ruppia cirrhosa (Petagna) Grande in mixed beds in Ringkøbing Fjord (Denmark). – Aquat. Bot. 9: 135–143.

Madsen, J. D., 1986: The production and physiological ecology of the submerged aquatic macrophyte community in Badfi sh Creek, Wisconsin. – Ph. D. Thesis Univ. Wisconsin, Madison (USA), pp. 1–449.

Menéndez, M., 2002: Net production of Ruppia cirrhosa in the Ebro delta. – Aquat. Bot. 73: 107–113.

Menéndez, M. & Comín, F. A., 1989: Seasonal patterns of biomass variation of Ruppia cirrhosa (Petagna) Grande and Potamogeton pectinatus L. in a coastal lagoon. – Proc. 22nd European Marine Biology Symp., Barcelona (Spain) 53: 633–638.

Menéndez, M., Hernandez, O. & Comín, F.A., 2002: Spatial distribution and ecophysiological characteristics of macro-phytes in a Mediterranean coastal lagoon. – Estuar. Coast. Shelf Sci. 55: 403–413.

Meyer, E., 1991: Pattern of invertebrate community structure, abundance and standing crop in a Black Forest stream: Re-sults of a 3-year study. – Verh. Internat. Verein. Limnol. 24: 1840–1845.

Nixon, S. W. & Oviatt, C. A., 1973: The ecology of a New Eng-land salt marsh. – Ecol. Monogr. 43: 463–498.

Ozimek, T., Prejs, K. & Prejs A., 1986: Biomass and growth rate of Potamogeton pectinatus L. in lakes of different trophic state. – Ekol. Poll. 34: 125–131.

Pérez, M. & Camp, J., 1986: Distribución espacial y biomasa de las fanerógamas marinas de las bahías del delta del Ebro. – Inv. Pesq. 50: 519–530.

Schiemer, F. & Prosser, M., 1976: Distribution and biomass of submerged macrophytes in Neusiedlersee. – Aquat. Bot. 2: 289–307.

Schwoerbel, J., 1994: Methoden der Hydrobiologie, Süßwas-serbiologie. – Gustav Fischer Verlag Stuttgart/Jena/New York, pp. 1–368.

Tamisier, A., Bonnet, P., Bredin, D., Dervieux, A., Rehfi sh, M., Rocamora, G. & Skinner, J., 1987: L’Ichkeul (Tunisie), quartier d’hiver exceptionnel d’Anatidés et de foulques. Im-portance, fonctionnement et originalité. – L’Oiseau et RFO (Revue Française d’Ornithologie) 57 : 296–306.

Tamisier, A. & Boudouresque, C. F., 1994: Aquatic bird popu-lations as possible indicators of seasonal nutrient fl ow at Ichkeul lake, Tunisia. – Hydrobiologia 279/280: 149–156.

Tamisier, A., Dehorter, O., Defosse, A., Poydenot, F., Gravez, V. & Boudouresque, C. F., 2000: Modelling aquatic ecosys-tems: benefi ts, costs and risks for a fi eld biologist. Ichkeul lake, Tunisia, a case study. – In: Comin, F. A., Herrera, J. A. & Ramírez, J. (eds): Limnology and aquatic birds. Monitor-ing, modelling and management. – Universidad Autónoma de Yucatán publ., Mérida (Mexico), pp. 185–203.

Van Wijk, R. J., 1988: Ecological studies on Potamogeton pectinatus L. I. General characteristics, biomass produc-tion and life cycles under fi eld conditions. – Aquat. Bot. 31: 211–258.

Verhoeven, J.T.A., 1980: The ecology of Ruppia-dominated communities in western Europe. III. Aspects of production, consumption and decomposition. – Aquat. Bot. 8: 209–253.

Submitted: 13 April 2005; accepted: 17 November 2006.

Documents

Biomass of Ruppia cirrhosa and Potamogeton pectinatus in a