Marine Biology

, Volume 143, Issue 6, pp 1229–1238

Short- and long-term effects of eutrophication on the secondary production of an intertidal macrobenthic community

Authors

    • Institute of Marine Research (IMAR), Department of ZoologyUniversity of Coimbra
  • M. A. Pardal
    • Institute of Marine Research (IMAR), Department of ZoologyUniversity of Coimbra
  • A. I. Lillebø
    • Institute of Marine Research (IMAR), Department of ZoologyUniversity of Coimbra
  • U. Azeiteiro
    • Institute of Marine Research (IMAR), Department of ZoologyUniversity of Coimbra
  • J. C. Marques
    • Institute of Marine Research (IMAR), Department of ZoologyUniversity of Coimbra
Article

DOI: 10.1007/s00227-003-1133-5

Cite this article as:
Dolbeth, M., Pardal, M.A., Lillebø, A.I. et al. Marine Biology (2003) 143: 1229. doi:10.1007/s00227-003-1133-5

Abstract

Secondary production of a macrobenthic community at an intertidal mudflat was estimated for 33 successive months. Sampling was carried out along a eutrophication gradient, including non-eutrophied Zostera meadows, an intermediate muddy area, and a strongly eutrophied sand-muddy flat, where macroalgal blooms of Enteromorpha spp. usually occur. The Zostera meadows were always the most productive habitat (145–225 g ash-free dry weight m−2 year−1). In the short term, the macroalgal bloom benefited the total estuarine production by enhancing the annual production in the eutrophied area. Nevertheless, our results show that this increase was short lived and in no way sufficient to match the production in the Zostera meadows. In the long term, the present study provides evidence that the disappearance of macrophyte beds, as a result of ongoing eutrophication, constitutes a major threat to the sustainability of the estuarine ecosystem.

Introduction

All over the world eutrophication, as a consequence of increased anthropogenic-derived activities, has become a prominent problem in coastal areas (Jørgensen and Richardson 1996; Flindt et al. 1999; Cloern 2001). The phenomenon of coastal eutrophication is complex: physical and biological characteristics of the ecosystem and climate combined with nutrient loading itself are thought to affect strongly the outcome of eutrophication (Jørgensen and Richardson 1996; Cloern 2001). Ecosystem processes (nutrient and trophic dynamics) are affected by increased eutrophication, which in turn affects the dynamics and energetics of the whole system (Jørgensen and Richardson 1996; Cloern 2001).

At the system level, one of the more drastic effects concerns modifications at the primary production level, which consequently affects heterotrophic organisms depending on that production. Frequently, the decline of macrophyte assemblages detrimental to macroalgae have been reported (Marques et al. 1997; Flindt et al. 1999; Raffaelli 1999; Asmus and Asmus 2000). This leads to changes in the composition and structure of macrobenthic communities (Norkko and Bonsdorff 1996; Pardal 1998; Bachelet et al. 2000) and species production (Marques et al. 1994; Lillebø et al. 1999; Pardal et al. 2000, 2002; Ferreira 2001; Cardoso et al. 2002). The amplitude of the responses will depend on the characteristics of the ecosystem itself and on species-specific responses (Jørgensen and Richardson 1996; Norkko et al. 2000).

Secondary production, as a quantification of the ecosystem dynamics, can be used to assess the response of the ecosystem to environmental stressors (Tumbiolo and Downing 1994). Considering that estuaries are one of the most productive natural systems (Levin et al. 2001), what happens with secondary production in a scenario of eutrophication? It is clear that changes in primary producers due to eutrophication affect the production of the other trophic levels (Flindt et al. 1999; Pardal et al. 2000, 2002; Beukema et al. 2002; Cardoso et al. 2002) but few quantitative studies have approached the topic for the whole macrobenthic community and from a long-term perspective. The main objective of the present work was to estimate intertidal macrobenthic secondary production along an eutrophication gradient, to understand the effects of eutrophication on the production dynamics. Special emphasis was given to the replacement of primary producers, namely macrophytes by green macroalgae. As a case study, the Mondego estuary (western Portugal) was used.

Material and methods

Study site

The Mondego estuary (Fig. 1) is located in a warm temperate region. It has two arms, north and south, with different hydrologic characteristics. Sampling occurred in the south arm, where water circulation is dependent on tidal activity and on small freshwater input from a tributary, the Pranto river, which is controlled by a sluice. The freshwater discharge from this tributary is regulated according to water needs of the rice crop of Pranto Valley. Consequently, the freshwater entering the system varies with both precipitation and river water management, which in turn determines the variation of other abiotic variables: salinity, nutrients in the water column and sediments, water-flow velocity, and light extinction coefficients (Marques et al. 1997; Pardal et al. 2000, 2002; Martins et al. 2001).
Fig. 1.

The location of the Mondego estuary, indicating sampling stations and percentage of Zostera noltii cover in 1986 and 1993

Anthropogenic activities in the Mondego estuary have been the cause of high environmental pressure on the estuary that, coupled with specific physical characteristics (water residence time, hydrodynamics, and depth) and climate conditions (precipitation), has resulted in an ongoing process of eutrophication over the past two decades. The downstream areas of the south arm still remain relatively unchanged, exhibiting Zostera noltii meadows. In comparison, in the inner areas the macrophyte community has completely disappeared (Fig. 1) and Enteromorpha spp. blooms have been observed during the last 20 years.

Sampling

Samples were taken during low water tide at three different environments: (1) Zostera meadows, where the macrophyte Z. noltii community was present; (2) an intermediate area, a muddy flat; and (3) a eutrophied area, a sand-muddy flat in the inner parts of the estuary. At the three sites ten cores (with 141-cm2 section, approximately 3 l sediment) were taken to a depth of 20 cm. Each sample was sieved through a 500-μm mesh sieve bag. At the laboratory, organisms collected were identified to the species level, counted, measured, and their ash-free dry weight (AFDW) assessed, after combustion for 8 h at 450°C (shells of mollusks included, Bachelet 1982).

From February 1993 to June 1994 samples were taken fortnightly, followed by monthly sampling until September 1995. In the eutrophied area, Enteromorpha spp. exhibited a typical spring bloom in 1993, followed by an algal crash in early summer. In 1994, an especially rainy year, no macroalgal bloom was observed owing to the strong hydrodynamics and low salinity, which inhibited Enteromorpha spp. growth (Lillebø et al. 1999; Pardal et al. 2000, 2002; Martins et al. 2001; Cardoso et al. 2002). In the spring and summer of 1995 the Enteromorpha spp. biomass again achieved relatively high values, but it was not enough to be considered a typical spring bloom (Ferreira 2001; Cardoso et al. 2002). More detailed descriptors of the macrophyte and macroalgal biomasses are reported elsewhere (Lillebø et al. 1999; Pardal et al. 2000, 2002; Ferreira 2001; Cardoso et al. 2002).

Biodiversity

Species richness was estimated by computing the number of species for each sampling date. Heterogeneity was calculated using species biomasses according to the Shannon–Wiener index:
$$ {H' = - {\sum\limits_{i = 1}^n {{\left( {p_{i} } \right)}{\left( {\log _{2} p_{i} } \right)}} }} $$
where n is the number of species and pi is the proportion of the biomass of species i in a community.

Secondary production estimates

Data on production estimated by the increment summation method, based upon cohort identification (Winberg 1971), were already available for Hydrobia ulvae (Lillebø et al. 1999, Cardoso et al. 2002), Cyathura carinata (Marques et al. 1994; Ferreira 2001), Ampithoe valida, and Melita palmata ( Pardal et al. 2000, 2002). For other representative species (Carcinus maenas, Capitella capitata, Hediste diversicolor, Heteromastus filiformis, Haminoe hydatilis, and Littorina littorea), Sprung's (1993) method was used. For Cerastoderma edule and Scrobicularia plana, Tumbiolo and Downing's (1994) equation was applied. To estimate the temporal variation of production both equations were applied with a temporal correction.

From Sprung's (1993) equation, P was estimated according to the expression
$$ P_{\Delta t} = \left( {{{P/\bar B_{{\rm{spec}}} } \over {365}} \times \bar w^{ - 0.25} } \right) \times \bar w_{\Delta t} ^{ - 0.25} \times \bar B_{\Delta t} \times \Delta t $$
where \({P/\bar{B}_{{spec}} }\) is the estimate of the typical annual \({P/\bar{B}}\) of the species found in the literature, is the annual mean body weight,Δt is the mean body weight difference between two sampling dates, \({\bar{B}_{{\Delta t}} }\) is the mean biomass difference between two sampling dates (all in grams AFDW m−2), and Δt is the difference between two sampling dates (days).
From Tumbiolo and Downing's (1994) equation, P was estimated according to the expression
$$ \log P_{\Delta t} = \left( {0.18 + 0.97\log \bar B_{\Delta t} - 0.22\log w_{\rm{m}} + 0.04\bar T_{\Delta t} - 0.014\bar T_{\Delta t} \log (Z + 1)} \right) \times {{\Delta t} \over {365}} $$
where \({\bar{B}_{{\Delta t}} }\) is the mean biomass of two sampling dates (grams DW m−2), wm is the maximum individual body weight (milligrams DW m−2),Δt is the mean bottom water temperature of two sampling dates (°C), Z is the mean depth, and Δt is the difference between two sampling dates (days). The final result was converted to grams AFDW m−2 per sampling day.

For both equations the total production corresponded to the sum of PΔt. For the species that presented lower densities and biomasses, production was estimated by summing increases from one sampling date to the other, as in Sardá et al. (1995).

Results

Macrozoobenthic community

Biodiversity, abundance, and biomass

Mondego estuary's south arm presented a total of 75 different taxa during the study period. In general, the number of species in the Zostera meadows was always higher than in the intermediate and eutrophied areas (Fig. 2A). Regarding biomass heterogeneity, higher values were found in the intermediate and eutrophied areas than in the Zostera meadows, contrary to expectations (Fig. 2B; further explanation in the Discussion).
Fig. 2A, B.

Spatial and temporal variation of biodiversity. A species richness, number of species; B heterogeneity of biomass, Shannon–Wiener index

Density and biomass parameters were studied for the main taxonomic groups: Annelida, Gastropoda, Bivalvia, Crustacea, and "others". This last group is composed of more rarefied populations, such as Pisces, Equinodermata, and Insecta. With regard to both parameters, the dominant group was Mollusca (Fig. 3), with the effectives of H. ulvae, Scrobicularia plana, and Cerastoderma edule as the most representative species (Table 1). Crustacea and Annelida were also important components of the communities, especially in the intermediate and eutrophied areas (Fig. 3), where Cyathura carinata and Hediste diversicolor were the main species (Table 1).
Fig. 3A–F.

Spatial and temporal logarithmic variation of density and biomass of the main taxonomic groups. A, D Zostera meadows; B, E intermediate area; C, F eutrophied area

Table 1.

Mean annual density (, individuals m−2) and biomass (B̄, grams AFDW m−2), with standard error in parentheses; annual production (P, grams AFDW m−2 year−1) with the relative percentage of total production (P%) and P/B̄ of the main species for Zostera meadows, intermediate, and eutrophied areas, in the south arm of Mondego estuary. Other polychaetes: Aphmaria romijni, Capitella capitata, Chaetozone setosa, Heteromastus filiformis, Glycera convoluta and Nephthys hombergi

Zostera meadows

Intermediate area

Eutrophied area

P

P%

P/

P

P%

P/

P

P%

P/

1993 (macroalgal bloom)

H.diversicolor

9 (±2)

0.1 (±0.04)

0.3

0

3.4

16 (±4)

0.3 (±0.1)

0.9

3

2.8

30 (±5)

0.4 (±0.11)

1.3

1

3.0

Other polychaetes

4,701 (±727)

1.2 (±0.11)

6.8

4

5.7

3,602 (±1,598)

0.4 (±0.06)

4.1

11

9.2

1,406 (±339)

0.3 (±0.04)

1.2

1

3.1

H. ulvae

55,741 (±5,053)

59.2 (±2.36)

122.6

78

2.1

29,757 (±8,505)

4.4 (±0.85)

17.6

48

4.1

47,551 (±9,719)

9.4 (±2.21)

45.3

49

4.8

L. littorea

25 (±4)

2.3 (±0.53)

5.6

4

2.4

2 (±1)

0.5 (±0.37)

1.0

3

1.9

C. edule

441 (±64)

5.3 (±0.88)

5.8

4

1.1

329 (±74)

0.1 (±0.03)

0.1

0

1.0

818 (±362)

0.2 (±0.15)

0.5

0

2.0

S. plana

535 (±131)

3.4 (±0.65)

2.6

2

0.8

921 (±182)

7.8 (±0.91)

7.4

20

0.9

2,795 (±317)

11.6 (±1.03)

16.1

18

1.4

C. carinata

30 (±6)

0.1 (±0.01)

0.2

0

2.5

99 (±24)

0.2 (±0.06)

0.5

1

2.5

1,948 (±207)

6.8 (±1.36)

23.2

25

3.4

C. maenas

69 (±10)

1.1 (±0.33)

5.9

4

5.2

33 (±7)

0.6 (±0.37)

2.8

8

4.6

61 (±25)

0.4 (±0.14)

2.2

2

5.0

Others

3,506 (±935)

1.1 (±0.15)

6.9

4

-

2,388 (±475)

1.0 (±0.26)

2.1

6

6,302 (±866)

0.6 (±0.13)

1.9

2

Total

65,057

73.8

156.7

2.1

37,148

15.4

36.6

2.4

60,911

29.5

91.7

3.3

1994 (almost no macroalgae)

H. diversicolor

32 (±8)

0.6 (±0.26)

2.8

1

4.3

108 (±20)

2.3 (±0.57)

8.1

13

3.5

57 (±12)

0.4 (±0.13)

2.2

4

5.1

Other polychaetes

839 (±217)

0.4 (±0.06)

1.6

1

4.5

938 (±454)

0.1 (±0.03)

0.6

1

4.8

189 (±44)

0.1 (±0.01)

0.5

1

2.8

H. ulvae

127,107 (±16,036)

74.1(±3.54)

204.3

91

2.8

68,338 (±8,607)

9.4 (±0.91)

38.2

62

4.1

13,966 (±5,306)

2.4 (±0.80)

10.9

22

4.5

L. littorea

15 (±3)

1.7 (±0.48)

2.7

1

1.6

C. edule

263 (±37)

3.2 (±1.04)

3.5

2

1.1

128 (±47)

1.2 (±0.97)

1.6

3

1.4

120 (±49)

0.0 (±0.00)

0.0

0

2.0

S. plana

1,725 (±236)

1.6 (±0.33)

3.0

1

1.8

1,849 (±406)

8.3 (±0.71)

9.2

15

1.1

2,218 (±606)

4.2 (±1.05)

4.9

10

1.2

C. carinata

180 (±49)

0.9 (±0.21)

2.7

1

2.9

259 (±47)

1.1 (±0.22)

2.6

4

2.3

2,082 (±365)

10.7 (±1.60)

30.0

61

2.8

C. maenas

84 (±25)

0.3 (±0.09)

1.9

1

7.4

37 (±13)

0.1 (±0.05)

0.4

0

3.6

20 (±9)

0.0 (±0.01)

0.1

0

7.2

Others

5,560 (±1,199)

0.6 (±0.16)

2.6

1

-

2,334 (±647)

0.3 (±0.17)

0.9

1

6,877 (±1,696)

0.1 (±0.04)

0.1

0

Total

135,805

83.4

225.2

2.7

73,990

22.8

61.6

2.7

25,528

15.4

48.7

3.2

1995 (some macroalgae)

H. diversicolor

21 (±8)

0.3 (±0.16)

1.2

1

4.0

85 (±22)

1.4 (±0.31)

3.6

7

2.5

79 (±22)

0.9 (±0.49)

1.6

4

1.8

Other polychates

1,276 (±497)

0.6 (±0.10)

3.0

2

4.8

1,288 (±614)

0.2 (±0.03)

1.5

3

7.6

684 (±202)

0.2 (±0.03)

4.1

10

15.1

H. ulvae

85,938 (±16,009)

82.1 (±8.97)

110.5

76

1.3

28,585 (±9,180)

3.9 (±0.96)

21.8

40

4.2

11,551 (±3,766)

3.6 (±0.77)

13.1

30

3.7

L. littorea

16 (±6)

1.2 (±0.55)

2.9

2

2.4

3 (±3)

0.4 (±0.41)

0.6

1

1.7

C. edule

969 (±359)

12.1 (±3.55)

14.7

10

1.2

66 (±19)

2.8 (±1.07)

2.2

4

0.9

4 (±4)

0.0 (±0.01)

0.0

0

1.9

S. plana

1,208 (±255)

2.5 (±0.39)

2.7

2

1.1

684 (±118)

16.2 (±2.46)

14.3

27

1.0

1,253 (±456)

5.1 (±1.91)

6.5

15

1.3

C. carinata

478 (±86)

1.5 (±0.16)

4.1

3

2.8

137 (±54)

0.5 (±0.15)

1.3

2

2.6

1,436 (±338)

6.8 (±1.20)

17.6

41

2.6

C. maenas

128 (±44)

0.6 (±0.29)

3.0

2

5.0

49 (±19)

0.9 (±0.57)

3.0

6

4.4

7 (±5)

0.0 (±0.01)

0.1

0

3.7

Others

8,595 (±2,100)

0.8 (±0.14)

2.7

2

2,774 (±1,270)

0.3 (±0.16)

5.5

10

7,205 (±3,494)

0.1 (±0.07)

0.3

1

Total

98,629

101.8

144.9

1.4

33,671

26.5

53.8

2.1

22,220

16.2

43.4

2.7

Secondary production estimates

The analysis was carried out for the following years: (a) 1993, when a macroalgal bloom occurred followed by a sudden crash of the algae in the eutrophied area; (b) 1994, with almost no macroalgae; and (c) 1995, with some macroalgae but no bloom.

Total production estimates were always higher at the Zostera meadows (Fig. 4; Table 1). In this area and in the intermediate area, secondary production was higher in 1994 than in 1993. In contrast, in the eutrophied area the trend was the opposite, with higher production in 1993 (Fig. 4; Table 1). In all areas, production decreased in 1995 compared to 1994 (Table 1).
Fig. 4.

Monthly variation of the secondary production of Zostera meadows, intermediate, and eutrophied areas

In the eutrophied area, production showed an initial overshoot, from February 1993 to May 1993, corresponding to a macroalgal bloom period, [mainly Enteromorpha spp., whose biomass reached more than 400 g AFDW m−2 in April 1993 (Lillebø et al. 1999; Martins et al. 2001)]. In fact, secondary production was substantially higher during the bloom period, representing about 65% of total production obtained in 1993. Accordingly, the production of the bloom period (4 months) was higher than the production of 1994 and 1995 (59 g AFDW m−2 per 4 months vs 48 and 43 g AFDW m−2 year−1).

Main species analysis

The dominant species usually accounted for more than 80% of the total secondary production for each site. For instance, in the Zostera meadows Hydrobia ulvae alone always represented more than 75% of the total yearly production (Table 1).

The intermediate area was characterised by the dominance of S. plana and H. ulvae during the whole study period (about 70% of yearly production). In 1993, the "other polychaete" production, namely, Aphmaria romijni, Capitella capitata, Chaetozone setosa, Heteromastus filiformis, Glycera convoluta, and Nephthys hombergi, was relatively high (11%), decreasing in the following periods. An important contribution of Hediste diversicolor was recognisable in 1994 (13%).

In the eutrophied area, the production was mainly due to Hydrobia ulvae and Cyathura carinata, which together represented about 75% of total yearly production (Table 1). The relative importance of these species changed during the study period. In 1993 H. ulvae's production was higher (49%) than C. carinata's (25%); in the following years C. carinata production became more important, accounting for 61% of total yearly production in 1994 and 41% in 1995 (Table 1). For almost all the main species the P/B̄'s were higher in the eutrophied area (Table 1).

To understand the ongoing changes in the Mondego estuary in a quantitative way, a mean intertidal production was estimated for the whole south arm. Through the comparison of aerial photographs, between 1986 and 1993–1995 (Fig. 1), it was concluded that the Zostera meadows declined from about 10% of the total intertidal area to 1%. Based on the production estimated for each site in 1994 (a year with almost no macroalgae), a hypothetical production for 1986 was also estimated. Results revealed that production may have been considerably higher in 1986, followed by 1993, 1994, and 1995 (Table 2). H. ulvae, C. carinata and S. plana were the most productive species (Table 2).
Table 2.

Mean macrobenthic community production estimates for the south arm of Mondego estuary, taking into account the area of each habitat relative to the whole intertidal area

South arm production

Total

H. ulvae

C. carinata

S. plana

(g m−2 year−1)

(%)

(%)

(%)

1986 (almost no macroalgae): 10% Z. noltii, 50% muddy, 40% sand-muddy

72.9

46

17

8

1993 (macroalgal bloom): 1% Z. noltii, 59% muddy, 40% sand-muddy

59.9

50

16

18

1994 (almost no macroalgae): 1% Z. noltii, 59% muddy, 40% sand-muddy

58.2

50

23

13

1995 (some macroalgae): 1% Z. noltii, 59% muddy, 40% sand-muddy

50.5

38

16

22

Trophic analysis

Detritivores were the trophic group that most contributed to secondary production. Spatially its importance becomes gradually greater from the Zostera meadows to the most eutrophied area. Herbivores showed the inverse pattern, prevailing in the Zostera meadows (Fig. 5). Still, results regarding Zostera meadows can be strongly influenced by H. ulvae, whose production was responsible for most of the total production and which belongs to both trophic groups. In fact, this species can behave as detritivore or herbivore depending on food resources (Cardoso et al. 2002). Over time, there were no pronounced changes in the trophic organisation regarding production of the different trophic groups. In the intermediate area the production was more distributed among all trophic groups (Fig. 5).
Fig. 5.

Relative contribution of the different trophic groups in the south arm of the Mondego estuary

Discussion and conclusions

Mondego estuary's secondary production

The Mondego estuary appears to be a very productive system when compared to other tidal flats (Table 3). Such comparisons may be biased because of differences in the sampling procedures (e.g. the schedule of sampling, mesh sieve used, etc.) and secondary production estimation methods used; nevertheless, the production results for the Mondego estuary are relatively high.
Table 3.

Macrobenthic community production and mean biomass estimates recorded on tidal flats, with reference to method used for the estimation of secondary production

Location

Production

Mean biomass

Production method used

Reference

(g AFDW m−2 year−1)

(g AFDW m−2)

Wadden Sea (Germany)

  Seagrass bed

48.2

30.2

Increment summation

Asmus and Asmus 1985

  Arenicola flat (sand)

50.2

27.6

  Nereis-Corophium belt

17.5

16.5

Bay of Cádiz (Spain)

  "Aquaculture bay"

13.3–27.3

10.6–25.3

Edgar (1990) (300-μm mesh sieve)

Arias and Drake 1994

Crib Point (Australia)

  Seagrass beds

15.3–106.0

3.7–43.1

Edgar (1990) (1-mm mesh sieve)

Edgar et al. 1994

  Unvegetated

5.3–82.5

3.7–32.7

Ria Formosa (Portugal)

  Seagrass bed

57.6

17.2

Increment summation and Sprung (1993) (500-μm mesh sieve)

Sprung 1994

  Sand flat

34.4

15.4

  Mud flat

71.5

25.1

Nauset Marsh (USA)

  Seagrass bed

23.0–139.1

27.4

Robertson (1979) (1.4-mm mesh sieve)

Heck et al. 1995

  Sand flat

5.5–9.1

1.0

  Mud flat

6.5–10.6

1.1

  Marsh pool

7.8–13.3

2.2

Sacca di Oro (Italy)

  Mud flat (clayey-silt)

49.8–75.0

46.5–73.5

Tumbiolo and Downing (1994) (500-μm mesh sieve)

Mistri et al. 2001

  Silty-sand flat

71.3

66.0

Mondego (Portugal)

  Seagrass bed

144.9–225.2

34.5–40.1

Increment summation and empirical methods (500-μm mesh sieve)

Present study

  Sand flat (eutrophied)

43.4–91.8

10.1–18.8

  Mud flat (intermediate)

36.6–61.6

16.0–23.3

Comparisons of the different sites clearly show that production in the intermediate and eutrophied areas when there is no macroalgal bloom is only 20–25% of the production in Zostera meadows, and about 50% when there is a bloom. This result is consistent with other studies comparing seagrass beds and non-vegetated environments (Table 3). Seagrass beds are characterised by high productivity and biodiversity (Edgar et al. 1994; Heck et al. 1995; Marques et al. 1997; Asmus and Asmus 2000). Sandy and muddy bare bottoms are less productive, as seen, for example, by Heck et al. (1995). Species seem to take advantage of the protection from predators, food resources (mainly the associated epiphytes), and availability of microhabitats supplied by the macrophytes (Edgar et al. 1994; Pardal et al. 2000, 2002; Cardoso et al. 2002). The intermediate and eutrophied areas do not offer the benefits associated with macrophyte presence, which may explain the lower production of those areas.

Macroalgal blooms: short-term and long-term effects on production

Production in the eutrophied area increased significantly during the bloom period (4 months), achieving higher values than in Zostera meadows in the same period (Fig. 4). In fact, during the bloom, abundance increased in the eutrophied area, probably owing to the adaptive strategies of opportunist species that recruit locally. We hypothesised that macroalgal blooms benefit the system while supplying increased habitat heterogeneity, shelter, and food resources (Raffaelli et al. 1998; Norkko et al. 2000), enhancing production. Nevertheless, soon after the macroalgal crash (post-bloom period), the system collapsed, resulting in a dramatic decrease in production. As in other studies, the fragile and unstable character of the macroalgal habitat (due to the dynamics of the macroalgae) was evidenced (Norkko and Bonsdorff 1996; Raffaelli et al. 1998; Lillebø et al. 1999; Norkko et al. 2000; Pardal et al. 2000, 2002).

The Zostera beds in the Mondego estuary have been gradually replaced by macroalgae over the last 20 years. Assuming that the level of production estimated for the different habitats studied would have been the same in 1986 (when Z. noltii occupied a larger area), a comparison between the mean intertidal productions of the study period showed a considerable decrease (about 30%). Accordingly, we may assume that the secondary production during macroalgal blooms cannot compensate for production losses due to seagrass decline (Table 2).

Main species

The dominance of H. ulvae was clear, with production reaching more than 200 g AFDW m−2 year−1 (90% of yearly production in Zostera meadows). The biomass dominance of this species in the Zostera meadows was reflected in a lower heterogeneity, despite higher species richness (Fig. 2). The same scenario was observed in other estuaries, where H. ulvae was the dominant species in density and production (Asmus and Asmus 1985). Reasons for the high productivity of H. ulvae are probably related to the location of the Mondego estuary, associated with warm temperatures, available food resources, and microhabitats. Together, these factors enable the optimisation of H. ulvae's reproduction, growth, and voltinism (Lillebø et al. 1999; Cardoso et al. 2002).

Annelid biomass and production were higher in the intermediate area. Hediste diversicolor production increased in the intermediate area in 1994 and 1995. This species is known for its tolerance to adverse conditions and as a potential recoloniser of disturbed habitats (Norkko and Bonsdorff 1996), so higher production estimates were expected. Nevertheless, our results are apparently consistent with the occurrence of lower densities found in drift algae obtained by Norkko et al. (2000).

In general, in the eutrophied area, all main species decreased in production in 1994 and 1995, with the exception of C. carinata. The population of C. carinata is more stable in the eutrophied area, probably because of to its preferences for sand-muddy bare bottoms (Ferreira 2001). Following the macroalgal crash, this species seemed to recover in the following year (1994), substantially increasing its production, probably owing to higher reproductive success in the absence of algae (Ferreira 2001).

Although infauna is recognised to be strongly affected by algal cover, showing high mortality (Norkko and Bonsdorff 1996; Norkko et al. 2000), in the present study, in the eutrophied area, the production of S. plana was higher in 1993. Nevertheless, such results are not very conclusive as deeper living species, such as S. plana, are forced to the surface as a result of the algal cover's stress, increasing their predation risk (Raffaelli et al. 1998) and feeding activity interference. On the other hand, macroalgae physically interfere with predator behaviour, as shown, for example, by Aarnio and Mattila (2000) and Cabral et al. (1999), making the possibility of lower predatory pressure from higher trophic levels also valid.

Trophic organisation production availability

Rooted-macrophyte systems can be seen as an integrated part of a grazing and nutrient-controlled stable environment. In comparison, systems in which the primary production is dominated by macroalgae and phytoplankton can be seen as unstable detritus/mineralisation environments, with a more dynamic turnover of oxygen and nutrients (Flindt et al. 1999). This appears to be reflected in the trophic organisation of the Mondego ecosystem. As in other tidal flats (Aníbal 1998), the dominant productive trophic group was the detritivore group, suggesting that a great part of the energy/biomass enters the system via the detritus food chain. Moreover, spatially, herbivores' importance gradually lessens from downstream to the inner parts of the estuary, following an increased eutrophication gradient and consequent progressive replacement of macrophyte producers by macroalgal producers. This difference could be even more pronounced if H. ulvae was considered strictly as herbivore, as in Asmus and Asmus (1985). Our results agree with those obtained in Ria Formosa, where in the eelgrass areas the main productive trophic groups were detritivores and herbivores (Boaventura et al. 1999), whereas the groups associated with the macroalgae were mainly detritivores (Aníbal 1998). No grazing effects on macroalgae with its potential control have been found for the Mondego estuary (Martins et al. 2002).

Implications for higher trophic levels

The Mondego estuary supports a well-developed fish community, including commercially important species such as Solea vulgaris, Platichthys flesus, Dicentrarchus labrax, and Sparus aurata, among others (Jorge et al. 2002), and constitutes an important habitat for a high diversity of birds, especially waders (Lopes et al. 2000; Múrias et al. 2002). Given the decrease in the macrobenthic production due to the associated effects of the macroalgal bloom, we might also reasonably expect a decline in its predators. By adding the effects of physical interference by macroalgae (as seen by Cabral et al. 1999) or post-crash anoxia, the prey/predator trophic transfer will be potentially reduced. Nevertheless, hypotheses regarding the upper trophic levels are still incipient. Additionally, as seen earlier, a great part of the production decrease was due to H. ulvae dynamics, and although some studies reveal this species as an important prey item in predators' diets (Raffaelli and Milne 1987; Múrias et al. 2002), others categorised it as a low-quality food item (Aarnio and Mattila 2000). Its "incorporation" in the secondary consumers' tissues, in the Mondego estuary, is yet unknown.

Conclusions

The present study reinforces the idea that the decrease of the macrophyte beds necessarily implies the decrease of estuarine production. Although total production was temporarily enhanced by the occurrence of the macroalgal bloom, the following macroalgal crash caused a drastic decrease in production, associated with the decrease in density and biomass.

From the management point of view important questions need to be addressed: are estuaries able to support repeated discharges of nutrient-enriched waters that induce the proliferation of green macroalgae? In the present study, it is clear that the macrophyte assemblages suffered as a consequence of eutrophication and increased macroalgal biomass. There was an evident decrease in secondary production; furthermore, the nursery grounds of economically important species are potentially affected (Asmus and Asmus 2000; Levin et al. 2001), as well as the nutrient control at the system level (Flindt et al. 1999). Ultimate consequences affect the entire food web itself (Raffaelli 1999; Levin et al. 2001), either by a diminished number of chains or by decreased energy/biomass flow through the chain itself, as a consequence of qualitative and quantitative changes in secondary producers. So, we can assume that in coastal waters the complete replacement of macrophyte beds by macroalgae might seriously compromise the integrity of the whole ecosystem.

Copyright information

© Springer-Verlag 2003