Diversity and Abundance Patterns of Benthic Invertebrate Assemblages on Intertidal Estuarine Seagrass Beds in Aveiro (Portugal) ^†

Abstract

Seagrass meadows are productive ecosystems and many animal species are dependent on them, including a wide diversity of invertebrates. This study aims to explore spatial diversity patterns of benthic invertebrates associated with Zostera noltei. Three areas with Z. noltei meadows were sampled along the Mira Channel (Ria de Aveiro). At each area, two sites were selected and four cores were taken at each site. Fauna was sorted, counted, and identified to the lowest taxonomical level. Results showed significant differences in the number of taxa among meadows. It was also observed that some taxa presented differences in the abundance among meadows.

1. Introduction

Seagrass meadows form one of the most productive ecosystems in the world [1] and fulfill key functions on coastal ecosystems [2]. Seagrasses are ecosystem engineers, due to their ability to modify environmental and biological drivers and enhance the local biodiversity [3,4]. Seagrass meadows are distributed in tropical and temperate waters around the globe, on marine and estuarine rocky, sandy and muddy substrates [2,3]. However, during the last decades, they became one of the most endangered ecosystems on coastal systems [5]. This may be the result of their high sensibility to disturbances—such as water pollution, sediment resuspension, eutrophication, invasive species and even exploitation of animal species—that, for instance, increases the dispersion of their seeds [2,5,6].

Seagrass meadows establish complex habitats that shelter many marine organisms such as epiphytic algae, fishes, and vertebrates, as well as a great diversity of invertebrates that can be associated with their leaves (epifauna) or their rhizomes in the sediment (infauna) [2]. Therefore, meadows may be tenfold more diverse than other similar systems less structurally complex [2]. This enhanced diversity plays a critical role in seagrass meadows. Epifaunal animals, such as amphipods and gastropods, feed upon epiphytic organisms that cover seagrass leaves, cleaning them and allowing to improve light incidence, oxygen and carbon dioxide intake, while infaunal organisms increase sediment oxygenation [7], and filter-feeders, such as annelids and bivalves, reduce water turbidity. Moreover, many of these animals serve as food for higher trophic levels [2].

The seagrass Zostera noltei Hornemann, 1832 creates extensive meadows along the Atlantic intertidal areas of the Iberian Peninsula, often dominant, especially in the Ria de Aveiro lagoon. The aim of this study is to explore changes in invertebrate diversity associated with Z. noltei meadows along an estuarine gradient.

2. Materials and Methods

2.1. Study Area

This study was carried out during July 2019 throughout the Mira channel of the Ria the Aveiro lagoon, a waterbody that runs parallel to the coastline for about 25 km. Freshwater supply, mainly on the upper area of the channel creates a salinity gradient, with low levels of salinity in the inner part and higher values at the mouth [8]. As a result of increasing anthropogenic influence, the channel has been transformed over the last decades becoming a more heterogeneous environment [9]. Several human activities, such as dredging, shellfish and bait harvesting, recreational navigation and oyster farming, have been advanced everywhere, but particularly along the margins.

2.2. Sampling and Sample Processing

To study the invertebrate diversity associated with Z. noltei meadows in the Ria de Aveiro lagoon, three areas were selected: (A, B and C, Figure 1). Meadow A was located upwards of the waterbody and meadow C was positioned downwards, while meadow B layed in the middle close to the oyster trestles of a commercial farm and at approximately the same distance from the other sites (Figure 1). The coverage of Z. noltei was 68.37 ± 6.037%, 90.12 ± 6.037% and 70.00 ± 10.32% in A, B and C meadows, respectively.

Two sites, separated by tens of meters, were selected at each meadow; and at each site, four sediment replicates were randomly collected. Samples were retrieved with a 0.02 m² corer to a depth of 30 cm, and subsequently sieved in situ through a 0.5 mm-mesh bag. Macrofauna retained was preserved in a 4% neutralized formaldehyde solution with Rose Bengal.

Fauna samples were sorted, identified to the lowest possible taxonomical level and counted, using a stereo microscope (Nikon SMZ800) and a microscope (Leica DM 2500 LED) for smaller organisms.

2.3. Data Analyses

Differences among meadows on the total number of individuals (N), Shannon index (H’) and the number of taxa (S) of macrofauna were explored by analysis of variance (ANOVA). These analyses considered two factors: Meadow (Me) as an orthogonal fixed factor with three levels (A, B, C) and Site (Si) as a random factor nested in Me with two levels (1, 2). Homogeneity of variance was tested by a Cochran’s C test and data were log (x + 1) transformed when necessary. Whenever ANOVA showed significant differences (p < 0.05), a post hoc Student–Newman–Keuls (SNK) test was performed for a posteriori multiple comparisons. In order to explore differences among meadows in the abundance of dominant taxa, ANOVA analysis was used following the same design previously described.

3. Results

A total of 5560 individuals represented by 41 different taxa were sampled. Oligochaetes constitute 44% of this abundance followed by Peringia ulvae (Pennant, 1777) (20%). ANOVA results (

Table 1. ANOVA results for the Number of Taxa (S), Abundance (N) and Shannon Diversity Index (H′).

		S			N			H′
Source	df	MS	F	P	MS	F	P	MS	F	P
Meadow	2	103.625	11.840	0.038	2.771	4.320	0.131	0.584	2.910	0.198
Site	3	8.750	1.450	0.261	0.641	4.460	0.017	0.201	2.480	0.094
Residuals	18	6.028			0.144			0.081
Total	23

) showed non-significant differences among meadows on the values of H’ and N but, significant variability among sites was detected for N (Table 1).

Values of S showed significant differences among meadows (Table 1) with significantly higher values in meadow A than in meadows B and C (Figure 2).

The most abundant taxa were: Cyanthura carinata (Krøyer, 1847), Hediste diversicolor (O.F. Müller, 1776), nematodes, Notomastus latricerus Sars, 1851, oligochaetes, P. ulvae and Scrobicularia plana (da Costa, 1778). However, ANOVA results only showed significant differences among meadows for H. diversicolor, nematodes and P. ulvae (

Table 2. ANOVA results for the more relevant species. Numbers in bold indicate significant differences.

		P. ulvae			H. diversicolor			Nematodes
Source	df	MS	F	P	MS	F	P	MS	F	P
Meadow	2	3.157	18.690	0.0202	2399.542	19.800	0.0187	15.551	49.280	0.0051
Site	3	0.169	0.520	0.6728	121.208	3.170	0.0495	0.316	0.750	0.5370
Residuals	18	0.323			38.236			0.421
Total	23

).

Moreover, post hoc analysis showed different patterns of abundance for these three species: P. ulvae had a lower abundance in meadow B, higher in meadow C and intermedial in meadow A, while H. diversicolor increased its abundance significantly from meadow A to meadow B and C (Figure 3). Finally, nematodes were significantly more abundant in meadow A (Figure 3).

4. Discussion

A total of 41 taxa was found in our study, lower than the invertebrate diversity associated with Z. noltei meadows in a Mediterranean intertidal mud flat. These authors found a total of 102 taxa with S values between 13 and 33 taxa per site and H values between 1.86 and 2.83 [10]. In general, these values were higher than those reported in our study and our highest values were nearly the lowest values found by [10]. This difference may be explained because they sampled a higher area (19 localities). The positive relationship between the area and the number of species may explain the lower value found in our study [10]. Zostera noltei populations on the Atlantic shores are frequently found in transitional systems such as our study area. In this scenario, when we consider other studies conducted on the European Atlantic shores, we find contrasting results. In a study conducted in Belgium with a similar sampling effort, [4] found lower values of S (22 taxa) and H (1.24) than in this study. On the other end, many studies conducted in Arcanchon bay (France) found similar values for the median number of taxa [11,12] and, in other cases, total or mean values of S were higher than in our study [3,13]. These differences may be the result of different sampling efforts, as aforementioned. Another plausible explanation could be that our Z. noltei meadows show a lower structural heterogeneity and thus a lower S than previous studies [3,13]. Finally, many dominant species in our study were also found in other Atlantic localities. However, their abundance and distribution patterns were very variable. This highlights the importance of investigating the community structure in stressful environments, particularly in estuarine seagrass meadows.