ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on March 2, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(3):446-452; doi:10.1093/icesjms/fsm018
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Quantitative assessment of the area of shallow habitat for fish on the Swedish west coast
Department of Marine Ecology, Göteborg University, Kristineberg Marine Research Station, 450 34 Fiskebäckskil, Sweden
Correspondence to J. Stål: tel: +46 523 18500; fax: +46 523 18502; e-mail: johan.stal{at}kmf.gu.se
Stål, J., and Pihl, L. 2007. Quantitative assessment of the area of shallow habitat for fish on the Swedish west coast. – ICES Journal of Marine Science, 64: 446–452.Much effort has been focused recently on juvenile and adult fish habitat use in shallow coastal areas. However, to understand fully the importance of such habitats for fish production it is necessary also to quantify the area of existing habitat types. We inventory and quantify the area of major habitat types in a 1000 km2 area of the Swedish west coast, on a scale appropriate for coastal-zone management. An echosounder and GPS-transmitter mounted on a small boat were used to estimate the distribution of habitat along transects in seven regions of differing coastal morphology. The signal from the echosounder separated major habitat types, and recordings were verified by video documentation and visually. The information was used with GIS-software to estimate the quantitative extent of bottom habitats at depth ranges of 0–3, 3–6, and 6–10 m. Of the major habitat types, soft substrata dominated all except one region, and increased in size with depth in all regions. There were rocky substrata in all regions, but as steep rock walls in the north and more gently sloping substrata with pebbles and boulders in the south. Approximately half the rocky habitat was in the shallowest depth range. Seagrass meadows on soft substrata were mainly in the shallow protected archipelago of the central coast.
Keywords: areas, fish habitat, fisheries management, habitat distribution, Swedish west coast
Received 29 September 2006; accepted 13 January 2007; advance access publication 2 March 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsData collectionResultsDiscussionReferences |
|---|
Shallow habitats in the coastal zone constitute just a fraction of the world's ocean floor, but they are important for the production of many marine fish species and may act as bottlenecks for population sizes of some species. The coastal zone has high primary and secondary production and supports a great diversity and often fast growth of many fish (Beck et al., 2001). On the Swedish west coast, some 60 fish species regularly utilize shallow (0–10 m) habitats of the coastal zone and, of these, approximately 30% are of commercial value (Pihl and Wennhage, 2002).
There already exists much information on small-scale variations (m2) in habitat and structure of faunal assemblages, as well as on the utilization of shallow coastal habitats by fish during various stages of their life history (Pihl and Rosenberg, 1982; Baden and Pihl, 1984; Möller et al., 1985; Pihl, 1989; Pihl and Ulmestrand, 1993; Pihl and Wennhage, 2002). However, there has to date been no attempt to quantify on a larger scale (km2) the area of fish habitat along Sweden's west coast <10 m. Therefore, we carried out a large-scale quantitative assessment of fish habitat, in the hope that this new information will help coastal-zone managers evaluate and value the importance of shallow fish habitats along the Swedish west coast. Ecological studies of the Swedish west coast have to date generally divided the coastal zone into three major habitat types: non-vegetated soft substrata (SB), seagrass (Zostera marina) meadows (Z), and vegetated rocky substrata (RB). The larger scale level of resolution is also appropriate for some of the management purposes discussed in this study. RB can, however, be further divided according to the dominating alga species and level of exposure, and soft substrata according to sediment grain size.
Attempts to map habitats in the coastal zone have used various multi-beam techniques that generally operate better at depths >2 m (Komatsu et al., 2003; Bekkby and Rosenberg, 2006), satellite remote sensing that only detects RB and seagrass meadows (Dahdouh-Guebas et al., 1999), and aerial photographs that can be used to detect habitats at smaller scales, e.g. specific habitats in certain locations (Kirkman, 1996; Pihl et al., 1999). Habitat inventories along the Swedish west coast have to date used an aquascope from a boat to quantify the extent of seagrass meadows (Baden et al., 2003). Here, we further develop this technique by adding GPS equipment and an echosounder. Our aim was to perform a quantitative assessment of area and depth distributions (within 0–10 m) of major fish habitat types in a 1000 km2 area of the Swedish west coast. We also subdivided the target area into regions, habitats, and depth ranges on a scale appropriate for use in coastal-zone management.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsData collectionResultsDiscussionReferences |
|---|
The Swedish west coast, including the Skagerrak and the Kattegat, extends from the Norwegian border (11°E 59°N) in the north to Öresund (12°E 55°N) in the south. The Skagerrak coast contains an archipelago with a complex structure of islands and fjords of different size, and a shoreline with a mixture of soft and rocky substrata. The Kattegat has an open, exposed coastline with an archipelago only in the north. The tidal amplitude on the Swedish west coast is around 0.2 m (Svansson, 1975), so exchange of surface water in the study area is ensured largely by diffuse wind-driven processes. As the characteristics of the shore vary along the coast, the study area was subdivided into seven regions (Figure 1a). Division was made according to drainage area, as suggested by the Swedish Hydrological and Metrological Institute, and regions 1–4 also followed the division used by Pihl et al. (1999). The distribution and structure of the vegetation and fish assemblage vary with depth and exposure (Pihl and Wennhage, 2002), so measurements were made within three depth ranges (0–3, 3–6, and 6–10 m) in accord with previous work. Division of the coast into three depth ranges and seven regions is considered as useful to support management of the coastal zone of Sweden's west coast.
|
|
|
Data collection |
|---|
|
TopIntroductionMaterial and methodsData collectionResultsDiscussionReferences |
|---|
The total area of seabed within the depth ranges of 0–3, 3–6 and 6–10 m was estimated for the seven regions of the coast using GIS software (ArcView, version 3.2) and digitized sea charts. This information was subsequently used in combination with estimates of habitat distribution to evaluate the area of major fish habitat by depth range for the whole region.
The distribution of different habitats within the three depth ranges and seven regions was estimated from random field sampling. A grid was superimposed on sea charts to produce rectangles of 0.5 x 0.5 nautical miles. Nine rectangles were selected randomly to represent each region (Figure 1a), and those without shoreline were omitted and replaced. Region 3 had a broad archipelago and was divided into an inner and an outer part, giving a total of 18 rectangles sampled there. The 72 selected rectangles together included 5.8% of the total area of seabed in the region investigated. Within each rectangle, between 4 and 19 transects were planned, representing the entire shore down to a depth of 10 m; in all, the results from 2130 transects were analysed. The total transect length within a rectangle varied between 230 m and 18 400 m. Transects were perpendicular to the shore, 0.5 cables (92.6 m) apart, and habitat type, position, and depth distribution were recorded along each (Figure 1b).
Measurements were conducted from a small boat equipped with a GPS-transmitter (Garmin eTrex venture), an echosounder (Simrad model CE 32, frequency 200 kHz, beam width 15°), and an aquascope (a submersible cylinder with a bottom made of glass). Average survey speed was 2–3 knots. Shallower than 0.2 m, measurements from the boat were not possible, so visual determination of habitat type and distance to shore had to be carried out. The signal from the continuously recording echosounder was used to separate major habitat types along transects, and the position and depth was registered at each habitat change. Habitat types were related to specific echosounder signals prior to the investigation. By comparing the trace pattern on the display of the echosounder with visual observations and video images of different habitat types, the trace pattern for each habitat on the display was determined. The signal was verified by direct observation with the aquascope whenever possible. When such visual observation was not possible (at depths >4 m) bottom samples were taken with a small grab, allowing us to distinguish between soft and rocky substrata. Habitats separated by this method were SB seabed, vegetated soft sediment seabed (i.e. seagrass, Zostera marina), RB, and beds of blue mussels. It was not possible to distinguish between different species of alga in the rocky habitat or between grain sizes of soft-sediment habitats.
Computation of habitat distribution and area
The distribution of the different habitats along transects was calculated from the positions and depths registered at each habitat transition using Pythagoras' theorem. The horizontal difference in position and the vertical difference in depth served as sides of a triangle in which habitat length was the hypotenuse. Thus, we were able to correct for differences in the vertical gradient of transects, a crucial correction when calculating, for example, habitat distributions along transects along almost vertical rock walls. Each transect was divided into several sections according to habitat type and depth range. The length of each habitat in each transect was then calculated, and these lengths were summarized by rectangle. Finally, the sum of habitat lengths in each rectangle was converted among rectangles into proportions (percentages) to represent a region. The proportions were then used to transform habitat length into area, by multiplying each habitat proportion by the bottom area (from the GIS-computations) within the three depth ranges for each region.
|
Results |
|---|
|
TopIntroductionMaterial and methodsData collectionResultsDiscussionReferences |
|---|
Area of bottom habitat
Our investigation included seven geographical regions together covering 940 km2 of shallow coastal habitat along 400 km of Swedish west coast. In each region, we mapped 4.7–9.2% of the total area of seabed habitat between the shore (0 m) and a depth of 10 m (Table 1). The sampling precision (standard error/mean) when estimating habitat distribution within regions was similar for RB and SB, varying between 20 and 46%, but for seagrass, it was less precise, between 40 and 100% (Table 2). The area of bottom habitat within the three depth ranges is summarized for each region in Table 1. Regions 3, 4, 5, and 7 had the largest areas of bottom habitat, a total of 150–176 km2. For regions 1–5 the largest areas of bottom habitat were in the shallowest depth range (0–3 m), but in regions 6 and 7 the largest areas were in the deepest range (6–10 m).
|
|
Distribution of habitats
SB were overall the dominant habitat, accounting for 53% of the area <10 m. Except in region 3, SB dominated, constituting 37–77% of seabed area in the seven regions (Figure 2); its greatest contributions were in regions 4 and 7. RB contributed 13–49% of the area, and was especially prevalent in regions 2 and 6. RB could be separated into two main habitat types according to their geographical appearance: the RB of regions 1–4 were steeply sloping, and that of regions 6 and 7 were mainly less steep and covered with pebbles and boulders. In region 5, both these habitat types were found. The distribution of seagrass beds varied geographically, but beds were found mainly in regions 3, 4, and 5. In region 3, seagrass was the most common habitat, accounting for 48% of the area of seabed. Seagrass was scarce in regions 6 and 7, in each case <0.5% of seabed. Blue mussel beds were scarce in all regions, but beds of this habitat-forming species covered 0.8 and 1.7% of the seabed in regions 1 and 3, respectively.
|
The percentage contribution of SB increased with depth in all regions, except in region 6, where there were only small differences in coverage by depth (Figure 3). Almost 50% of the SB in that region was within the deepest depth range (6–10 m); the rest was equally distributed between the other two ranges. There was a similar distribution pattern of SB between regions in the two depth ranges shallower than 6 m, in contrast to the situation in the deeper range (6–10 m), the contribution of SB being large (46–89%) throughout. RB's percentage contribution varied between 15 and 54% in the three depth ranges in all seven regions (Figure 3). RB dominated regions 2 and 6 at all depths, and the distribution pattern was similar between regions for the three depth ranges. Overall, half the RB habitat was in the shallowest depth range (0–3 m). Seagrass was more prevalent shallower than 6 m, 86% of seagrass meadow being in the 0–3 m depth range, and most of the remaining meadow between 3 and 6 m (just 0.5% of seagrass meadow was deeper than 6 m). In regions 3 and 5, seagrass dominated shallower than 6 m.
Region 3 contains a wide archipelago, so this part of the coast was divided into an inner and an outer area (Figure 4). In its inner area, SB was mainly between 6 and 10 m deep, but RB was limited (<10%) at all depths. Seagrass dominated shallower than 6 m. In the outer area of region 3, SB contributed 24–40% of bottom habitat, and its distribution increased with depth. RB's contribution also increased with depth, from around 5% to 60%, whereas seagrass was the dominating habitat (70%) shallower than 3 m.
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsData collectionResultsDiscussionReferences |
|---|
To our knowledge, this is the first quantitative attempt to estimate the area of shallow fish habitat at a large spatial scale and covering the entire Swedish west coast. In combination with biological data, the results can be used to manage habitats through protection and conservation, and by aiding the development of sustainable fisheries.
Our method quantifies habitats in a large area with limited effort at low cost, and the method is accurate at evaluating major habitat types shallower than 10 m. As Sweden's west coast tidal amplitude is small (<0.2 m), variations in water level can be considered as insignificant and not requiring correction when recording water depth. Recordings were only made during calm weather (because the observations were from a small boat), further reducing the influence of waves or swell on depth recordings.
On some occasions, e.g. when there were patches of macroalgae (and hence RB) in seagrass beds, or in areas where sediment grain size was a mixture of small (sand and gravel) and large (pebbles and stones) sizes, determination of the habitat (SB or RB) may have been somewhat arbitrary. These problems were often resolved visually or by sampling the sediment formally, but in areas with long transects, there may still have been some bias in our results.
The general appropriateness of our assessment is dependent on its resolution, the degree of representation of sampled habitats, and the precision of sampling. Overall, 5.8% of the total seabed area of the coast shallower than 10 m was sampled, and the precision (the ratio between the standard error and the mean) varied between 20 and 46% for RB and SB habitats, and between 40 and 100% for seagrass habitat in the seven regions. The poorer precision for seagrass habitat can be attributed to the great natural variation of this habitat (i.e. several squares where seagrass was absent), resulting in more variation in area between rectangles. Habitat types exhibited scale-dependent variation in structure, species composition among vegetation, and sediment grain size, all of which may affect the fauna utilizing them. Such detailed information already exists (Pihl and Rosenberg, 1982; Möller et al., 1985; Möller, 1986; Pihl, 1986; Wennhage and Pihl, 1994; Baden and Boström, 2001; Pihl and Wennhage, 2002; Baden et al., 2003; Pihl et al., 2006) for the study area, but mainly in small-scale resolution (from cm2 to m2). Such resolution is often too tight for management purposes and is also impossible to include when quantifying a large coastal area. However, we consider that our resolution, spatial representation, and sampling precision were sufficiently good for our type of investigation (Veijola et al., 1996; Fraser and Williams, 1997; Drummond and Connell, 2005), so we believe that the results have value in coastal zone management.
SB were present in large areas along the coast and dominated the habitat in all except one region. Moreover, half of the total SB was in water 6–10 m deep. The biomass of macrobenthic fauna living in soft sediments is high at such depths (Stål et al., 2007), particularly in exposed areas where biomass increases with depth down to 10 m (Möller, 1986). However, in less exposed areas, production of benthic fauna is also high in water depth <1 m above soft sediments (Möller et al., 1985). Although this habitat was treated as a single bottom type in this study, the difference in exposure levels between regions may result in there being a variety of grain sizes between areas. Such difference in grain size and level of exposure may have important implications for the marine life inhabiting these habitats, because sediment structure influences both fish assemblages and the food these fish take (Möller, 1986; Pihl, 1986).
Generally, there was little variation in the distribution of rocky substrata between regions, with an archipelago in the north and the open coast in the south. However, the characteristics of the rocky habitat differed between the two geographical areas. In the archipelago of regions 1–4 in the north, the rocky habitat was characterized by steep vegetated rock walls with an abundance and large biomass of macroalgae, whereas in regions 5–7 in the south, it consisted mainly of pebbles and stones, with a shallow slope and few macroalgae. The rocky habitats in the south are directly exposed to waves from the Kattegat, which in combination with the gently sloping substratum may explain the absence of large macroalgae, such as Laminariales and Fucales (Eriksson and Bergstrom, 2005). Therefore, although both habitat types are classified as rocky, their function may vary through different regimes of exposure, vegetation assemblages, and slope. Stål et al. (2007) found significantly greater values of motile epibenthic macrofauna, in terms of number of species, abundance, and biomass, in the 0–3 m depth range than in deeper rocky substrata along the Swedish Skagerrak coast. Therefore, considering that some half the area of rocky substrata were within that depth range and that the motile epibenthic fauna constitutes an important food resource for coastal fish assemblages (Stål et al., 2007), the value of shallow rocky habitats is obvious.
The difference in exposure levels between regions in the north and the south may also be key in governing the distribution of seagrass, which was virtually absent in the exposed regions 6 and 7. Seagrass cover was greatest in the shallow depth ranges of regions 3, 4, and 5, and these three regions have an archipelago with lower levels of exposure, creating optimum conditions for seagrass (Baden and Boström, 2001). Further, there was a difference in the area of seagrass habitat within region 3's two parts. The area of seagrass meadows was similarly high in the 0–3 and 3–6 m depth ranges in region 3's inner area, but in its outer area the distribution of seagrass was low in the 3–6 m depth range. In the outer part of region 3, the area of rocky habitat in the 3–6 m depth range was 5–6 times greater than in the inner area. Previous studies have reported that seagrass meadows constitute important nursery areas for several species of fish (Pihl et al., 2006), emphasizing the importance to fish of the protected shallow areas of the archipelago. Juvenile fish of several species tend to use Sweden's inshore seagrass habitat as nursery areas.
To summarize, the greatest area of SB was in the 6–10 m depth range, whereas RB and seagrass were proportionally more dominant in the 0–3 m depth range. Further, seagrass beds were most abundant in the protected archipelago in the central regions of the coast. The availability and good quality of benthic macrofauna, in combination with the large area of suitable habitat, makes these areas extremely valuable as fish habitat. However, to understand better the importance of shallow coastal habitats to fish, knowledge of the degree of habitat dependence of the various fish species is needed. Strongly habitat-specific species will be more dependent on their preferred habitat than species that make use of various habitats, so a strong habitat affinity may imply that the habitat could be acting as a bottleneck for certain such fish species. For example, plaice (Pleuronectes platessa) and Atlantic cod (Gadus morhua) are common and economically important fish species along Sweden's west coast. Plaice are very habitat-dependent (Pihl and Rosenberg, 1982; Pihl, 1989), so if that preferred habitat is altered (e.g. by eutrophication) or destroyed (e.g. by physical alteration of the habitat), plaice will not be able to complete their life cycle, with potentially detrimental effects for the stock. Although Atlantic cod are less habitat-specific and may use several alternative habitats as nursery, RB and seagrass meadows offer the species a high-quality environment allowing for high densities, fast growth, and good survival (Keats et al., 1987; Gotceitas et al., 1995; Fjosne and Gjosaeter, 1996; Tupper and Boutilier, 1997; Pihl et al., 2006). Therefore, there is potential for the local cod stock to be negatively impacted if such habitats are altered or diminished. In conclusion, therefore, the results from this study provide crucial information to underpin management decisions on how and where best to protect the shallow coastal habitats of fish along the Swedish west coast
|
Acknowledgements |
|---|
We acknowledge funding of this work by the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning, and thank Therese Jansson, Maria Bodin, and Florin Floruta for valuable assistance during field measurements. We also thank Per Nilsson for kindly lending us equipment and for GIS support, two anonymous referees for valuable comments on an early draft of the manuscript, and the editor for grammatical improvement.
|
References |
|---|
|
TopIntroductionMaterial and methodsData collectionResultsDiscussionReferences |
|---|
-
Baden S. and Boström C. (2001) In Reise K. (Ed.). The leaf canopy of seagrass beds: fauna community structure and functional in a salinity gradient along the Swedish coast. In Ecological Comparisons of Sedimentary Shores: Ecological Studies. Springer, Dordrecht 151:213–236.
Baden S., Gullstrom M., Lunden B., Pihl L., Rosenberg R. (2003) Vanishing seagrass (Zostera marina, L.) in Swedish coastal waters. Ambio 32:374–377.[Medline]
Baden S. P. and Pihl L. (1984) Abundance, biomass and production of mobile epibenthic fauna in Zostera marina (L.) meadows, western Sweden. Ophelia 23:65–90.[Medline]
Beck M. W., Heck J. K. L., Able K. W., Childers D. L., Eggleston D. B., Gillanders B. M., Halpern B., et al. (2001) The identification, conservation, and management of estuarine and marine nurseries for fish and invertebrates. Bioscience 51:633–641.[CrossRef][Web of Science]
Bekkby T. and Rosenberg R. (2006) Marine habitaters utbredelse- terrengmodellering i Gullmarsfjorden. Länsstyrelsen i Västra Götalands län, Vattenvårdsenheten. Rapport 2006 7: (in Swedish).
Dahdouh-Guebas F., Coppejans E., van Speybroeck D. (1999) Remote sensing and zonation of seagrass and algae along the Kenyan coast. Hydrobiologia 400:63–73.[CrossRef][Web of Science]
Drummond S. P. and Connell S. D. (2005) Quantifying percentage cover of subtidal organisms on rocky coasts: a comparison of the costs and benefits of standard methods. Marine and Freshwater Research 56:865–876.[CrossRef][Web of Science]
Eriksson B. K. and Bergstrom L. (2005) Local distribution patterns of macroalgae in relation to environmental variables in the northern Baltic proper. Estuarine. Coastal and Shelf Science 62:109–117.[CrossRef]
Fjosne K. and Gjosaeter J. (1996) Dietary composition and the potential of food competition between 0-group cod (Gadus morhua L.) and some other fish species in the littoral zone. ICES Journal of Marine Science 53:757–770.
Fraser B. G. and Williams D. D. (1997) Accuracy and precision in sampling hyporheic fauna. Canadian Journal of Fisheries and Aquatic Sciences 54:1135–1141.
Gotceitas V., Fraser S., Brown J. A. (1995) Habitat use by juvenile Atlantic cod (Gadus morhua) in the presence of an actively foraging and non-foraging predator. Marine Biology 123:421–430.[CrossRef]
Keats D. W., Steele D. H., South G. R. (1987) The role of fleshy macroalgae in the ecology of juvenile cod (Gadus morhua L.) in inshore waters off eastern Newfoundland. Canadian Journal of Zoology 65:49–53.
Kirkman H. (1996) Baseline and monitoring methods for seagrass meadows. Journal of Environmental Management 47:191–201.[CrossRef][Web of Science]
Komatsu T., Igarashi C., Tatsukawa K., Sultana S., Matsuoka Y., Harada S. (2003) Use of multi-beam sonar to map seagrass beds in Otsuchi Bay on the Sanriku Coast of Japan. Aquatic Living Resources 16:223–230.[CrossRef][Web of Science]
Möller P. (1986) Physical factors and biological interactions regulating infauna in shallow boreal areas. Marine Ecology Progress Series 30:33–47.[Web of Science]
Möller P., Pihl L., Rosenberg R. (1985) Benthic faunal energy flow and biological interaction in some shallow bottom habitats. Marine Ecology Progress Series 27:109–121.[Web of Science]
Pihl L. (1986) Exposure, vegetation and sediment as primary factors for mobile epibenthic faunal community structure and production in shallow marine soft bottom areas. Netherlands Journal of Sea Research 20:75–83.
Pihl L. (1989) Abundance, biomass and production of juvenile flatfish in southeastern Kattegat. Netherlands Journal of Sea Research 24:69–81.
Pihl L., Baden S., Kautsky N., Ronnback P., Soderqvist T., Troell M., Wennhage H. (2006) Shift in fish assemblage structure due to loss of seagrass Zostera marina habitats in Sweden. Estuarine, Coastal and Shelf Science 67:123–132.[CrossRef]
Pihl L. and Rosenberg R. (1982) Production, abundance and biomass of mobile epibenthic marine fauna in shallow waters, western Sweden. Journal of Experimental Marine Biology and Ecology 57:273–301.[CrossRef][Web of Science]
Pihl L., Svenson A., Moksnes P. O., Wennhage H. (1999) Distribution of green algal mats throughout shallow soft bottoms of the Swedish Skagerrak archipelago in relation to nutrient sources and wave exposure. Journal of Sea Research 41:281–294.[CrossRef]
Pihl L. and Ulmestrand M. (1993) Migration pattern of juvenile cod (Gadus morhua) on the Swedish west coast. ICES Journal of Marine Science 50:63–70.
Pihl L. and Wennhage H. (2002) Structure and diversity of fish assemblages on rocky and soft bottom shores on the Swedish west coast. Journal of Fish Biology 61:(Suppl. A), 148–166.[Web of Science]
Stål J., Pihl L., Wennhage H. (2007) Food utilisation by coastal fish assemblages in rocky and soft bottoms on the Swedish west coast: inference for identification of essential fish habitats. Estuarine, Coastal and Shelf Science 71:593–607.[CrossRef]
Svansson A. (1975) Physical and Chemical Oceanography of the Skagerrak and the Kattegat. (Institute of Marine Research, Lysekil)88.
Tupper M. and Boutilier R.G. (1997) Effects of habitat on settlement, growth, and postsettlement survival of Atlantic cod (Gadus morhua). Canadian Journal of Fisheries and Aquatic Sciences 52:1834–1841.
Veijola H., Merilainen J. J., Marttila V. (1996) Sample size in the monitoring of benthic macrofauna in the profundal of lakes: evaluation of the precision of estimates. Hydrobiologia 322:301–315.[CrossRef][Web of Science]
Wennhage H. and Pihl L. (1994) Substratum selection by juvenile plaice (Pleuronectes platessa L) – impact of benthic microalgae and filamentous macroalgae. Netherlands Journal of Sea Research 32:343–351.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||