to environmental gradients - variability at temporal and different - PowerPoint PPT Presentation

Coastal fish abundance in relation to environmental gradients - variability at temporal and different spatial scales L Bergström*, U Bergström*, J Olsson*, J Carstensen** *Dept Aquatic Resources, Swedish University of Agricultural Sciences , SWE **Dept Bioscience - Applied Marine Ecology and Modelling, Aarhus Univ., DK www.waters.gu.se

www.waters.gu.se

Coastal fish indicators • Not included in the WFD • Assessed as part of the MSFD in Sweden and in the HELCOM region Key species Cyprinids Piscivores www.waters.gu.se

Assessment units for coastal fish Coastal areas of 17 sub-units in the Baltic Sea Example gives the current status of Key Species based on long term monitoring data. (HELCOM Core Indicator Fact Sheet 2015) www.waters.gu.se

How about spatial variation? www.waters.gu.se

Available environmental monitoring data • Poor density in relation to the biology of the species www.waters.gu.se

Coastal fish in the Baltic Sea • Areas less than 20 m depth • Dominated by relatively stationary species, which may respond to environmental pressures originating from local as well as larger geographical scale. www.waters.gu.se

How does the status relate to changes in pressures? Topography Predation Salinity Fishing Eutrophication NOT MUCH EXPLORED YET Habitat availability Temperature Food availability Habitat quality www.waters.gu.se

Questions: 1. How important is inter-annual versus geographical variation? 2. What is the effect of small-scale environmental variability? (temperature, salinity, wave exposure) 3. How does fish abundance relate to natural environmental variables and anthropogenic pressures? (climate, salinity, eutrophication, fishing) www.waters.gu.se

Data included • 41 areas from the Bothnian Bay to the Bornholm Basin (Lat. 56-66 o N) • 2002-2013 (1-12 years per area, 30-45 stations per area) • Various level of anthropogenic disturbance Type Number of areas Reference area for monitoring 13 Marine protected area 5 Fish no-take area 3 Commercial harbour 7 Seal protection area 8 High nutrient levels 4 Other 8 www.waters.gu.se

Sampling method: test fishing with Nordic costal multimesh gill nets 9 panels Mesh sizes 10-60 mm common ratio 1.25 Depth X Length = 1.8 X 45 m Fish > 11 cm Photo: Ulf Bergström www.waters.gu.se

Studied response variables Abundance of Abundance of Perch Cyprinids Proportion of Perch above 25 cm www.waters.gu.se

Model 1: Components of variation Temporal and spatial sources of variation (GLMM): 𝐽 = 𝜈 + 𝐵 + 𝑇 𝐵 + 𝑍 + 𝑍 × 𝐵 + 𝑓 Indicator value = mean level + random variation among areas + random variation among stations within areas + random interannual variation + random changes in interannual variation + dispersion factor Where mean level ( μ ): a function of depth (d), temperature (T) and wave exposure (SWM) 𝑒 + 𝑙 2 ∙ 𝑒 + 𝑙 3 ∙ 𝑈 + 𝑙 4 ∙ log 𝑇𝑋𝑁 + 𝑙 5 ∙ log 2 (𝑇𝑋𝑁) 𝜈 = 𝑙 1 ∙ Perch and Cyprinids analysed as Poisson variables, Proportion of large perch analysed as binomial variables . www.waters.gu.se

Result 1a: Variation among areas dominated over interannual variation Relative variation explaned (%) Source CYP PER Large PER Between areas V[A] 10.3 3.8 19.3 Between stations V[S(A)] 5.0 1.7 8.6 Between years V[Y] 0.0 0.2 0.0 Between years and area V[Y×A] 1.9 1.3 7.9 Residual V[e] 82.8 93.0 64.2 www.waters.gu.se

Result 1b: Variation among stations was partly explained by variation in depth, temperature at fishing and wave exposure www.waters.gu.se

80 80 80 A) B) C) Cyprinids Cyprinids Cyprinids Cyprinidae Cyprinidae 60 60 60 Abundance Fish catch Cyprinids Cyprinidae 40 40 40 20 20 20 0 0 0 0 5 10 15 20 25 0 5 10 15 20 25 1000 10000 100000 1000000 Temperature (  C) Depth (m) Wave exposure () 50 50 50 D) E) F) Perch Perch Perch Perca fluviatilis Perca fluviatilis 40 40 40 Abundance Perch Fish catch 30 30 30 Perca fluviatilis 20 20 20 10 10 10 0 0 0 0 5 10 15 20 25 0 5 10 15 20 25 1000 10000 100000 1000000 Temperature (  C) 100% Depth (m) 100% 100% Wave exposure () G) H) I) Large perch Perch Perch Perch Proportion of larger perch Proportion Perca fluviatilis Perca fluviatilis 80% 80% 80% Perca fluviatilis 60% 60% 60% 40% 40% 40% 20% 20% 20% 0% 0% 0% 0 5 10 15 20 25 1000 10000 100000 1000000 0 5 10 15 20 25 Temp ( o C, 5-25) Depth (m, 0-30) Wave exposure Temperature (  C) Wave exposure () Depth (m) www.waters.gu.se

Model 2: Area specific means 𝐽 = 𝜈 𝑏𝑠𝑓𝑏 + 𝑙 ∙ 𝑈 + 𝑡(𝑒) + 𝑡(𝑇𝑋𝑁) + 𝑓 Indicator value = mean value of all 41 areas + a linear temperature effect + spline model of depth + spline model of wave exposure (GAM) . Area-specific marginal means were calculated for average values of T=15.9 o C, d=7.4 m and log(SWM)=8 to allow for direct comparison between areas. www.waters.gu.se

Calculated for average values of T=15.9 o C, d=7.4 m and log(SWM)=8 to allow for direct Area-specific means comparison between areas. www.waters.gu.se

How does fish abundance relate to environmental variables? Natural variables: • mean seasonal temperature (climate) • salinity Anthropogenic variables • water transparency, as seen in eutrophic areas • commercial catches of perch www.waters.gu.se

Relationship between fish abundance and the studied environmental variables Cyprinids Perch Large perch Salinity ns decrease ns Temperature ns ns ns Water transparency decrease plateau ns Commercial fisheries ns Increase ns www.waters.gu.se

Cyprinids increase in areas of low water transparency  Cyprinidae 40 Fish catch 30 Abundance 20 10 0 0 1 2 3 4 5 6 7 8 Secchi depth (m) Water transparency (Secchi depth, m) www.waters.gu.se

Perch decrease with salinity and increase with commercial catches 10 10 D) B) Commercial Salinity 8 8 catches Fish catch Fish catch 6 6 4 4 Perch 2 Perca fluviatilis 2 Perch Perca fluviatilis 0 0 0 5 10 15 20 25 30 2 3 4 5 6 7 8 Commercial perch landing (kg km -2 ) Salinity Temperature Water 10 10 A) C) Perch Perca fluviatilis Perch transparency 8 8 Perca fluviatilis Fish catch Fish catch 6 6 4 4 2 2 0 0 10 11 12 13 14 15 16 17 0 1 2 3 4 5 6 7 8 Temperature (  C) Secchi depth (m) www.waters.gu.se

Conclusions • Method development: extend the geographical applicability of coastal fish status assessment • Abundance of Cyprinids is a useful indicator for assessing ecological status in relation to eutrophication • No negative relationship between commercial fisheries and perch abundance www.waters.gu.se

Thank you for listening! Photo: Lena Bergström www.waters.gu.se