A. Gimpel et al. 
Fable 1 
Surveys, sampling techniques and results from data taken inside the Offshore Wind Farm during summer (OWF summer), during winter (OWF spawn & OWF winter) and at 
stations outside the OWF in the German Bight during summer (GB summer) and winter (GB spawn & GB winter). Given are the mean values and the standard deviation for the 
catch per unit effort angling per hour (CPUE), age, length, weight, condition K®, the stomach fullness index Ip”, the carbon isotopes 8°C and nitrogen isotopes 8!°N, maturity 
stages and sex rations, the origin of eggs (OWF or GB), the age of the eggs and the distance the eggs travelled. ION = modified Indian Ocean Net; TD logger = Temperature 
and Depth logger). 
Survey IWF summer 
Sample Zod adults 
Nind farm area Inside 
Gear used Fishing rod 
"imeline Jun 2019 
Stations 1 
Jepth at station (m) 20.7 (+ 9.5) 
Sample size (N) 122 
ZPUE 5.45 (+ 2.67) 
Age (y) „7 (+ 0.7) 
„ength (cm) 12.4 (+ 10.1) 
Neight (g) 109.5 (+ 689.5) 
Zondition K „01 (+ 0.13) 
F „46 (+ 1.33) 
Mean 5'°N (%o; N = 5) „9.36 (+ 0.10) 
Mean 8*°C (%o; N = 5) 17.27 (+ 1.02) 
Sex ratio (F/M/U) 72/50 
Maturity stage (N) NA 
Origin of eggs (%) 10.3 89.7 
Distance (km) 4.5 (+ 2.3) 12.6 (+ 8) 
Age (hr) 90.3 (+ 55.5) 178.4 (+ 121.1) 
* In order to compare the fish condition across the sampling areas, the body condition index K was calculated for every individual, 
X = 100 W42*100 
Science of the Total Environment 878 (2023) 162902 
where K is the Condition, W is the Total weight (gr) and L is the Length of the individuals (total length, cm) (Ovegärd et al., 2012). 
> To investigate the pattems in feeding rates, we calculated a stamach content fullness index (Ir) for each stomach: 
Fr = 10000CL’— 3 
where C is the stomach content mass (full stomach —- empty stomach in g) and L the length of the individual in cm (Reubens et al., 2013c). 
‘chthyoplankton surveys to investigate presence and potential source loca- 
ions of cod eggs in the OWF and the GB (Aurich, 1941; Fox et al., 2008). 
We performed the surveys inside the OWF (OWF spawn; Fig. 1 and 
Table 1) and in the wider GB (GB spawn; Fig. 1 and Table 1) in January 
2019 and January 2020. We carried out vertical plankton hauls using an 
'ndian Ocean Net (ION) from 6 m above the seafloor to the surface. The 
jet had a mouth opening of 1 m diameter and was equipped with 405 um 
nesh. In-situ temperature and depth were recorded with an attached 
Temperature-Depth-logger (TD logger, Star-Oddi). 
2.1.3. Database 
In total, 217 cod adults were available for our analysis: 134 specimens 
were caught by angling in the OWF during summer (n = 122 from a total 
of 11 stations) and winter (n = 12 from a total of 5 stations). Fifty-four 
specimens were available from a total of 145 bottom trawl stations in the 
GB during summer. Further, data from the German Bight in winter were re- 
trieved from the ICES Database of Trawl Surveys for 2019 and 2020 
(DATRAS; http: //datras.ices.dk/; Extraction 7 December 2022 of Interna- 
:jonal Bottom Trawl Survey (IBTS). ICES, Copenhagen), including all ICES 
rectangles, that overlap with the German EEZ. Altogether, 29 specimens 
were available via DATRAS from 53 stations in winter (Table 1, Fig. 1). 
In total, 214 cod-like eggs were further sampled during our 
ichthyoplankton surveys at 34 stations in the OWF (OWF spawn) and 36 
stations in the wider GB (GB spawn) during winter (Table 1, Fig. 1). 
2.2. Length frequencies and diet composition 
In order to study the effect of offshore wind turbines with a scour pro- 
tection, we tested if size classes, diet composition and trophic niches of 
cod found in the OWF differ from individuals sampled in the GB. We col- 
lected information on total length (to the nearest 1 cm) and weight (to 
che nearest 1 g) and extracted ear stones (ie. otoliths) for age determination 
5y reading annual rings of all individuals caught in the OWF and the GB 
(Table 1). 
In order to investigate if food availability differs between the OWF and 
he surrounding area, which might affect local cod densities, we identified 
stomach contents. No stomach samples were given for the GB winter 
/DATRAS) data, leading to an overall sample size ofn = 188. After capture, 
;tomachs were removed from the abdominal cavity and frozen on board. 
Analyses of stomach contents were carried out in the laboratory onshore. 
The total weights of full and empty stomachs was determined and contents 
.dentified to the lowest possible taxonomic level (Hayward and Ryland, 
‚990a; Hayward and Ryland, 1990b). For fish, we used the species- 
specific shape of the otolith to identify the species (Campana, 2004). The 
wet weight of each prey item (to the nearest 1 mg) was determined. The sta- 
istical analyses and the comparison of diet compositions required the 
grouping of some taxa into taxonomically or functionally distinct groups. 
Taxa were assigned to one of the four groups ‘other crustaceans’, 
cephalopoda’, ‘flat fish’ or ‘fish others’. Due to their high frequency of oc- 
currence and ease in identification, edible crab (Cancer pagurus), swimming 
crab (Liocarcinus spp.), common hermit crab (Pagurus bernhardus), masked 
crab (Corystes cassivelaunus), porcelain crab (Porcellanidae), shrimps 
“Crangonidae) and butterfish (Pholis gunnellus) represented separate, indi- 
vidual categories. We identified diet items as ‘benthic remains’, when 
:hey could not be clearly identified but related to benthic feeding (ie. 
stones or sponges). 
A total of 12 prey categories were used in the statistical analysis. Items 
where the degree of digestion prevented identification were excluded from 
‘he analysis (Reubens et al., 2014a; Werner et al., 2019). All other samples 
‚egarded acceptable for analyses (n = 167) were standardized through con- 
version to proportional data (sum equal to 1.0) (Appendix A, Table A.1). 
We examined patterns in diet composition using Canonical Correspondence 
Analysis (CCA), a multi-variate approach. For the CCA we used gradients in 
an independent set of explanatory variables in order to explain variability