East Anglia ONE Offshore Windfarm


April, 2016


8.


Input data for the simulations were the coordinates of the wind farm boundary and proposed turbine locations and the estimated monthly average densities of the seabirds recorded during the baseline characterisation surveys (EAOW 2012). The months agreed for surveys with the MMO were October to March, however initial tests also included April in order to confirm that the omission of this month did not affect the results obtained (Table 1). It should be noted that the densities in Table 1 reflect the relevant species-specific data in relation to the primary impacts of concern. Hence, for gannet and kittiwake which are at risk of collision with turbines the key metric is the density of birds in flight. The values given for these species were extracted from the revised collision risk assessment for East Anglia ONE presented in the technical appendix for East Anglia THREE (EAOW 2015). Auks, which typically fly below rotor height, are of concern in relation to displacement. Hence, the auk densities used were those for birds recorded on the sea surface. The key feature for analysis is the density, therefore on review of the values recorded in the baseline surveys it was apparent that selection of a low density, medium density and high density species would cover the range recorded. Thus, simulations were based on the density of gannet, kittiwake and guillemot for these species respectively, and there was no additional benefit from modelling razorbill.


Table 1. Seabird densities recorded during baseline surveys of the East Anglia ONE site (EAOW 2012). Month October


November December January February March April


Density (average birds/km2) Gannet (in-flight) Kittiwake (in-flight) Guillemot (on sea) Razorbill (on sea)


0.286 1.452 0.055 0.016 0


0.018 0


2.1 Spatial Modelling 9.


To investigate the effect of alternative survey designs on the power to detect change in seabird distributions from a before- after comparison the following survey parameters were modified: transect separation (e.g. 1.4km or 2,2km), survey frequency (e.g. 1 or 2 per month) and impact magnitude expressed as the percentage of individuals displaced from the wind farm (e.g. 30% or 50%). To test for effect detection, 100 iterations of each scenario were undertaken, with the outcome of each scored as 1 or 0 to indicate success or failure of effect detection (respectively). The summed outcome across the 100 simulations provided an estimate of the power of the survey design to detect an effect with the standard threshold for acceptance of a design set at 80%.


10. The modelling was conducted using R (Cran) and followed these steps:


1. The survey region was defined from the wind farm site boundary, buffered by approximately 4km to the east, south and west and 10km to the north (as this is the direction from which autumn migrants are expected to originate). The planned turbine locations were used to align survey transects across rows in an approximate east-west orientation (aligned at a bearing of approximately 80°; Fig 1.);


2. The transect width was set at 250m and was divided into cells 500m long. Two different transect spacing distances were tested, 1.4km and 2.2km which gave transects aligned with every 2nd or 3rd turbine row respectively;


3. Randomly generated bird locations (x/y) were simulated using a uniform distribution for each month and year of surveys (i.e. October to April in the pre-construction and post-construction years). The total number of bird locations generated was calculated as the bird density for that month multiplied by the total area of the survey region. It should be noted that for species-month combinations which had estimated zero density (e.g. gannet in February) no survey was simulated, reducing the number of months included in the analysis for those species;


4. Each bird location was identified as being within or outside the wind farm (site boundary); 5. Points assigned to pre-construction months (i.e. pre-construction) were left unmodified while a proportion of the points within the wind farm assigned to post-construction months were relocated outside the wind farm simulating


0.032 1.628 1.205 0.346 0.210 0.259 0


0.135 0.675 2.905 6.015 2.675 1.795 0.205


0.045 0.255 0.32 0.92


0.435 0.53 0.07


Page 5


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35