East Anglia ONE Offshore Windfarm
April, 2016
Density 0-500m
Density 500-1000m
0.2 0.4 0.6 0.8 1.0 Density (birds/km2)
Density 1000-1500m
0.2 0.4 0.6 0.8 Density (birds/km2)
Density 1500-2000m
0.2 0.4 0.6 0.8 Density (birds/km2)
0.2 0.4 0.6 0.8 Density (birds/km2)
Figure 3. Histograms of gannet density within 500m buffers around 1,000 randomly relocated wind farms, with the density within the buffers around the actual wind farm indicated by red lines, all data generated using before data (i.e. no wind farm effect).
22.
The after data were generated with birds avoiding the wind farm by up to 2,000m (simulated as up to 500m, 1,000m, 1,500m or 2,000m ). In Figures 4 to 7, the position of the red lines were offset to the left relative to the resampled distribution (histogram) indicating a difference in avoidance between the ‘real’ wind farm location and the randomly offset ones. However, it is important to note that this effect was only seen in the histograms which matched the actual distance used to generate the data (and smaller distances). Thus, these tests identify the avoidance distance used to generate the data. For example, if the data were simulated on the basis of bird avoidance up within 1,000m from the wind farm the red lines were offset to the left of the bootstrap distribution in the 0-500m and 500-1000m plots, but not at higher distances.
23.
A permutation test was used to calculate how likely it was that the shifts to the left in the ‘true’ density relative to the simulated wind farm locations would be obtained by chance. This probability of a chance effect was calculated as the proportion of random samples which had density estimates as small, or smaller, than the observed estimates (Table 4).
24.
The results of the permutation tests indicate that if birds avoid the wind farm the bootstrap simulation method will be expected to identify the distance (to the nearest 500m) over which avoidance is occurring (this can be seen in the threshold probabilities in the shaded cells of Table 4). For example, if birds avoid the wind farm by 1000m (Table 4 column 2), in none of the 1,000 simulated wind farm relocations was a lower density obtained.
Page 10
Frequency 0 100 200 0
Frequency 200
Frequency 0 50 150
Frequency 0 50 150
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
