East Anglia ONE Offshore Windfarm Generation Assets Monitoring Plan


September, 2016


51.


Once the survey has been completed the data will be backed-up and quality checked to ensure the survey has been completed successfully and all lines and images have been collected. The imagery will be processed to alter the colour balance to optimise target detection, then georeferenced using the GPS Log and batched into folders containing 25 images per batch. The image analysis process comprises screening the entire image to identify any possible targets (birds, marine mammals, vessels and other notable objects). Once a target has been identified the analyst measures the body length and wingspan and selects the species from a predefined list to ensure data consistency. The survey target species, set through an agreement between EAOL, MMO and Natural England, will be gannet, kittiwake, great black-backed gull, guillemot and razorbill. All birds (and marine megafauna) will be geo-referenced with a GPS coordinate that can be exported into ArcGIS Shapefiles for the mapping of all individual bird locations. In addition to standard information including image time, additional species-dependent information will be recorded including species, age, age class and sex (where possible), whether a bird is part of a group or alone, behaviour (flying, sitting or diving), flight direction and flight height, with associated confidence intervals.


52.


The presence of anthropogenic features that may influence the occurrence, location and behaviour of seabirds and marine megafauna (e.g. fixed structures and vessels, including type etc.) will also be recorded.


53.


Once the image analysis process has been completed for a survey, the target data will be exported to an excel format (which contains information including length, species, sex and gender where applicable, bearing, and location coordinates) and will be converted into an ArcGIS shapefile.


3.1.1.7 Quality Assurance 54.


All birds detected will be Quality Assured and assigned to a species group and where possible, each of these will be identified to species level. Where identification to species is not possible individuals will be assigned to taxonomic groups such as:


• ‘black-backed gull sp.’ (lesser black-backed or great black-backed gull); • ‘grey gull sp.’ (common, herring, black-headed or little gull); and • ‘gull sp.'


4 Analysis and Reporting 4.1 Summary survey results 55.


In addition to the modelling methods described below, summary outputs from the surveys (e.g. raw counts, proportions of each species at rotor height, on the sea surface, etc.) and figures providing the locations of key species will be provided. Flight heights will be analysed with respect to location to identify patterns in relation to the wind farm (noting that this will not identify any trends in the pre-construction surveys). As the monitoring surveys have been designed to address specific questions (as discussed above), direct comparison with data collected during the baseline site characterisation surveys is not appropriate or straightforward to undertake. Hence, comparison of the results of the two surveys (current surveys and characterisation surveys) will be based on a qualitative review. This will discuss similarities and differences between the datasets thereby providing context for the monitoring survey results.


4.2 Analysis Methods 56.


Two complementary methods will be used to analyse the survey data. The first will use the spatial modelling techniques developed by St. Andrews University (Mackenzie et al. 2013) to estimate seabird density, abundance and spatial distributions (surface maps) and the second is designed to estimate seabird avoidance distances at two spatial scales: from the wind farm (as a whole) and turbines. As the current surveys will be undertaken prior to turbine installation these analyses will not identify wind farm induced effects, however they will permit illustration of how these methods will be used once post- construction surveys have been conducted.


4.2.1 Spatial modelling 57.


The data will be analysed using the spatially adaptive statistical methods CReSS (Complex Region Spatial Smoother) and SALSA (Spatially Adaptive Local Smoothing Algorithms) developed by the Centre for Research into Ecological and Environmental Modelling (CREEM) at St. Andrews University (Mackenzie et al. 2013). The method fits a two-dimensional


Page 13


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