Detecting displacement effects of offshore windfarms on red-throated divers wintering in the Outer Thames Estuary

Stephanie McGovern, Mark Rehfisch, Bethany Goddard
APEM, Stockport, UK


Separating natural fluctuations in avian distribution and abundance from changes caused by human activities is a key challenge in environmental impact assessments (EIA). Displacement effects may preferably be analysed using model-based approaches that include physical habitat features since count data are spatially limited and do not allow a complete characterization of an area. Here, we present the successful application of a spatial model to assess distributions of red-throated divers (Gavia stellata) in relation to the construction and operation of offshore windfarms in the Outer Thames Estuary (McGovern et al. 2016). This area contains several offshore windfarm sites that have been developed over the last fifteen years. The area also supports the largest aggregation of wintering red-throated divers in the UK.


Spatial models were based on bird count data from aerial high-resolution digital images collected since 2009 in the Outer Thames Estuary. Pre- and post-construction survey data were analysed using the Complex Region Spatial Smoother (CReSS) and the Spatially Adaptive Local Smoothing Algorithm (SALSA). Models were developed using all available data from the pre-construction, during-construction and post-construction periods for the London Array offshore wind farm. CReSS models included both static and variable environmental factors to categorise and determine a relationship between the environment and bird abundance. The model allowed extrapolations from the surveyed area to the rest of the area of interest using a prediction grid. The model building, selection and testing followed the latest guidance for CReSS/SALSA using the MRSea package in R.


Divers showed a significant decline in density between the pre-construction and during-construction periods. Diver distributions altered with proportionally fewer birds being seen in the wind farm and surrounding areas during the construction period than were recorded during the pre-construction period. First results from the post-construction period suggest that divers may recolonize the wind farm after construction has ceased.


The CReSS/SALSA method provides a suitable model framework on which to base further analysis of displacement effects of offshore windfarms on avian distribution and abundance relative to the effects of other human activities (e.g. boat traffic). Incorporating additional post-construction data will allow more definite conclusions on the net effect of offshore wind farms on birds.


EIA of offshore projects should incorporate spatial models based on existing census data to test the significance of changes in avian distribution and abundance over time and to define the scale and source of potential impact objectively.