Abstract
With the increasing expansion of offshore wind farms in the German North Sea concerns have been raised about the protection and preservation of the winter and spring resting populations of the two diver species, red-throated (Gavia stellata) and black-throated divers (Gavia arctica). The German North Sea is an important habitat for both species. As various studies suggest a stronger response of divers to offshore wind farms than previously expected and due to the high conservation status of these two species, there is a strong interest in assessing whether the development of diver populations is influenced by the expansion of wind power.
The present study is based on a high-quality data set on the distribution and occurrence of divers in the German North Sea over the last 18 years and represents the most extensive data set currently available. For spring, 16 years of data were available (no data for 2006 and 2007), for winter, 17 years were available (2006 missing). This data set considers only high-quality aerial survey data and not data from ship surveys, which is considered less suitable for the detection of divers. Through the application of a Bayesian spatial and spatio-temporal model approach, which is well suited for the complex problem of quantifying diver displacement, a reliable analysis of the development of diver populations between 2001 and 2018 is possible. Furthermore, for the first time, data on diver abundance during the winter season were also analysed in addition to spring data.
The main aim was to analyse trends in diver populations over the last 18 years, before and after the expansion of offshore wind farms (OWF) in the German North Sea. In addition, we wanted to find out up to which distance divers are displaced by offshore wind farms and what theoretical habitat loss might result from this, as well as quantifying possible variation between sub-areas and seasons.
The four key aspects were:
1. Reliable calculation of diver population size over the 18-year study period for the entire German North Sea as well as for a northern and a southern sub-area and examination of whether changes in the population might be related to the expansion of offshore wind farms.
2. Investigation of the spatial distribution of divers in relation to the location of offshore wind farms, taking into account seasonal (spring/winter) and local (north/south) factors.
3. Reliable analysis of the displacement distance of divers from offshore wind farms, taking into account seasonal (spring/winter) and local (north/south) factors.
4. Calculation of a theoretical habitat loss based on the current dataset.
The main result of this study is that over the entire study period, the spring abundance of divers fluctuated between individual years without any clear trend, with overall stable population numbers between 2001 and 2018. No connection with the expansion of wind power in the German North Sea and the inter-annual variability in diver abundance was found. In spring, divers reached the highest numbers, and on average 16,500 divers were estimated in the German North Sea. The northern part of the German North Sea and the main concentration area therein, as defined by the BMU (2009), are of the greatest importance here, as it accounts for approx. 60 % of the German North Sea population over all the years examined. The abundance also fluctuated in the North and South sub-areas without a decline being observed with the expansion of offshore wind power. In winter, there was an increase in the population over the years, especially for the southern sub-area. However, compared to spring, there were significantly fewer divers in the German North Sea in winter (n = 4.833).
After the expansion of offshore wind farms in the northern part, a less variable distribution of divers was observed. The birds concentrated relatively consistently in a central area of the main concentration area, which has the longest possible distance to all surrounding wind farms. However, this area was also used by divers before construction of the first wind farms, with high densities in some years (e.g. 2003, 2010). Despite this partial redistribution, the number of divers in the main concentration area is not declining and the area is still very important for resting divers in spring. In winter, the spatial distribution is more variable and the birds are more likely to be much nearer to the coast and in significantly lower densities overall.
The results showed that divers keep different distances from the wind farms, depending on season (spring/winter) and area (north/south). The most reliable calculation of the avoidance distance is based on spring data, as this is the period with by far the highest density of divers. In spring, a displacement distance (gradient) of 10.2 km was calculated for the entire study area and for all available data which therefore forms the most robust result (see Box 1 for details on calculation). In the two sub-areas North and South, slightly shorter distances were calculated. Due to a significantly flatter displacement curve in the southern area, where less than 20 % of the spring population of divers were present, a theoretical habitat loss of 2 km (radius around a model OWF) was calculated here, while a theoretical habitat loss of 5 km was calculated for the northern area (comprising >75 % of the diver population in spring) and for the entire data set.
In winter, large differences in the displacement distance to offshore wind farms were observed between the northern and southern sub-area, potentially due to the considerably lower diver densities and the resulting greater uncertainties in the analyses. Nevertheless, these differences show that seasonal and spatial factors may play a role in the specific response of divers to offshore wind farms and results found here are therefore not directly transferable to areas other than those considered in this study.
The shape of the wind farm footprint has only a minor effect on the theoretical habitat loss. The total number of divers displaced can be reduced by combining individual OWF projects into clusters so that displacement radii overlap.
In addition to the Bayesian distance model, a before-after approach was also applied for the northern sub-area to calculate avoidance-distances. The results for two different before periods (2001-2005 and 2008-2011) show larger displacement distances (11 and 13 km, respectively), but also reveal uncertainties regarding the reference period to be chosen. A comparison of the diver distributions before and after the wind farm extension shows, depending on the selected reference period, areas of different sizes with significant diver decrease and increase, which are not circular around the respective wind farms. There is limited knowledge about the mechanisms behind the displacement effect and so possible physiological consequences and subsequent possible effects on long-term population development cannot be assessed. But it is apparent, however, that the local population within the German North Sea is stable during the time period analysed.