As the number of offshore wind farms increases, it is important to correctly assess the impact that these developments can have on wildlife. New research led by the BTO examines this situation for seabirds, considering the current environmental impact assessment process in light of the key factors that determine seabird population dynamics. The construction of offshore wind farms is rapidly increasing as governments aim to reduce carbon emissions. However, since there is a growing body of evidence linking offshore wind farms to increased rates of mortality and displacement in seabirds, many countries require a full assessment of any potential impacts to seabird populations before giving consent for wind farm development. These impact assessments follow the precautionary principle, so that consent decisions are based on estimates of the maximum harm that could be caused to seabird populations. Such assessments typically exclude any consideration of density-dependent processes on the population concerned. Density-dependence can be both positive and negative. Negative density-dependence occurs when, for example, the death of breeding adults frees up space in the population for young birds to join it and become breeders themselves. This process can therefore offset the loss of individuals from a breeding population. Positive density dependence can take place once a population falls below a critical threshold. Small populations might not have the ‘safety in numbers’ benefits of seeing off predators, or have a sufficient numbers of potential mates to make the population sustainable. Populations can change from negative to positive density-dependence should they become too depleted. Seabirds are one of the most threatened bird groups in the world, so understanding the implications of wind farm impact assessments for their breeding populations is crucial. New research by the BTO and the JNCC collated evidence for density-dependent population processes in 31 species. The results found widespread evidence for negative density-dependence, especially amongst the large gulls and auks, meaning that populations may be able to withstand small and infrequent losses associated with renewable energy developments. However, extinctions will still occur if the number of losses is greater than the number of new breeding recruits, and species that are highly vulnerable to wind farm developments include large gulls, small gulls, Gannets, seaducks and divers. The study also showed that positive density-dependence was prevalent in the smaller species that breeding in colonies, such as terns, small gulls and auksThese species are likely to experience accelerated rates of population decline at low densities due to increased predation from large gulls and corvids. For such populations impact assessments that ignore density-dependent processes will underestimate the projected impact of a wind farm and could therefore overlook potential extinction events. Scientifically robust estimates of the expected impacts of wind farms on seabird populations are critical. The evidence for density-dependence in seabird populations indicates that the current impact assessment process does not offer a fully precautionary approach. A more robust approach would be to compare the projected population size with and without the expected population changes associated with the proposed development, and test how this changes under a range of density-dependent scenarios. For more information about this research, please see this blog for the Journal of Applied Ecology.

Populations of many species of migratory bird are declining in Britain. However, the picture is not equally gloomy across the country. Many species are doing much better in northern Britain than they are in the south. Recent research, led by Cat Morrison at the University of East Anglia in collaboration with BTO staff, has used BTO data to understand why this difference occurs. Among the species faring better in the north is the Willow Warbler. This tiny inter-continental traveller used to be one of our commonest species, but it is now very scarce indeed in some places in the south-east of England. This work combined data from several BTO schemes to better understand the demographic causes of these patterns. By constructing an integrated population model (IPM) for each region, the authors untangled the different effects that productivity (from nest record visits) and survival (from CES captures) have on the number of breeding birds (from the BBS counts). This work shows that while changes in the number of breeding birds are primarily affected by the survival between years, the difference in the overall population trend between the two regions arises as a consequence of differences in productivity. Between 1994 and 2012, annual survival and productivity rates ranged over similar levels in the two regions, but years of good productivity (i.e. lots of chicks fledged) were rarer in the south, where the population is declining. In particular, years of good productivity never coincided with years when the survival rate was also high. In contrast, population growth in the north was fuelled by several years in which good productivity coincided with high survival rates. To assess the importance of this difference we modelled what the population changes might have been in the south using a realistic range of productivity values (including those achieved by birds in northern Britain). This showed that, with productivity similar to their northern cousins, populations in the south would have recovered. Consequently, actions to improve productivity on breeding grounds, for example improving the size and quality of available habitat, especially in areas (such as southern Britain), where there are currently population declines, are likely to be a more fruitful and achievable means of reversing migrant declines than actions to improve survival on the breeding, passage or African wintering grounds.

A common issue that many analysts of biological data encounter is that of detectability. For a human population we can (in principle) count every individual. For wildlife though, things are trickier, and only rarely is this possible. Bird’s nests are a good example of this - we cannot find every nest. Some are well hidden, some are out of reach, and some we just miss. When we do find a nest, it is rarely right at the start of egg-laying, mostly we find them when they already have eggs or chicks in them. This makes coming up with unbiased estimates of nest survival and success tricky, especially when considering that the chances of a nest surviving differ depending on whether it contains eggs, or noisy chicks. Fortunately, there are statistical methods that enable us to overcome many of these problems, but they suffer from one problem. They assume that the dates when the eggs were laid and the chicks hatched and subsequently fledged are known. This though is rarely the case, since most nests are not visited every day. Recently our statistician Mark Miller has developed an extension of these statistical methods that enable one to account accurately for this imperfect information. This new method was applied to the thousands of Blackbird nest record cards collected by BTO volunteers between 2003 and 2011. The study particularly focused on whether there was a difference between Blackbirds nesting in gardens and those nesting in the countryside. The results showed that the nest survival of Blackbirds nesting in suburban habitats was higher than in either urban or rural areas. There was also an effect of rainfall. In both town and countryside nests survived better when it was wetter, presumably reflecting better conditions for birds to find the soil invertebrates which make up much of their diet. However, in towns, nest survival was related to rainfall in the weeks preceding nest-laying, so a very immediate effect. In the countryside, by contrast, overall wetness of the soil, which is influenced by rainfall several months previously, was more important. This difference probably reflects the much greater run-off and drainage of surface water in urban areas (because of tarmac), whereas in the countryside moisture is much more evenly distributed in the soil. As rainfall patterns are expected to alter as part of wider climate changes birds in different habitats are likely to respond differently, and some may be constrained in how they are able to do this. Nest recorders will continue to gather data to help understand these changes as we devise solutions to help our birds adapt to the changing environment.

Climate change is a much discussed topic. There has been significant warming in the UK since the 1960s, with land temperature from 2005-2014 0.9°C higher than the 1961-1990 mean, and detectable shifts in rainfall patterns. During this time, there have been significant changes in biodiversity too, with long-term declines in some of our bird species, such as on farmland and in woodland, and in our moths. Other taxa have seen increases however, including some of our mammal species like deer. An important component of our work at BTO is to identify the causes of population changes in our biodiversity. Here, we consider the role that climate change may have played in driving some of these long-term trends. In a collaborative project led by BTO, we analysed data on population trends from the Rothamsted Insect Survey (aphids and moths), the National Bat Monitoring Programme, the UK Butterfly Monitoring Scheme, and the Breeding Bird Survey (birds and mammals). We modelled annual changes in the abundance of over 500 species from these schemes as a function of monthly variation in temperature and precipitation, summarised into a small number of principal components. These models were then used to identify the extent to which long-term population trends were consistent with the trend expected from changes in temperature and precipitation. We used this approach to infer the potential role of climate change in driving these population trends, although a proportion of this contribution may have been related to natural variability in the weather. Our results suggest that climate change may have had a significant impact on the long-term trend of 79 species since the 1970s. Trends of eight rapidly declining species matched the negative impacts of climate change expected from our models, including two birds (Lesser Redpoll and Common Snipe) and six moths (Mottled Umber, Little Emerald, Northern Winter Moth, Twin-spot Carpet, Broom Moth and Minor Shoulder-knot). Positive population trends of four species were consistent with the modelled increases in abundance expected from climate change (Greylag Goose, Canada Goose, Lesser-spotted Pinion and Reeves Muntjac). Across species, moth populations declined by an average of 1.4% per year, half of which was consistent with the expected impact of climate change. Conversely, winged aphid abundance increased annually by 0.7%, of which over 60% may have been caused by climate change. Although it is difficult to definitively attribute long-term trends to climate change, this study suggests that at least some long-term trends in terrestrial biodiversity may have been caused by climate change. Although overall changes in bird, mammal and butterfly populations were not strongly related to climate change, matching the results of previous work on farmland birds, and suggesting that other drivers of change have probably been more important for these groups, our results strongly suggest that climate change may have significantly contributed to the decline of moths, particularly in southern Britain, and to increases in winged aphids. This information helps identify the species that may be most vulnerable to future impacts, and importantly suggests that we need to be closely monitoring trends in the abundance of bird species that rely on moth caterpillars for food.