Modelling seabird breeding habitat selection

海鸟在近海岛屿和/或海岸悬崖上繁殖，以减少与陆地捕食者的接触，并靠近觅食区。然而，对于悬崖筑巢的物种来说，很明显不是所有的悬崖都是一样的：鸟类以每平方米70多个鸟巢的密度聚集在一些地方，而其他悬崖则是空的，即使它们显然有适合筑巢的壁架。暴露在风中可能是一个关键因素，因为它对体温调节、飞行能力和强风将鸟类赶出巢穴的能力有影响，例如，不同物种的体重和飞行方式不同，风的重要性也不同。为了支持这一点，最近的一项研究表明，普通海鸠（以下简称海鸠）的大型繁殖地可以通过简单的代理来预测两个相邻小岛上的避难所。尽管如此，气流和适应性之间的联系还有待检验，而且庇护所是否能预测区域范围内的繁殖分布，包括在不同暴露于强风的种群中，仍然是未知的。这代表了重要的知识空白，因为(i)蚁群可以代表成千上万个体的繁殖活动的联系，（ii）风力正在发生变化。本文将以海鸠（Uria aalge）、海鸠（以下简称海鸠）、刮嘴鸟（Alca torda）、黑腿三趾鸥（Rissa tridactyla）和北方管嘴鸟（Fulmarus glacialis）为研究对象，确定风对悬崖筑巢海鸟种群水平过程的影响。我们把重点放在悬崖筑巢的物种上，因为它们最容易暴露在风里。它们也代表了英国75%以上的海鸟（在JNCC海鸟2000年报告中记录的大约800万只海鸟中超过600万只）。因此，从保护的角度和对海洋群落的更广泛影响来看，风海鸟的影响都是令人感兴趣的。研究风对筑巢地影响的一个关键限制是在陡峭和难以到达的地形上测量风的困难，比如筑巢的悬崖。我们将采用一种雄心勃勃的跨学科方法，使用计算流体动力学软件（CFD）来描述英国9个地区鸟类在繁殖悬崖上选择和避免的气流。CFD被广泛应用于其他学科（例如风能行业），并被证明是理解洋流对河流鱼类行为和生态影响的关键。然而，在动物生态学中还缺乏更广泛的应用。我们将使用超级计算设备来模拟我们研究区域的风，这代表了在鸟类生态中使用这种模型的规模增加了3个数量级。这将使我们能够确定(i)鸟类是否选择与其提供的庇护所相关的繁殖栖息地，庇护所是否可以预测（ii）亚种群之间繁殖成功率的差异，以及（iii）十年尺度上的种群趋势。最后，我们将测试悬崖方向的简单测量是否可以用作庇护所的代理，这将使我们的方法应用于更大的区域，而不需要相关的计算成本。我们的项目利用了英国各地繁殖海鸟的丰富数据，包括至少十年的计数和繁殖成功数据，这些数据来自9个地区，这些地区在强风下的暴露程度各不相同，这代表了我们研究物种的英国繁殖种群总数的很大一部分。产出将包括第一个区域尺度的海鸟栖息地选择模型，以及对风影响更广泛种群水平过程的机制的新见解。

Seabirds breed on offshore islands and/ or coastal cliffs to reduce exposure to terrestrial predators and be close to feeding areas. Nonetheless, for cliff nesting species, it is clear that not all cliffs are equal: Birds crowd into some areas at densities of over 70 nests per metre squared, while other cliffs are left empty, even when they apparently contain ledges suitable for nesting. Exposure to wind is likely to be a key factor due to its impact on thermoregulation, flight capacity and the ability of strong winds to displace birds from the nest, with the importance of wind varying between species according to body mass and flight style, for instance. In support of this, a recent study has shown that large breeding colonies of common guillemots (hereafter guillemots) can be predicted from simple proxies for shelter on two small, neighbouring islands. Nonetheless, the links between airflows and fitness have yet to be tested, and whether shelter can predict breeding distributions at a regional scale, including in colonies that vary in their exposure to strong winds, remains unknown. These represent important knowledge gaps because (i) colonies can represent the nexus of reproductive activity for tens of thousands of individuals and (ii) wind regimes are changing.We will establish the impact of wind on population level processes in cliff-nesting seabirds, taking common guillemots (Uria aalge), hereafter guillemots, razorbills (Alca torda), black-legged kittiwakes (Rissa tridactyla) and northern fulmars (Fulmarus glacialis) as our study systems. We focus on cliff-nesting species as these will be the most exposed to the wind. They also represent over 75% of the seabirds breeding in the UK (over 6 million of the approximately 8 million recorded by the JNCC Seabird 2000 report). The influence of wind seabirds is therefore of interest both from a conservation perspective and from the wider impact on marine communities. A key constraint in the ability to study the impact of wind at the nest site has been the difficulties of measuring wind over steep and inaccessible terrain, such as nesting cliffs. We will take an ambitious, interdisciplinary approach using computational fluid dynamics software (CFD) to characterise the airflows that birds select and avoid at their breeding cliffs in 9 areas of the UK. CFD is widely used in other disciplines (e.g. in the wind energy industry) and has proved pivotal for understanding the impact of currents on the behaviour and ecology of riverine fish. However, wider applications in animal ecology are lacking. We will use supercomputing facilities to model the wind in our study areas, which represents an increase in 3 orders of magnitude in the scale at which such models have been used in avian ecology. This will allow us to establish (i) whether birds select their breeding habitat in relation to the shelter it provides, whether shelter can predict (ii) differences in breeding success between sub-colonies, and (iii) the population trend over a decadal scale. Finally, we will test whether simple measures of cliff orientation can be used as a proxy for shelter, which would allow our approach to be applied to even larger areas without the associated computational cost. Our project capitalises on the rich data available for breeding seabirds across the UK, including at least a decade of count and breeding success data from 9 areas that vary in their exposure to strong winds, and that represent a substantial proportion of the total UK breeding populations of our study species. The outputs will include the first regional-scale habitat selection model for seabirds, as well as novel insight into the mechanisms by which wind impacts wider population-level processes.