Breeding Woodcock Ireland survey data and analysis

This data contains the data and code required to replicate analyses for a two-step species distribution modelling approach to estimate the distribution and abundance of breeding Eurasian woodcock (Scolopax rusticola) across the Republic of Ireland. Quantitative species occurrence data was collected at 238 randomly-selected forest sites by volunteers who counted displaying (roding) male woodcock at dusk in May and June of 2017, 2019 and 2021. Environmental data used consisted of historical bioclimatic variables downloaded from www.worldclim.org, and landcover (forest and urban settlement) derived from datasets provided by Coillte, DAFM Forestry Division, DAERA Forest Service and CORINE landcover. The first step was an ensemble model using biomod2 in R, to estimate the geographical range of breeding woodcock in Ireland using bioclimatic variables at a 1 * 1 kilometre resolution. The second step was a Generalised Additive Model (GAM) using 'mgcv' in R to determine the bioclimatic and landcover drivers behind woodcock abundance, conditional on presence (falling within the geographic range estimated in the first step). The GAM was then projected across 1 km squares with a minimum threshold of forest habitat (0.1 km2) within the estimated range to predict the distribution and abundance across Ireland. Finally, the predicted density from all relevant 1km squares were summed to produce a national population estimate, with standard errors.

本数据集涵盖用于复现两步法物种分布模型分析的全部数据与代码，该模型旨在估算爱尔兰共和国境内繁殖期欧亚丘鹬（Scolopax rusticola）的分布范围与种群丰度。
2017、2019及2021年的5月至6月间，志愿者于238个随机选取的林地样点中，在黄昏时段统计正进行求偶振翅（roding）行为的雄性欧亚丘鹬数量，以此获取定量的物种出现数据。
所用环境数据包含从www.worldclim.org下载的历史生物气候变量，以及由Coillte、DAFM林业司、DAERA林业服务处、CORINE土地覆盖数据集衍生得到的土地覆盖数据（涵盖林地与城镇聚落两类）。
分析分为两步：第一步基于R语言的biomod2包构建集成模型，以1×1千米分辨率的生物气候变量为输入，估算爱尔兰境内繁殖期欧亚丘鹬的地理分布范围；第二步基于R语言的mgcv包构建广义可加模型（Generalised Additive Model, GAM），以落在第一步估算的地理分布范围内的物种存在区域为条件，解析影响欧亚丘鹬种群丰度的生物气候与土地覆盖驱动因子。
随后将该GAM模型投影至估算分布范围内、且森林栖息地面积不低于0.1平方千米的1平方千米网格中，以此预测爱尔兰全境的欧亚丘鹬分布与种群丰度。最终将所有符合条件的1平方千米网格的预测密度求和，并结合标准误得到全国种群数量估算值。