Patterns and process in population trends of UK herpetofauna

由于无法对许多神秘类群的全国状况、趋势和威胁做出可靠的估计，对生物多样性的可靠估计受到限制。因此，有效的决策需要(1)对生物多样性进行更好的平衡评估；(2)整理和分析使用各种协议和利益相关者小组收集的“杂乱无章”的物种数据。有证据表明，英国许多两栖和爬行动物(或爬行动物)的数量减少，特别是在以前分布广泛的物种，如毒蛇和常见的蟾蜍。国家两栖和爬行动物记录计划(NARRS)旨在为英国两栖动物的趋势提供证据，但一直无法对这些趋势的驱动因素提供实质性的见解。幸运的是，现在已经出现了能够从非结构化数据(包括公民科学记录)中提供强劲趋势的统计工具。该项目将把来自一系列组织的现有数据集整合到一个通用的建模框架中，基于分层的贝叶斯占有率检测模型。这些产出将用于：1.根据不同的数据(包括NARR)揭示国家状况和趋势2.确定分布和丰度趋势的驱动因素3.预测未来变化情景下的趋势4.综合监测组合的设计建议学生将获得全面的培训经验，涵盖广泛的分析技能，包括使用公民科学数据进行时空建模的贝叶斯统计技术，以及可转让的技能，包括利益攸关方参与和知识交流。

Reliable estimates of biodiversity are constrained by an inability to produce robust estimates of nationwide status, trends and threats for many cryptic taxa. Effective decision-making therefore requires (1) a better balanced assessment of biodiversity; and (2) collation and analysis of 'messy' species data that are collected using a variety of protocols and stakeholder groups.Anecdotal evidence points to declines in many of the UK's amphibians and reptiles (or herpetofauna), particularly in formerly widespread species such as the adder and common toad. The National Amphibian and Reptile Recording Scheme (NARRS) was designed to provide evidence on trends in UK herpetofauna, but has been unable to deliver substantive insights into drivers of these trends. Fortunately, statistical tools have now emerged that are capable of providing robust trends from unstructured data, including citizen science records.This project will bring together existing datasets from a range of organisations into a common modelling framework, based on a hierarchical Bayesian occupancy-detection model. These outputs will be used to:1. Reveal national status and trends from disparate data (including NARRS)2. Determine the drivers of trends in distribution and abundance3. Forecast trends under scenarios of future change4. Advise on the design of an integrated monitoring portfolioThe student will receive a comprehensive training experience, covering a broad range of analytical skills including Bayesian statistical techniques for spatio-temporal modelling using citizen science data, as well as transferable skills including stakeholder engagement and knowledge exchange.