The improvement of climate change investigations by developing and applying innovative evolutionary subset time series modelling using semi-parametric sparse-patterned approaches

通过使用半参数稀疏模式方法开发和应用创新的进化子集时间序列建模来改进气候变化调查

• 批准号: DP1092801
• 负责人: Prof Terence O'Neill
• 金额: $ 16.09万
• 依托单位: Bond University
• 依托单位国家: 澳大利亚
• 项目类别: Discovery Projects
• 财政年份: 2011
• 资助国家: 澳大利亚
• 起止时间: 2011-01-31 至 2016-12-31
• 项目状态: 已结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/DP1092801
• 关键词: improvement; climate; change; investigations; developing

With an estimated US$6.98 trillion loss indicated in the Stern review, severe climate change will make world climate conditions harsher and more likely include large natural climate disasters. The health of the Australian economy is critically dependent on decisions of environmental managers. However, most problems of complexity arising in climate change involve issues on which we do not possess a deep understanding. This project draws upon a set of inter-disciplinary concepts and models centred in neural networks that enable us to advance our understanding of complexity, leading to superior quantitative tools and models to allow for improved environmental decision-making.

《斯特恩报告》估计造成6.98万亿美元的损失，严重的气候变化将使世界气候条件更加恶劣，更有可能包括大规模的自然气候灾害。澳大利亚经济的健康发展严重依赖于环境管理者的决策。然而，气候变化所产生的大多数复杂问题涉及我们没有深刻理解的问题。该项目借鉴了一套以神经网络为中心的跨学科概念和模型，使我们能够推进对复杂性的理解，从而产生上级定量工具和模型，以改进环境决策。