IDR-Collaborative Research: The impact of green infrastructure on urban microclimate and energy use

1134580/1133590 Pardyjak/Willemsen将使用大型模拟科学来调查绿色基础设施项目对城市能源使用和小气候的影响。绿色基础设施项目有多种形式，包括开发公园、改造建筑屋顶，以及在街道和停车场使用新型沥青和混凝土材料。它们都有共同的目标，即减少能源消耗、减少污染排放和改善城市小气候。由于很难模拟长度尺度(m到公里)的巨大差异，人们对它们的影响知之甚少。一个跨学科团队将利用一套基于计算的策略来弥合这些尺度，并提高对绿色基础设施如何在地方(邻里)、城市和中尺度上与城市环境相互作用的理解。具体地说，将研究如何影响热量、水分和污染物的分布。PI假设可以使用大规模模拟科学来寻找最优的绿色基础设施设计。将调查城市形态和绿色基础设施之间的复杂相互作用，并制定指导未来项目的战略。为了充分解决支配热量、水蒸气和污染物在大范围尺度上分布的基本传输过程，将需要进行千万亿级计算。将在能够利用最大计算平台的大规模并行计算框架(Uintah)内实施新的城市流体动力学和大规模交通模拟代码。该模拟工具将与中尺度模拟相结合，以产生具有实际大气强迫条件的城市核心的高分辨率(约1米)模拟。中尺度模拟还将与运行速度极快的基于GPU的城市小气候和弥散工具相结合，以根据不同的标准(如能源使用、空间)优化绿色基础设施项目的设计。目标模拟场景是2003年城市联合战役期间俄克拉荷马市的一个完整的日循环，具有高保真的物理效果，在1米到1000 S公里的长度尺度上分辨物理效果。这一目标场景利用了在实地试验期间从俄克拉荷马城获得的大量数据集，从而允许进行详细的验证。这些模拟工具将帮助城市规划者为设计和实施绿色基础设施项目制定有用和独特的战略。为了确保这一点，将在整个模型开发过程中继续与城市规划者合作。来自高分辨率模拟的数据将以档案形式提供给其他致力于城市气象学应用的研究人员。这些数据将跨越史无前例的范围，并详细描述物理过程。这些数据预计将对本项目范围之外的广泛模型开发和理论工作有用。此外，在这两所合作的大学中，还包括一个重要的外展部分，旨在向美国印第安人、阿拉斯加原住民和其他少数民族介绍模拟科学和环境工程。学生将被邀请参加为期一周的互动学习研讨会，在项目的三年中的每一年。此外，该项目将通过合作研究活动和参与犹他大学全球变化与生态系统中心，为研究生和博士后研究人员提供大气、工程、计算机和社会科学方面的跨学科培训。

1134580/1133590 Pardyjak/WillemsenLarge-scale simulation science will be used to investigate the impact of green infrastructure projects on urban energy use and microclimate. Green infrastructure projects come in many forms including the development of parks, alteration of building rooftops, and the use of novel asphalt and concrete materials for streets and parking lots. They all share the common goals of reducing energy usage, mitigating pollution emissions and improving the urban microclimate. Due to difficulty in simulating the large disparity in length scales (m to km), little is known about their impact. An interdisciplinary team will utilize a suite of computationally based strategies to bridge these scales and improve understanding of how green infrastructure interacts with the urban environment at local (neighborhood), city, and meso- scales. Specifically, how the distribution of heat, moisture and pollutants can be effected will be investigated. The PIs hypothesize that large-scale simulation science can be used to find optimal green infrastructure designs. The complex interaction between urban form and green infrastructure will be investigated and strategies to guide future projects will be developed. To adequately resolve the fundamental transport processes that govern the distribution of heat, water vapor and pollutants across a wide range of scales will require petascale computing. A new urban fluid dynamics and mass transport simulation code inside of a massively parallel computational framework (Uintah) capable of taking advantage of the largest computing platforms will be implemented. This simulation tool will be combined with mesoscale simulations to produce high-resolution (on the order of 1 m) simulations of the urban core with realistic atmospheric forcing conditions. Mesoscale simulations will also be combined with an extremely fast-running GPU based urban microclimate and dispersion tool to optimize the design of green infrastructure projects based on different criteria (e.g., energy usage, space). The target simulation scenario is a full diurnal cycle of Oklahoma City during the Joint Urban 2003 field campaign, with high fidelity physics, resolving physical effects at length scales ranging from ~1m to 1000?s km. This target scenario takes advantage of the extensive datasets from Oklahoma City taken during the field experiment allowing for detailed validation. These simulation tools will aid urban planners in developing useful and unique strategies for the designing and implementation of green infrastructure projects. To ensure this, work with urban planners will continue throughout the model development process. The data from the high-resolution simulations will be made available in an archival form to other researchers working on urban meteorology applications. These data will span an unprecedented range of scales and have a detailed representation of the physical processes. The data are anticipated to be useful for a wide range of model development and theoretical work outside the scope of this project. In addition, included is a substantial outreach component at both collaborating universities designed to introduce American Indians, Alaskan Natives, and other minorities to simulation science and environmental engineering. Students will be invited to a weeklong interactive learning symposium during each of the three years of the project. In addition, this project will provide interdisciplinary training in the atmospheric, engineering, computer and social sciences for graduate students and post-doctoral researchers through collaborative research activities and from involvement in the Global Change and Ecosystem Center at the University of Utah.