Frameworks: Collaborative Proposal: Software Infrastructure for Transformative Urban Sustainability Research

框架：合作提案：变革性城市可持续发展研究的软件基础设施

• 批准号: 1931363
• 负责人: Shrideep Pallickara
• 金额: $ 200.03万
• 依托单位: Colorado State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1931363&HistoricalAwards=false
• 关键词: Frameworks; Collaborative; Proposal; Software; Infrastructure

The United States is highly urbanized with more than 80% of the population residing in cities. Cities draw from and impact natural resources and ecosystems while utilizing vast, expensive infrastructures to meet economic, social, and environmental needs. The National Science Foundation has invested in several strategic research efforts in the area of urban sustainability, all of which generate, collect, and manage large volumes of spatiotemporal data. Voluminous datasets are also made available in domains such as climate, ecology, health, and census. These data can spur exploration of new questions and hypotheses, particularly across traditionally disparate disciplines, and offer unprecedented opportunities for discovery and innovation. However, the data are encoded in diverse formats and managed using a multiplicity of data management frameworks -- all contributing to a break-down of the observational space that inhibits discovery. A scientist must reconcile not only the encoding and storage frameworks, but also negotiate authorizations to access the data. A consequence is that data are locked in institutional silos, each of which represents only a sliver of the observational space. This project, SUSTAIN (Software for Urban Sustainability to Tailor Analyses over Interconnected Networks), facilitates and accelerates discovery by significantly alleviating data-induced inefficiencies. This effort has deep, far-reaching impact. It transforms urban sustainability science by establishing a community of interdisciplinary researchers and catalyzing their collaborative capacity. Hundreds of researchers from over 150 universities are members of our collaborating organizations and will immediately benefit from SUSTAIN. Domains where spatiotemporal phenomena must be analyzed benefit from this innovative research; the partnership with ESRI and Google Earth amplify the impact of SUSTAIN, giving the project a global reach and enabling international collaborative initiatives. The direct engagement with middle school students in computer science and STEM disciplines has well-known benefits and, combined with graduate training, produces a diverse, globally competitive STEM workforce. SUSTAIN targets transformational capabilities for feature space exploration, hypotheses formulation, and model creation and validation over voluminous, high-dimensional spatiotemporal data. These capabilities are deeply aligned with the urban sustainability community's needs, and they address challenges that preclude effective research. SUSTAIN accomplishes these interconnected goals by enabling holistic visibility of the observational space, interactive visualizations of multidimensional information spaces using overlays, fast evaluation of expressive queries tailored to the needs of the discovery process, generation of custom exploratory datasets, and interoperation with diverse analyses software frameworks - all leading to better science. SUSTAIN fosters deep explorations through its transformative visibility of the federated information space. The project reconciles the fragmentation and diversity of siloed data to provide seamless, unprecedented visibility of the information space. A novel aspect of the project's methodology is the innovative use of the Synopsis, a spatiotemporal sketching algorithm that decouples data and information. The methodology extracts and organizes information from the data and uses the information (or sketches of the data) as the basis for explorations. The project also incorporates a novel algorithm for imputations at the sketch level at myriad spatiotemporal scopes. The effort creates a collaborative community of multidisciplinary researchers to build an enduring software infrastructure for urban sustainability.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

美国是高度城市化的国家，80%以上的人口居住在城市。城市利用并影响着自然资源和生态系统，同时利用庞大而昂贵的基础设施来满足经济、社会和环境需求。美国国家科学基金会在城市可持续发展领域投资了几项战略研究，所有这些研究都产生、收集和管理了大量的时空数据。在气候、生态、健康和人口普查等领域也提供了大量数据集。这些数据可以刺激对新问题和新假设的探索，特别是在传统上不同的学科之间，并为发现和创新提供前所未有的机会。然而，数据以不同的格式编码，并使用多种数据管理框架进行管理——所有这些都导致了观测空间的破坏，从而阻碍了发现。科学家不仅要协调编码和存储框架，还要协商访问数据的授权。其结果是，数据被锁定在机构孤岛中，每个孤岛只代表了观测空间的一小部分。该项目名为SUSTAIN（通过互联网络定制城市可持续性分析的软件），通过显著缓解数据导致的低效率，促进并加速了发现。这一努力具有深远的影响。它通过建立跨学科研究人员的社区和促进他们的合作能力来改变城市可持续发展科学。来自150多所大学的数百名研究人员是我们合作组织的成员，他们将立即受益于SUSTAIN。必须分析时空现象的领域受益于这一创新研究；与ESRI和谷歌Earth的合作扩大了SUSTAIN的影响，使该项目具有全球影响力，并使国际合作倡议成为可能。与中学生直接接触计算机科学和STEM学科具有众所周知的好处，并且与研究生培训相结合，可以培养出多样化的，具有全球竞争力的STEM劳动力。SUSTAIN的目标是对大量高维时空数据进行特征空间探索、假设制定、模型创建和验证的转换能力。这些能力与城市可持续发展社区的需求密切相关，它们解决了阻碍有效研究的挑战。SUSTAIN通过实现观测空间的整体可见性、使用覆盖层对多维信息空间进行交互式可视化、根据发现过程的需要对表达性查询进行快速评估、生成自定义探索性数据集以及与各种分析软件框架进行互操作来实现这些相互关联的目标——所有这些都有助于更好的科学研究。SUSTAIN通过其对联合信息空间的变革性可见性促进了深入的探索。该项目协调了孤立数据的碎片性和多样性，提供了无缝的、前所未有的信息空间可见性。该项目方法论的一个新颖方面是对synosis的创新使用，这是一种将数据和信息解耦的时空素描算法。该方法从数据中提取和组织信息，并使用信息（或数据草图）作为探索的基础。该项目还采用了一种新颖的算法，用于在无数时空范围内的草图水平上进行imputation。该项目创建了一个多学科研究人员的协作社区，为城市可持续发展建立一个持久的软件基础设施。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Concerto: Leveraging Ensembles for Timely, Accurate Model Training Over Voluminous Datasets

• DOI: 10.1109/bdcat50828.2020.00024
• 发表时间: 2020-12
• 期刊: 2020 IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT)
• 作者: Walid Budgaga;Matthew Malensek;S. Pallickara;S. Pallickara

Lightweight, Embeddings Based Storage and Model Construction Over Satellite Data Collections

• DOI: 10.1109/bigdata50022.2020.9377764
• 发表时间: 2020-12
• 期刊: 2020 IEEE International Conference on Big Data (Big Data)
• 作者: Kevin Bruhwiler;Paahuni Khandelwal;Daniel Rammer;Samuel Armstrong;S. Pallickara;S. Pallickara

Attention-based convolutional capsules for evapotranspiration estimation at scale

基于注意力的卷积胶囊用于大规模蒸散发估计

• DOI: 10.1016/j.envsoft.2022.105366
• 发表时间: 2022
• 期刊: Environmental Modelling & Software
• 影响因子: 4.9
• 作者: Armstrong, Samuel;Khandelwal, Paahuni;Padalia, Dhruv;Senay, Gabriel;Schulte, Darin;Andales, Allan;Breidt, F. Jay;Pallickara, Shrideep;Pallickara, Sangmi Lee
• 通讯作者: Pallickara, Sangmi Lee

Infrastructure autopoiesis: requisite variety to engage complexity

基础设施自创生：应对复杂性的必要多样性

• DOI: 10.1088/2634-4505/ac4b48
• 发表时间: 2022
• 期刊: Environmental Research: Infrastructure and Sustainability
• 作者: Chester, Mikhail V.;Allenby, Braden
• 通讯作者: Allenby, Braden

A synthetic water distribution network model for urban resilience

城市复原力的综合配水网络模型

• DOI: 10.1080/23789689.2020.1788230
• 发表时间: 2020
• 期刊: Sustainable and Resilient Infrastructure
• 影响因子: 5.9
• 作者: Ahmad, Nasir;Chester, Mikhail;Bondank, Emily;Arabi, Mazdak;Johnson, Nathan;Ruddell, Benjamin L.
• 通讯作者: Ruddell, Benjamin L.