MCA: Career Advancement in Polar Cyberinfrastructure: Permafrost Feature Mapping and Change Detection using Geospatial Artificial Intelligence and Remote Sensing

MCA：极地网络基础设施的职业发展：使用地理空间人工智能和遥感进行永久冻土特征映射和变化检测

• 批准号: 2120943
• 负责人: Wenwen Li
• 金额: $ 35.98万
• 依托单位: Arizona State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-09-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2120943&HistoricalAwards=false
• 关键词: MCA; Career; Advancement; Polar; Cyberinfrastructure

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).Polar regions play a vital role in Earth’s climate, ecosystems, and economy. Unfortunately, climate change is driving dramatic changes in the Arctic ecosystem, endangering its natural environment, infrastructure, and lives. Arctic permafrost, ground that remains below 0°C for at least two consecutive summers, is at the center of this change. Covering nearly 1/4 of the land in the northern hemisphere, thawing permafrost is causing a significant local and regional impact in the Arctic. Severe impacts include land subsidence resulting in costly damage to the built environment and increased release of greenhouse gases which further exaggerates the greenhouse effect and global warming. To improve our understanding of permafrost dynamics and its linkages to other Arctic ecosystem components in the midst of rapid Arctic change, it is critically important to have geospatial data readily available that provide high-resolution mapping of permafrost features, their geographical extent, distribution, and change. Although a coarse classification of pan-Arctic permafrost has been developed, fine granularity, local to regional-scale mapping of major permafrost features, is largely unavailable. This data gap inevitably constrains us from gaining a holistic view of the space-time dynamics of permafrost degradation across the Arctic. The goal of this project is to bridge this existing data gap by developing new analytical solutions to support intelligent and automated delineation of permafrost features at scale.Through a partnership with colleagues at Woodwell Climate Research Center, this project will explore novel ways to deepen the integration of cutting-edge AI, geospatial analysis, and cyberinfrastructure into Arctic permafrost research. Specifically, novel GeoAI (Geospatial Artificial Intelligence) solutions will be developed to empower the ongoing efforts of AI-based, high-resolution mapping of pan-Arctic permafrost thaw from Big Imagery. By enabling location-aware and multi-source deep learning and the integration of key spatial principles (i.e., spatial dependency and spatial autocorrelation), the proposed GeoAI model will create polar data products with high veracity and automation, thereby accelerating the scientific navigation of the New Arctic. A joint initiative, “Women in Polar Cyberinfrastructure,” will broaden the participation of women and underrepresented minorities in Arctic AI research. It will also serve as an important avenue for openly sharing knowledge and resources and provide mentorship to early-career scholars in Arctic science, GeoAI, and cyberinfrastructure. All datasets and tools produced in this project will be open-sourced and made available in the NSF Permafrost Discovery Gateway to increase their reuse and inspire further innovation.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.

该奖项全部或部分由2021年美国救援计划法案（公法117-2）资助。极地地区在地球的气候，生态系统和经济中发挥着至关重要的作用。不幸的是，气候变化正在推动北极生态系统的急剧变化，危及其自然环境，基础设施和生命。北极永久冻土，至少连续两个夏天保持在0°C以下的地面，是这种变化的中心。永久冻土融化覆盖了北方近四分之一的土地，正在对北极地区造成重大的局部和区域影响。严重的影响包括地面沉降，对建筑环境造成代价高昂的破坏，以及温室气体排放增加，进一步加剧温室效应和全球变暖。在北极快速变化的过程中，为了提高我们对永久冻土动态及其与其他北极生态系统组成部分的联系的理解，至关重要的是要随时提供地理空间数据，以提供永久冻土特征、地理范围、分布和变化的高分辨率地图。虽然泛北极永久冻土的粗分类已经开发，细粒度，主要永久冻土特征的局部区域尺度映射，在很大程度上是不可用的。这一数据缺口不可避免地限制了我们对整个北极地区永久冻土退化的时空动态的整体看法。该项目的目标是通过开发新的分析解决方案来弥合现有的数据差距，以支持大规模的永久冻土特征的智能和自动化描绘。通过与伍德韦尔气候研究中心的同事合作，该项目将探索新的方法，以深化尖端人工智能，地理空间分析和网络基础设施与北极永久冻土研究的整合。具体而言，将开发新的GeoAI（地理空间人工智能）解决方案，以支持正在进行的基于AI的高分辨率泛北极永久冻土融化测绘工作。通过实现位置感知和多源深度学习以及关键空间原则（即，空间依赖性和空间自相关性），拟议的GeoAI模型将创建具有高准确性和自动化的极地数据产品，从而加速新北极的科学导航。一项名为“极地网络基础设施中的女性”的联合倡议将扩大女性和代表性不足的少数民族在北极人工智能研究中的参与。它还将成为公开分享知识和资源的重要途径，并为北极科学，GeoAI和网络基础设施的早期职业学者提供指导。该项目中产生的所有数据集和工具都将开源，并在NSF Permafrost Discovery Gateway中提供，以提高其重用性并激发进一步的创新。该奖项反映了NSF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

GeoImageNet: a multi-source natural feature benchmark dataset for GeoAI and supervised machine learning

• DOI: 10.1007/s10707-022-00476-z
• 发表时间: 2022-09
• 期刊: GeoInformatica
• 影响因子: 2
• 作者: Wenwen Li;Sizhe Wang;S. Arundel;Chia-Yu Hsu

GeoAI for Large-Scale Image Analysis and Machine Vision: Recent Progress of Artificial Intelligence in Geography

• DOI: 10.3390/ijgi11070385
• 发表时间: 2022-07
• 期刊: ISPRS Int. J. Geo Inf.
• 作者: Wenwen Li;Chia-Yu Hsu

Explainable GeoAI: can saliency maps help interpret artificial intelligence’s learning process? An empirical study on natural feature detection

可解释的 GeoAI：显着图可以帮助解释人工智能的学习过程吗？

• 发表时间: 2022
• 期刊: International journal of geographical information science
• 影响因子: 5.7
• 作者: Hsu, C.Y.;Li, W.
• 通讯作者: Li, W.

Replication across space and time must be weak in the social and environmental sciences

• DOI: 10.1073/pnas.2015759118
• 发表时间: 2021-08-25
• 期刊: PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
• 影响因子: 11.1
• 作者: Goodchild, Michael F.;Li, Wenwen
• 通讯作者: Li, Wenwen