Workshop: The subterranean macroscope: sensor networks for understanding, modeling, and managing soil processes (University of Chicago-Hyde Park, Illinois - October 2017)

科学对地球上生命最重要的组成部分之一知之甚少：地球地下区域土壤的物理、生物和化学性质，地下区域通常达到地表以下0.6至3米。土壤提供食物、纤维和淡水，对能源和气候的可持续性作出重大贡献，并有助于维持生物多样性和全面保护生态系统。科学家们缺乏对土壤和生长在土壤中的植物的准确理解和预测模型，因为很难绘制出大片土地上土壤随时间变化的生物和地球化学特性的高分辨率实验图。然而，传感器网络的出现、使用纳米技术的更好的传感和数据分析有可能很快使这种高分辨率的实验性土壤制图能力触手可及。本次研讨会的目标是将工程、计算机科学、土壤和植物科学界聚集在一起，为地下无线传感器网络开发一个愿景，该网络将能够创建这样的地下土壤地图，然后使用这些数据开发下一代土壤和植物模型。除了拓展基础科学的边界之外，这些网络还将对粮食安全和我们管理环境和生态系统的方式产生重大影响。科学家对渗透带土壤的物理、化学和生物转化与循环及其对植物科学和粮食安全的影响了解甚少。由于缺乏足够的高时空尺度土壤数据，目前的模型是不充分的。然而，传感器和纳米技术、传感器网络、通信和微电子技术以及数据分析方面的革命性进步，将使可扩展和负担得起的地下传感网络成为可能，进而可能彻底改变土壤、植物和生态系统科学。这种进步将对环境、粮食安全和粮食管理产生重大影响。基于最近在各个科学和工程领域取得的成功，本次研讨会汇集了土壤科学研究人员（包括土壤生物，化学和物理性质方面的专家），动态土壤建模专业知识，植物科学，传感器网络，微电子和无线研究人员以及机器学习/数据分析。研讨会的目标是为如何在不同的地理尺度上建立这样一个地下传感器网络创建一个愿景和框架，这些传感器将产生密集、有用的数据，为土壤科学、植物科学和建模工作提供信息。这些努力反过来将使人们进一步了解土壤的物理、化学和生物性质及其对植物科学和粮食安全的影响。讨论将围绕以NSF为基础的三个关键主题展开。年代?伟大的想法?(1)融合，将上述多学科社区聚集在一起；(2)大数据（Big Data），即对长期密集的原位土壤数据进行高效传输、管理和分析；(3)理解生命的规则，通过测量与基因型相互作用的土壤环境条件（gx E相互作用）的进展，预测不同环境下基因型的表型。研讨会还考虑了技术和数据如何最好地与计算模型相结合，从而为现实世界的决策提供信息。研讨会的长期影响可能包括促进利用前所未有的时空、多模式数据，这些数据可以彻底改变我们对土壤的理解，开发比目前更准确的土壤和植物模型，并对环境可持续性、农业产量、气候模型、水和农业管理以及全球粮食安全产生深远影响。

Science has a poor understanding of one of the most important components of life on earth: the physical, biological and chemical nature of the soil in the subsurface region of the earth that typically reaches to 0.6 to three meters below the surface. Soils provide food, fiber and fresh water, make major contributions to energy and climate sustainability, and help maintain biodiversity and the overall protection of ecosystem. Scientists lack an accurate understanding and predictive models for soil and for the plants that grow in it, because it has been difficult to create high resolution experimental maps of the biological and geochemical properties of the soil over large tracts of land over time. However, the emergence of sensor networks, better sensing using nanotechnology, and data analytics has the potential soon to bring such high resolution experimental soil mapping capabilities within reach. The goal of this workshop is to bring together the engineering, computer science, and soil and plant science communities to develop a vision for underground wireless sensor networks that will enable the creation of such subterranean soil maps, followed by using these data to develop the next generation of soil and plant models. Besides furthering the boundaries of fundamental science, these networks would have a major impact on food security and the way we manage the environment and ecosystems. Scientists have a poor understanding of the physical, chemical and biological transformations and cycling of soil in the vadose zone and its influences on plant science and food security. Lacking adequate soil data at high spatial and temporal length scales, current models are inadequate. However, revolutionary advances in sensors and nanotechnology, sensor networks, communications and microelectronics technologies, and data analytics are poised to enable scalable and affordable subterranean sensing networks that, in turn, can potentially revolutionize soil, plant, and ecosystem sciences. Such advances would have a major impact on the environment, food security and its management. Building on recent successes across various science and engineering fields, this workshop brings together researchers in soil science (including experts in the biological, chemical and physical nature of soil), dynamic soil modeling expertise, plant sciences, sensor networks, microelectronics and wireless researchers, and machine learning/data analytics. The goal of the workshop is to create a vision and framework for how such a subterranean sensor network could be built across different geographical scales, with sensors that will generate dense, useful data that will inform soil science, plant science and modeling efforts. These efforts, in turn would lead to the next level of understanding of the physical, chemical and biological nature of soil, and its impact on plant science and food security. The discussion will be framed by three key topics that build upon NSF?s ?Big Ideas?: (1) Convergence, bringing together the aforementioned multidisciplinary communities; (2) Big Data, the efficient transmission, curation, and analysis of dense, in situ soil data over time; and (3) Understanding the Rules of Life, predicting phenotype from genotypes in diverse environments through advances in measuring soil environmental conditions that interact with genotype (G x E interactions). The workshop also considers how technology and data can best interface with computational models to inform real-world decisions. Longer-term impacts of the workshop may include facilitating the use of unprecedented levels of spatial and temporal, multi-modal data that can revolutionize our understanding of soils, developing more accurate soil and plant models than are available today, and enabling a profound impact on environmental sustainability, agricultural yields, climate models, water and agricultural management, and global food security.