An Interdisciplinary approach to Finding and Using Data for Complex Environmental Modelling Problems:A Soil System Example
An Interdisciplinary approach to Finding and Using Data for Complex Environmental Modelling Problems:A Soil System Example
复制标题
DOI：
--
复制
发表时间：
2017-09
期刊：
影响因子：
--
通讯作者：
G. Dean;Victoria Janes Bassett;R. Towe;Vatsala Nundloll;J. Davies;G. Blair
中科院分区：
文献类型：
--
作者：
G. Dean;Victoria Janes Bassett;R. Towe;Vatsala Nundloll;J. Davies;G. Blair
环境过程往往是复杂的，因为它们涉及生物、化学和物理过程之间在一系列时间和空间尺度上的非线性相互作用。土壤系统就是这种复杂性的一个例子。土壤是地质学、生物学（植物和微生物过程）、地球化学、水文学和气候之间相互作用的产物。土壤和人类系统之间的反馈带来了进一步的复杂性。土壤支撑着社会：它们提供我们绝大多数的食物，调节水的流量和质量，作为碳储存，它们对气候调节很重要。通过农业、土地使用的变化和污染，过去几个世纪以来人类的行为改变了土壤。然而，土壤的可持续性往往被政策和私营部门忽视[1]。为了理解这些复杂的系统，需要模拟模型来整合土壤形成和影响其可持续性以及我们的生态系统和社会的过程。到目前为止，土壤模型大多存在于一个领域内（例如碳循环模型、土壤水流量或质量、土壤侵蚀或形成）。我们的目标是通过将土壤生物地球化学（N14 CP，参见[2]）与侵蚀和水文过程耦合，开发更全面的土壤系统模型，以创建可用于模拟土壤功能和变化环境条件下的响应的模型。然后，该模型可以作为综合风险评估过程的一部分（洪水管理领域使用的综合风险评估方法的示例见[3]），为决策提供信息，实现可持续的土壤管理，并确保持续提供重要的土壤服务，如粮食供应和碳储存。需要来自不同来源的大量甚至巨大的数据，不仅作为模型输入，而且还用于确定初始条件、参数值以及校准和验证模型。搜索这些数据可能非常耗时，并且需要时间来进一步了解环境科学。标准的数据搜索方法往往是不成功的，并错过了有用的数据源。我们感兴趣的语义Web技术的潜力，使更有效的发现和查询的数据源。本文讨论了如何管理与这种大规模综合建模实验相关的数据的重要和研究不足的领域。特别是，我们正在采用跨学科的方法来确定数字技术如何为环境科学做出贡献。本文的总体目标是评估语义网技术在满足复杂环境建模需求方面的潜在有效性。
Environmental processes are often complex, as they involve non-linear interactions between biological, chemical and physical processes across a range of temporal and spatial scales. Soil systems are one example of this complexity. Soils emerge from the interaction between geology, biology (plant and microbial processes), biogeochemistry, hydrology, and climate. Further complexity arises from the feedbacks between soil and human systems. Soils underpin society: they provide the vast majority of our food, the regulate water flows and quality, as a carbon store they are important for climate regulation. Through agriculture, land use change and pollution, human actions over the last have modified soils for centuries. Yet soil sustainability is often overlooked in policy and private sectors [1]. To understand these complex systems, simulation models are needed to integrate the processes that shape soils and influence their sustainability and that of our ecosystems and societies. To date, much of the modelling of soils has existed within one domain (e.g. either carbon cycle models, soil water flow or quality, soil erosion or formation). We aim to develop a more comprehensive model of the soil system by coupling soil biogeochemistry (N14CP, see [2]) with erosion and hydrological processes, to create a model that can be used to simulate soil functioning and response under changing environmental conditions. The model can then be used as part of an integrated risk assessment process (for an example of an integrated risk assessment approach used in the flood management domain see [3]) to inform decision making, enabling sustainable soil management and ensuring continued provision of vital soil services such as food provision and carbon storage. Large, or even vast, amounts of data from diverse sources are required to not only act as model input, but also to determine initial conditions, parameter values and to calibrate and validate models. Searching for this data can be incredibly time consuming and takes time away from furthering our understanding within environmental science. Standard methods of data searching can often be unsuccessful and miss sources of useful data. We are interested in the potential of semantic web technologies to enable more efficient discovery and querying of data sources. This paper addresses the important and under researched area of how to manage data associated with such large-scale integrated modelling experiments. In particular, we are adopting an interdisciplinary approach to determine how digital technologies might contribute to environmental science. The overall goal of the paper is to evaluate the potential effectiveness of semantic web technology in addressing the needs of complex environmental modelling.