Application of novel atmospheric measurement technology to wider environmental monitoring

下一代气溶胶检测，实现更准确的环境和健康监测：应用新型大气测量技术，发展气溶胶特性的最新技术，更好地跟踪污染和空气传播疾病。空气中的颗粒物通过各种来源排放和传播，覆盖从室内到全球的所有地理范围。我们能够识别气溶胶的来源并模拟它们如何传输，这对于确定它们可能的影响和实施缓解措施以保护人类，动物和植物健康至关重要。为了实现这一目标，需要更好地识别和识别各种环境中的颗粒类型：城市，农业，工业，临床，室内。然而，现有的实时监测技术并不能提供足够的信息作为采取有效行动的依据。虽然用于空气质量监测的光学粒子计数器可以确定气溶胶的大小，但它们无法区分气溶胶的形状和类型，例如柴油液滴和刹车灰尘之间的差异，或者致病真菌孢子和来自田间的简单灰尘之间的差异。它们也无法确定源解析，例如，急性污染事件的主要原因是交通、工厂还是天气。实时检测技术仅限于低成本的OPC或昂贵的研究级仪器，只能识别特定的气溶胶类型。我们的长期愿景是设计一种新型的监测系统，将机器学习技术与最初为对大气冰进行分类而开发的光散射技术相结合，不仅可以通过大小，而且更重要的是，还可以通过形状和表面结构来区分气溶胶。我们相信，这种能力将允许使用一个单一的仪器来检测广泛的空气污染物和病原体，并确定特定气溶胶流入的来源，它可以传输多远及其对环境或健康的影响。

Next-generation aerosol detection for more accurate environment and health monitoring: Applying novel atmospheric measurement techniques to progress the state of the art in the characterisation of aerosols to better track pollution and airborne disease transmission.Airborne particulates are emitted and transmitted by a wide range of sources and over all geographic scales, from indoors to globally. Our ability to identify the source of aerosols and model how they are transported is critical to both determining their likely impacts and implementing mitigation measures to protect human, animal and plant health. To achieve this, there is a need to better identify and characterise the types of particle across a range of environments: urban, agricultural, industrial, clinical, indoor. However, existing real-time monitoring techniques do not provide sufficient information on which to base effective action. While optical particle counters, used for air quality monitoring, can determine aerosol size, they are unable to differentiate between aerosol shape and therefore type, for example the difference between diesel droplets and brake dust, or pathogenic fungal spores and simply dust from a field. Nor can they identify source apportionment, for example whether the principal cause of an acute pollution episode is traffic, a factory or the weather.Real-time detection technologies are limited to low-cost OPCs or expensive research-grade instruments that only identify a specific aerosol type. Our long-term vision is to engineer a novel monitoring system, by combining machine learning techniques with light-scattering technologies originally developed to classify atmospheric ice, that discriminates aerosols by not just size but, crucially, by shape and surface structure too. We believe this capability will allow the use of a single instrument to detect a broad range of airborne pollutants and pathogens, and identify the source of an influx of a particular aerosol, how far it may be transported and its environmental or health impact.