ClearSky: cloud-free monitoring of UK agriculture

ClearSky：英国农业的无云监控

• 批准号: ST/V002252/1
• 负责人: James Geach
• 金额: $ 21.37万
• 依托单位: University of Hertfordshire
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2021
• 资助国家: 英国
• 起止时间: 2021 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=ST%2FV002252%2F1
• 关键词: ClearSky; cloud; free; monitoring; UK

Earth Observation satellites can monitor every part of the globe, offering a detailed view of the Earth's surface on a daily basis. This is a valuable source of intelligence for the agricultural industry, since one can - in theory - map all farmland to determine, for example, the rate of growth of crops, signatures of crop stress and disease, the presence of flooding or signs of drought. These factors are vital to the efficient production of food and maintaining food security. On one hand, a precise view of crop health allows farmers to optimise the use of fertiliser and water, and the rapid identification of threats allows food producers to respond in a timely manner to mitigate their impact on the food supply.There is a problem: cloud cover. Clouds obscure the view of Earth Observation satellites in the visible and infrared spectral range, and this is where the vast majority of remote sensing analysis is performed. This limits the value of imagery, because if it is cloudy when the satellite passes over, the intelligence is lost. Our solution is to exploit radar data. Satellites equipped with radar can reflect radio waves off the ground and detect their reflectance. Since radio waves are not affected by cloud, this offers an uninterrupted and reliable view of the ground. However, the challenge has been in the interpretation of radar data. We have solved this challenge by developing an algorithm that can accurately predict what would have been seen in the traditional V/IR images (e.g. the RGB image our eyes would see) for cases where the surface is obscured by cloud. We call our algorithm ClearSky.In this project we will use ClearSky to monitor every single field in the UK, every week. Our predictions will allow us to assess the presence and density of vegetation and its rate of growth, as well as determining local threats such as flooding and the onset of drought. Our aim is to deliver this intelligence direct to farmers and the food supply industry in order to aid the decision making process with regards the most efficient use of resources in producing food, and enabling us to be agile to potential threats to food security associated with a changing climate.

地球观测卫星可以监测地球的每一个部分，每天提供地球表面的详细视图。这对农业行业来说是一个宝贵的情报来源，因为从理论上讲，人们可以绘制所有农田的地图，以确定例如农作物的生长速度、作物压力和疾病的特征、洪水的存在或干旱的迹象。这些因素对于高效生产粮食和维护粮食安全至关重要。一方面，对作物健康的准确看法使农民能够优化化肥和水的使用，而快速识别威胁使粮食生产者能够及时做出反应，以减轻他们对粮食供应的影响。在可见光和红外光谱范围内，云层遮挡了地球观测卫星的视线，这是进行绝大多数遥感分析的地方。这限制了图像的价值，因为如果卫星经过时天气多云，情报就会丢失。我们的解决方案是利用雷达数据。配备雷达的卫星可以反射离地面的无线电波，并检测它们的反射率。由于无线电波不受云层的影响，这提供了一个不间断和可靠的地面图像。然而，挑战一直是在雷达数据的解释方面。我们已经解决了这一挑战，开发了一种算法，可以准确地预测在表面被云遮挡的情况下，传统V/IR图像(例如，我们的眼睛将看到的RGB图像)中会看到什么。我们称我们的算法为ClearSky。在这个项目中，我们将使用ClearSky每周监控英国的每一个领域。我们的预测将使我们能够评估植被的存在和密度及其生长速度，并确定当地的威胁，如洪水和干旱的开始。我们的目标是将这一情报直接提供给农民和粮食供应行业，以帮助决策进程，最有效地利用资源生产粮食，并使我们能够灵活应对与气候变化相关的粮食安全潜在威胁。