In-situ spectroscopic characterization of soil properties for precision agriculture**

精准农业土壤特性的原位光谱表征**

• 批准号: 533710-2018
• 负责人: Biswas, Asim
• 金额: $ 1.82万
• 依托单位: University of Guelph
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2018
• 资助国家: 加拿大
• 起止时间: 2018-01-01 至 2019-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=645547
• 关键词: situ; spectroscopic; characterization; soil; properties

SoilReader®, a precision agriculture equipment manufacturing company, is revolutionizing soil analysis techniques with highly precise and in-situ measurements of multiple soil constituents at depths in real-time utilizing the principle of soil spectroscopy. SoilReader, a soil spectrometer attached to a disc, records real-time soil spectral response at multiple depths (within top 9-in) as it moves through the field with high-resolution analysis (~3500 samples/acre) of multiple (9) soil constituents providing solutions for fast and non-destructive soil characterization for precision management of crop inputs. However, during each rotation of the SoilReader disc, the spectrometer records spectral response continuously both in dark environment within the soil and the bright environment outside of soil. This creates difficulty in processing the soil spectra and the prediction of soil properties. The joint venture between the University of Guelph and SoilReader aims to develop strategies to separate the spectra and improve prediction of soil properties. More specific questions that this project aims to answer are 1) how well spectroscopic measurements can predict multiple soil properties in the field (in-situ) conditions? and 2) how can we develop a multivariate statistical model to process above/below ground spectral responses for improved prediction of soil properties? The research, the technique, and the information generated through this project will provide an inexpensive means for soil characterization in short time for the growers, service providers, researchers, and other data users for precision agriculture services. Additionally, the mathematical model developed from this project will guide SoilReader to develop and strengthen the analytical services and will improve the equipment in future to support better crop growth and development from the optimized use of agronomic input while improving environmental sustainability.**

SoilReader® 是一家精密农业设备制造公司，利用土壤光谱学原理，对深层多种土壤成分进行高精度、现场实时测量，正在彻底改变土壤分析技术。 SoilReader 是一种附在圆盘上的土壤光谱仪，可在穿过田地时记录多个深度（顶部 9 英寸内）的实时土壤光谱响应，并对多种 (9) 土壤成分进行高分辨率分析（约 3500 个样本/英亩），为快速、无损的土壤表征提供解决方案，从而实现作物投入的精确管理。然而，在土壤读取器盘的每次旋转期间，光谱仪连续记录土壤内黑暗环境和土壤外明亮环境中的光谱响应。这给处理土壤光谱和预测土壤特性带来了困难。圭尔夫大学和 SoilReader 之间的合资企业旨在开发分离光谱并改进土壤特性预测的策略。该项目旨在回答的更具体的问题是：1）光谱测量能够在多大程度上预测现场（现场）条件下的多种土壤特性？ 2）我们如何开发多元统计模型来处理地上/地下光谱响应，以改进土壤特性的预测？通过该项目产生的研究、技术和信息将为种植者、服务提供商、研究人员和其他精准农业服务数据用户提供一种在短时间内进行土壤表征的廉价手段。此外，该项目开发的数学模型将指导SoilReader开发和加强分析服务，并将在未来改进设备，通过优化农艺投入的使用来支持更好的作物生长和发育，同时提高环境可持续性。 **