Determining and Predicting Soil Chemistry with a Point-of-Use Sensor Toolkit and Machine Learning Model
Determining and Predicting Soil Chemistry with a Point-of-Use Sensor Toolkit and Machine Learning Model
复制标题
使用使用点传感器工具包和机器学习模型确定和预测土壤化学
DOI：
10.1101/2020.10.08.331371
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Grell M
中科院分区：
文献类型：
--
作者：
Grell M
过度使用氮肥破坏了环境和土壤健康;产量下降，而人口继续增加。土壤是一个复杂的活的有机体，它在物理、化学和生物学方面不断进化。土壤的标准实验室测试，以确定氮（主要是NH 4+和NO3−）的水平是不常见的，因为它是昂贵的，缓慢的，氮的水平在短时间内变化。因此，目前的测试实践对指导施肥没有帮助。我们证明，当结合土壤电导率、pH值、容易获得的天气条件时，NH 4+的使用点（PoU）测量（在这项研究中，我们在实验室模拟天气）和定时数据（即自受精后经过的天数），允许使用机器学习（ML）模型对土壤中NO3−的水平进行瞬时预测，R2=0.70（使用更高精度的实验室测量值而不是PoU测量值将同一模型的R2增加到0.87）。我们还表明，长短期记忆递归神经网络模型可用于预测NH 4+和NO3−的水平，直到未来12天，从第一天的单次测量，R2 NH 4 += 0.64和R2 NO3-= 0.70，对于看不见的天气条件。为了在PoU轻松且廉价地测量土壤中的NH 4+，我们还开发了一种新的传感器，该传感器使用化学功能化的近“零成本”纸基电气传感器。该技术可检测土壤中3± 1 ppm的NH 4+（R2=0.85）。由于气相样品的复杂性降低，气相感测提供了感测NH 4+的稳健方法。我们基于机器学习的方法消除了使用专用的，昂贵的传感仪器来确定土壤中的NO3−水平的需要，这是难以用廉价的技术可靠地测量的;此外，可以确定和预测关键的含氮土壤养分，其准确度足以预测气候对施肥计划的影响，并调整作物需求的时间，减少过度施肥，同时提高作物产量。
Overfertilization with nitrogen fertilizers has damaged the environment and health of soil; yields are declining, while the population continues to rise. Soil is a complex, living organism which is constantly evolving, physically, chemically and biologically. Standard laboratory testing of soil to determine the levels of nitrogen (mainly NH4+and NO3−) is infrequent as it is expensive and slow and levels of nitrogen vary on short timescales. Current testing practices, therefore, are not useful to guide fertilization. We demonstrate that Point-of-Use (PoU) measurements of NH4+, when combined with soil conductivity, pH, easily accessible weather (in this study, we simulated weather in the laboratory) and timing data (i.e. days passed since fertilization), allow instantaneous prediction of levels of NO3−in soil with of R2=0.70 using a machine learning (ML) model (the use of higher-precision laboratory measurements instead of PoU measurements increase R2to 0.87 for the same model). We also show that a long short-term memory recurrent neural network model can be used to predict levels of NH4+and NO3−up to 12 days into the future from a single measurement at day one, with R2NH4+= 0.64 and R2NO3-= 0.70, for unseen weather conditions. To measure NH4+in soil at the PoU easily and inexpensively, we also developed a new sensor that uses chemically functionalized near ‘zero-cost’ paper-based electrical gas sensors. This new technology can detect the concentration of NH4+in soil down to 3±1ppm (R2=0.85). Gas-phase sensing provides a robust method of sensing NH4+due to the reduced complexity of the gas-phase sample. Our machine learning-based approach eliminates the need of using dedicated, expensive sensing instruments to determine the levels of NO3−in soil which is difficult to measure reliably with inexpensive technologies; furthermore, crucial nitrogenous soil nutrients can be determined and predicted with enough accuracy to forecast the impact of climate on fertilization planning, and tune timing for crop requirements, reducing overfertilization while improving crop yields.
登录
查看更多内容
DOI：
--
发表时间：
2016
期刊：
影响因子：
--
作者：
Richmond Narh Tetteh
通讯作者：
Richmond Narh Tetteh
影响因子：
6.2
作者：
Rogovska, Natalia;Laird, David A.;Bond, Leonard J.
通讯作者：
Bond, Leonard J.
影响因子：
29.4
作者：
Hamedi, Mahiar M.;Ainla, Alar;Whitesides, George M.
通讯作者：
Whitesides, George M.
DOI：
10.1016/j.agee.2016.06.004
发表时间：
2016-08
期刊：
Agriculture, Ecosystems & Environment
影响因子：
--
作者：
Rory Shaw;R. Lark;A. P. Williams;D. Chadwick;David L. Jones
通讯作者：
Rory Shaw;R. Lark;A. P. Williams;D. Chadwick;David L. Jones
玉米对氮的反应受土壤质地和天气的影响
影响因子：
2.1
作者：
N. Tremblay;Y. Bouroubi;C. Bélec;Robert W. Mullen;N. Kitchen;W. Thomason;S. Ebelhar;D. Mengel;W. Raun;D. Francis;E. Vories;I. Ortiz
通讯作者：
I. Ortiz