Tracking infrared plant emissions from productive land
Monitoring plant health is a vital part of crop management across our agricultural, horticultural and viticultural industries. As part of our aerial data recording options for land health, Land IQ Insights can now offer greater insight into foliage health through the use of visible and near-infrared (Vis-NIR) spectroscopy.
Nature has the ability to employ much of the infrared range from the spectrum for plant health. While we as humans are unable to see the effects our new system can capture, much of the spectrum that is used to evaluate plant health. We have applied this to an operating vineyard in Central Otago, using a ‘false colour’ key to show the variation.
Let’s look at this in action. (With thanks to Grey Ridge Vineyard, Alexandra)
On the left is a standard photo image, On the right we are looking at a base NDVI (Normalized Differential Vegetation Index). Indexed number range is between -1 and +1: think very unhealthy (Red) through to very healthy (green).
We measured the yield from each row section of the vineyard when the grapes were picked (the part of row between the fence posts, usually about 5 plants). We were able to see many sections that produce 1-2 Kg, yet the section beside or next to that section can produce fundamentally different yield.
Less intervention, early
Changes in industry practices are demanding non-destructive, rapid, and reliable detection methods at an early stage.
When we overlay the harvest weight over the Indexed drawing created in the growing season, we can see a high correlation between harvest weight and indexed value.
Until recently, the number of techniques able to detect plant stress and disease at an early stage had been limited to visual inspections and sampling, both of which are quite labour intensive, involve specialist skillsets, and are often only picking up on changes once the problem has already manifested symptoms. Prevention of diseases and stress, while the plants are still in an asymptomatic stage could lead to better crop management in agricultural industries.
Science backs spectral data for early detection
Recent studies have been done on applications of visible and near-infrared (Vis-NIR) spectroscopy in disease detection and the implications of stress in various species of plants. These studies support Vis-NIR monitoring as a rapid and non-destructive technique that requires only minimal sample processing (sometimes none at all) before measurements and data analysis.
The visible and near-infrared region proved able to detect almost all functional groups and compounds making it a promising tool for data analysis.
With an increasing need to achieve better yields whilst sustaining good soil health, early detection of disease and plant stress provides more options to isolate and take action in a targeted way.
How reliable is the data and how can it be used?
The ability to apply large-scale real-time monitoring of disease and other stressors in plants presents many opportunities for implementing effective strategies to increase productivity and prevent greater losses in the agricultural industries.
The spectral data in the visible and near-infrared regions has the potential to provide diverse information on chemical parameters via the absorption pattern and characteristics, enabling the early detection of abnormal conditions caused by disease infection, or stress.
In some fields of agriculture, spectroscopy methods are already among mostly used methods for plant disease detection owing to the non-destructiveness, non-invasiveness, high-speed, high sensitivity, and specificity for particular disease types. Vis-NIR spectroscopy, which is what Land IQ Insights uses, is a proven measurement technology that had been implemented for the non-invasive analysis and identification of several parameters across a wide range of agricultural commodities.
This method had been demonstrated to discriminate between healthy and unhealthy conditions in plants based on the infrared light frequencies emitted by the plants when stressed.
Capabilities of our multi-spectral image capturing technology
Vis-NIR spectroscopy is a non-destructive, and rapid method that provides the prediction of the chemical and biological composition of a system. The Vis spectroscopy is capable to examine the colour and pigment (e.g. chlorophyll) analysis, whereas the NIR spectroscopy measures the macro components, mainly water. The visible region (400–750 nm) provides information on the spectral features of the pigments and is used by plants during the process of photosynthesis.
Troubleshooting
The Land IQ Insights drone is fitted with multi-spectrum image capturing equipment that has six cameras in one, which allows separation of channels for greater control of evaluation.
The aerial passes by our drone are precisely mapped and controlled using GPS tracking.
Other factors that can influence readings include ambient noise and light scattering, which can be factored in during pre-processing calibrations.
Outcomes
Our readings enable you to generate objective data using complete analysis of the surveyed area (not just spot sampling), which you can apply to triaging of potential disease outbreak or store for deferred analysis of outputs, and year-on-year comparisons.
It is cost effective, reducing the involvement of the workforce in the execution of visual monitoring and sampling tasks, and allowing re-use of the same recorded data to extrapolate new information.
For more information about how aerial Vis-NIR spectroscopy imaging can benefit your operations, contact Neill.
________________________________________________________
This article draws references from a 2022 article: “A review of visible and near-infrared (Vis-NIR) spectroscopy application in plant stress detection” Siti Anis Dalila Muhammad Zahir, Ahmad Fairuz Omar, Mohd Faizal Jamlos, Mohd Azraie Mohd Azmi, Jelena Muncan, published in Sensors and Actuators A: Physical, Volume 338, 1 May 2022, 113468. Online source: https://doi.org/10.1016/j.sna.2022.113468
Further reading:
“The potential of spectral reflectance technique for the detection of Grapevine leafroll-associated virus-3 in two red-berried wine grape cultivars”,Rayapati A. Naidu, Eileen M. Perry, Francis J. Pierce, Tefera Mekuria
Extract: “The differences in leaf reflectance measurements at specific wavelength intervals between virus-infected and uninfected grapevines and their correlation with RT-PCR results for the presence of GLRaV-3 suggest spectral reflectance technique as a promising tool for cost-effective, non-destructive method for diagnosis of GLD in the field…”
Online source: https://doi.org/10.1016/j.compag.2008.11.007