Plants experience stress in response to changes in the environment including drought, heat, pests and diseases (see my blog post about plant stresses). When plants experience extreme stress it can be relatively easy to detect. Wilting can indicate drought and heat stress, yellowing of leaves can suggest nutrient deficiency, whilst browning of leaves can indicate pest or disease attack. However, by the time these are clear often the plant is severely damaged. This is an issue for farmers, who have already lost crop yields by this point.
So how can we measure subtle changes to detect stress in whole fields of crops?
Drones and remote sensing! Satellites and drone technology are an emerging method to monitor crops on a large scale. Images of crops can be used to determine the health and growth stage of the crops, without the need for examining every part of a field. However, visual images can be difficult to interpret, as crop colour can be affected by ambient light and weather conditions.
Another approach is needed that doesn’t rely solely on visual light. Hyperspectral cameras use specific wavelengths including: green, red, blue, red-edge and near infrared to monitor crop health. By measuring the amount of light absorbed and reflected at these wavelengths a number of things can be determined about crops. For example, stress leads to an increase in infrared light absorbance and a decrease in reflection. Nutrient stress causes subtle changes in the blue and red light bands, which can inform the application of fertiliser. Infections can reduce the level of chlorophyll in leaves. This can be detected in the visible light and near-red wavelengths and browning of leaves using near-infrared wavelengths.
Using drone technology allows plant stresses to be detected earlier than conventional methods, before farmers suffer significant crop losses. This could lead to more efficient application of fertiliser and pesticides, reducing farmers costs and environmental impact. As the cost of drone technology decreases there are more opportunities to make this available to farmers, and improve crop efficiency using technology.
Check out these links for really good reviews of hyperspectral methods to detect crop stress: