Use of Miniature Thermal Cameras for Detection of Physiological Stress in Conifers

Tree growth and survival predominantly depends on edaphic and climatic conditions, thus climate change will inevitably influence forest health and growth. It will affect forests directly, for example, through extended periods of drought, and indirectly, such as by affecting the distribution and abundance of forest pathogens and pests. Developing ways of early detection and monitoring of tree stress is crucial for effective protection of forest stands. Thermography is one of the techniques that can be used for monitoring changes in the physiological state of plants; however, in forestry, it has not been widely tested or utilized. The main challenge rises from the need for high spatial resolution data. Newly emerging technologies, such as unmanned aerial vehicles (UAVs) could aid in provision of the required data. However, their main constraint is the limited payload, requiring the use of miniature sensors. This paper investigates whether a miniature microbolometer thermal camera, designed for a UAV platform, can provide reliable canopy temperature measurements of conifers. Furthermore, it explores whether there is a distinction in whole canopy temperature between the control and the stressed trees, assessing the potential of low-cost thermography for investigating stress in conifers. Two experiments on young larch trees, with induced drought stress, were performed. The plants were imaged in a greenhouse setting, and readings from a set of thermocouples attached to the canopy were used as a method of validation. Following calibration and a basic normalization for background radiation, both the spatial and temporal variation of canopy temperature was well characterized. Very mild stress did not exhibit itself, as the temperature readings for both stressed and control plants were similar. However, with a higher stress level, there was a clear distinction (temperature difference of 1.5 ◦C) between the plants, showing potential for using low-cost sensors to investigate tree stress.