Today’s main challenge to world’s food security is water scarcity, intensified by climate change that affects temperatures and precipitation patterns. This situation is forcing traditionally rainfed areas into irrigation and changing traditional irrigation standards. Considering that globally, irrigated grain and fruit production tend to over irrigate their plots to ensure high yields, raises the need for new irrigation approaches that are more specific to crop and conditions, precise, manageable and water-saving is increasing.
Plant monitoring is a technology that embraces continuous monitoring of soil, plant and climate fluctuations. Even though it has been available for almost two decades, this technology is mainly used for plant physiology research or as a monitoring tool for water induced stress. Accumulated knowledge, gained throughout the years, enabled the development of an algorithm that avoids water deficits through plant sensing rather than soil moisture alone. This algorithm was integrated into an operational platform that gathers data from close-range sensors, applies the developed algorithm and sends irrigation commands to the irrigation controller. During the last three years, several experiments have been conducted in research stations and commercial plots worldwide. This research focuses on two datasets: (1) irrigation trial in Corn, and (2) results of Mango, Apple, Avocado, Citrus, Tomatoes and Almonds commercial orchards, irrigated by the non-stress based algorithm.
The first dataset is a two-year irrigation trial in grain maize, conducted in an experimental site in northern Israel, in which three irrigation treatments were applied continuously throughout the season: (a) FAO and local experts’ recommendations (Kc), (b) 50% stress, and (c) autonomous non-stress-based algorithm. Measurements included continuous data from sensors and manual measurements of total dry biomass and overall water use at the end of the season.
The second dataset includes six different crops in which the non-stress-based algorithm was applied for one full growing season. Results were compared to an identical neighboring plot (control) that was irrigated standardly by the local grower.
Results proved the suitability of using trunk or stem diameter for autonomous irrigation operation. For example, a significant increase in corn stem growth rate was observed after each irrigation event, from around 0.8 mm/day to 1.1 mm/day after irrigation. Additionally, the non-stress based algorithm managed to increase yields by 30%, in comparison to the FAO treatment. This system is accurate enough to enable a precise response to minor changes in plants’ water demand, along with weather and phenological changes. The capability of this algorithm to detect and adapt to small stress events resulted in a reduction of 20 - 40% in water use, with an average increase of 5% in yield in the monitored commercial plots. These results confirm the suitability of this non-stress based irrigation platform for both continuously monitoring plant health status as well as for commercial applications.
