Welcome to the CrackSense Spotlight Series, where we speak with experts working at the intersection of fruit production, technology, and innovation. In this edition, we spoke with the two coordinators of the CrackSense project from the Volcani Institute: Avi Sadka, Ph.D., Professor of Plant Sciences, and Victor Alchanatis, D.Sc., Research Scientist. They shared their perspectives on fruit cracking, the progress of the project, and the importance of data-driven methods in today’s agricultural practice.
Could you tell us a bit more about your background and areas of expertise?
Avi Sadka:
We are from the Volcani Institute, which is the research arm of the Ministry of Agriculture and Food Security, the Israeli Ministry. We both co-coordinate the CrackSense project. Victor is in charge of the technical part, and I am in charge of the horticultural part. I am a Citrus Physiologist, and I work on different aspects of citrus productivity and fruit quality. Cracking is very important in citriculture, as it is in other crops, because it may cause damage to the yield. Some cultivars were so severely affected by cracking that they were simply eradicated.
Victor Alchanatis:
I am an Agricultural Engineer from the Volcani Institute in Israel. My basic expertise is sensing and data technologies. Cracking is a severe agricultural problem, but the technologies that we have today can be used to solve it using data-driven methods. We are trying to use sensors and data to put everything together in modern algorithms that can explain the risk of cracking, the phenomenon of cracking, so that the farmers will have the opportunity to use whatever management procedures they have in order to mitigate the cracking.
Could you share a few key outcomes from the experimental phase of the project?
Avi Sadka:
The first two years of the project were devoted to generating experimental plots in which cracking intensity shows variation in the same plot. We did experiments in four crops: pomegranate, citrus, sweet cherry, and table grapes, at several locations in Europe as well as in Israel. The idea is that these experimental sites would serve as an infrastructure to test the sensing technologies. In most cases, the experiments were successful, with the plots generating varied cracking intensity and sensing tools applied at these sites. Now we can say that we have started to model the relationship between cracking intensity and data driven by sensing technologies.
The limitation of sensing tools is that they cannot detect fruit cracking directly. So, we have to monitor tree physiology, tree health, and correlate it to cracking intensity. In that sense, we are in the right direction. We already have proof of concept that we can use sensing tools, especially remote sensing tools on UAVs, to detect or to predict cracking intensity.
Victor Alchanatis:
CrackSense is a data science project. The physiological and agronomical part, which was the main part in the first two years, provided extensive data. Now we are at the stage where we have to build the models. We intend to build models of AI, artificial intelligence, and machine learning in order to find the connections between the prediction of the risk of cracking and the data that we collect. We will try to do this at multiple scales: the scale of the tree, the plot, and the region. The data for those models have already been acquired and collected during the first two years.
The rest of the project will be devoted to implementing those models and testing them in pilot plots – plots where we no longer have experiments, but where we will try to apply the models and see the accuracy of the predictions for risk of cracking.
What are the main outcomes you expect the project to deliver?
Avi Sadka:
The integrated solution of CrackSense should result in a prediction model that can serve the growers and policy makers, to assess the risk of cracking. This assessment is very important. If the prediction is for high cracking intensity, then the farmer or grower can decide to take certain measures or avoid different measures in order to reduce the economic damage.
Conclusion
This discussion with Avi Sadka and Victor Alchanatis provides a clearer picture of how CrackSense connects horticultural knowledge, sensing technologies, and data science. With experimental sites established, extensive datasets collected, and model development underway, the project is progressing towards tools that can help growers better assess and respond to fruit cracking risk. Future phases will show how these models perform in pilot plots and how they may support more informed decision-making in orchards across different regions.
More Spotlight On interviews are on the way. Subscribe to our YouTube channel and follow our Newsroom and LinkedIn to keep up with every update.