Estimating Kc on a Plot Level Using Remote Sensing and Artificial Intelligence

*Session presented at the 13th European Conference on Precision Agriculture (ECPA2021) virtual event on July 2021.

  • Ran Pelta Ph.D.
  • Remote sensing and data scientist Manna Irrigation

13 minutes

This is a recording of a presentation of a study, Led by Ran Pelta Ph.D., Remote sensing and data scientist at Manna Irrigation. It was presented at the 13th European Conference on Precision Agriculture (ECPA2021).

Study abstract

Traditional methods for irrigation decision-making might not be efficient or relevant nowadays. In this study, remote sensing and meteorological data were collected for almost 600 commercial processing tomato plots for better irrigation decision-making. An artificial intelligence model was trained on 2017-2019 growing seasons and validated for 2020. At the beginning of the season, the model estimated the crop behavior in terms of crop coefficient (Kc) for the entire season. Additionally, a piecewise regression model was employed to estimate the crop growth stage in terms of days from the season start. The results of this study show improvement in both Kc and growth stage estimation, at the beginning of the season, compared to traditional crop protocols. The results can help to design the irrigation regime at the plot level and thus improve the ability to allocate the required water amounts between plots in real-time and even to plan it before the season starts.

Who we are?

Manna irrigation – a sensor-free, software solution that provides irrigation recommendations and crop monitoring tools based on remote sensing and AI model.

  • Founded in 2016 by Rivulis Irrigation
  • 17 employees
  • HQ and R&D in Israel – PhDs in agronomy, remote sensing, data scientists and SW development
  • 3 Patents
  • 7 Published papers

What do we do?

  • Hardware-free
  • Daily irrigation recommendations
  • Continous Crop monitoring with near-daily satellite images
  • Automatic notifications of irrigation and crop development issues
  • Better Water Use Efficiency and improved yield

What is the problem?

Up To Date Plot specific Kc protocol is needed

Crop coefficient (Kc) protocol, provides just a partial picture:

  • At season start
  • Crop behavior and periods
  • Experimentally determined
  • Time-consuming, skilled personnel
  • Regional/country scale
  • Might be outdated, inaccurate, irrelevant
  • Reduction of water use efficiency

What Manna’s solution offer?

  • Data-driven approach
  • Remote sensing
  • Hyperlocal weather data
  • Artificial intelligence model

Up To Date Plot-specific Kc protocol