Combining satellite imagery and bioagressor phenology modelling for sustainable crop protection

Tunisia is a Mediterranean country, situated in the northern part of the African continent where agriculture plays a leading role in its economy (approximately 12 % of its GDP). Moreover, In Roman times, it was known as the “granary of Rome”. Nowadays, the importance of this sector is suffering from several threats such as the decrease of the yields of different cultivated crops caused by several phytopathogenic fungi and pests. These harmful damages are increasing from year to year due, principally, to climate change.

I am currently a student and I will get my get my engineering degree this year (June 2019), I am working on pest modelling under the supervision of Dr. Mohamed ELIMEM. We started by understanding the Tunisia’s agricultural system, which is based on small family farms that grew subsistence crops with little market integration. We released that farmers do not know how to identify the disease nor to distinguish between damages caused by a bioagressor and a deficit in a mineral element. In this case and most of the time, they will use pesticides without asking for a recommendation. This behavior will never guarantee a sustainable farming system.

This approach started from an idea that speaks about the right timing of application of the phytosanitary treatment (ex. pesticides) to control pests or phytopathogenic fungi, which should coincide with the maximum percentage of the population that causes harmful losses in terms of agronomy (yield) and economy. For example, a pest phenology model is done by predicting the time of events in an organism’s development. We all know that insect is a poikilothermic animal, so its development depends on temperature, relative humidity and photoperiod to which they are exposed in the environment. The “degree-days” or “effective temperature summation” method can predict population phenology from climatic data of the real time and allow to forecast the set-in of a phenological stage in a region of the world. Normally, the “degree-days” method is applied to determine the time of the year when the population’s proportion reaches its maximum. To routinely apply this procedure we should correlate phenology data obtained experimentally with effective temperature summations above a pre-established threshold.

To achieve good control of a bioagressor, we should also include satellite imagery (Remote Sensing (RS) & GIS) and the use of drone. Our government and the African Development Bank have signed an agreement for the launch of a pilot project to use drones for data collection to enhance management of agricultural projects in the country. This project is also setting up of a training center equipped with training drones as well as computer simulation tools for drone control. This center is expected to be upgraded to a center of excellence in drone technology. The training also focuses on promoting drone-centered activities in Tunisia in view of promoting efficiency and effectiveness.

Last summer (2018), we collected some information about pests in a citrus orchard located in the region of Mograne (Tunisia) and we will use this to predict pest development this year (during my final project work). We already have a weather station which belongs to the National Meteorological Institute of Tunisia (NIM).

This start-up will be composed of a team divided into two groups, the first will help farmers by doing a weekly vulgarization to give them instructions and recommendations by following a full technical package from sowing to harvesting to fulfill efficient integrated pest management. The second group will realize experiments and statistical tests to develop algorithms to know more about the phenology of bioagressors that affect crops.

To be realistic, all outcomes depends on the number of farmers that accept us to control their crops, but we can give an estimation about that. So, we can say, if we had 10 clients, we would release 3000 €/month.

Combining these techniques (application of satellite imagery and bioagressor’s phenology modelling) will let farmers to use less pesticides and raise up the efficacy of treatments, by identifying the threatened zones.

Finally, I am applying to meet experts and to learn from their rich experience in several domains and I will be grateful to be selected for the the IUPAC Next Generation Agri Summit and my project for the N-GAGE champions

Mohamed Guesmi, Tunisia


  1. Very interesting project.
    The usage of drones is pretty smart.
    I hope this idea will come true and will be accessible for the majority
    Keep up the great work!


Comments are closed.

%d bloggers like this: