From mathematical models to bee-friendly plant protection

There is safety in numbers

Many factors affect the conditions of a beehive. With the computer program BEEHAVE (Becher et al. 2014)*, users can set different parameters and test their effects on a simulated bee colony. Bayer researchers hope the program will advance their understanding of the factors influencing bee health and the development of bee colonies.

From mathematical models to bee-friendly plant protection


// The European Food Safety Authority has formulated specific protection goals for honey bees and wants to make the impact assessment procedure for plant protection products more extensive.

// However, many experts believe that the protection goals – and particularly their translation into regulatory testing requirements – need to be revised.

// Mathematical models can help to establish more realistic goals. Virtual bee colonies can be created by using simulation software such as BEEHAVE*.

// The model observes the impact of various factors, including Varroa mite infestation, the role of the beekeeper and environmental factors.

// On the basis of the findings, researchers want to make more realistic predictions of the potential impact of plant protection products on honey bee colonies.

Dr Thomas Preuss - The Beehave Model

Dr Thomas Preuss can simulate different scenarios in the BEEHAVE* program and thus calculate how a bee colony will develop in future.

You could hear a pin drop – and yet, the scene on the screen is a literal hive of activity. It is peak season for the virtual bee colony living inside the computer. The landscape of bits and bytes offers an abundance of nectar and pollen, honey production is on point and there are no Varroa mites to affect the health of the pollinators. That is the scenario Dr Thomas Preuss planned when he set the parameters for his virtual bee world. The biologist works in Environmental Modelling at Bayer’s division Crop Science and has already created, observed and analyzed thousands of virtual colonies or, in other words, the digital models of beehives.

“Just like in a computer game, we can set different starting conditions. The BEEHAVE* simulation model then uses a series of scientific processes to work out how those conditions affect the colony,” explains Dr Preuss.

The background: In the context of pesticide regulations, the European Food Safety Authority, EFSA for short, formulated specific protection goals for honey bees and recommended that they be translated into testing requirements. When these goals were established, the model from which they were derived was still rather basic. Now, a much more refined and realistic model is available developed by scientists from the University of Exeter (UK), the Helmholtz Centre for Environmental Research (Germany), and Syngenta (Becher et al 2014). Bayer is very interested in this development and is committed to investigating the potential effects of plant protection products under realistic field conditions by means of monitoring studies and the BEEHAVE* model.

Some simulations last a few days, depending on the complexity and starting conditions. However, Dr Preuss can often tell after only ten minutes how his virtual beehive is going to develop over the following years. BEEHAVE* knows exactly how many drone bees there are in the colony, how many eggs the virtual queen lays in a day and what percentage of worker bees is looking after the brood.

The biologist can even predict the lifespan of worker bees and how far they fly, how much pollen is in the hive and how much nectar is consumed in a day. “At the end, we also work out how much honey was produced, how many Varroa mites there were in the beehive and how many of the parasites were transmitters of a virus,” says Dr Preuss. Many of the calculated parameters, such as how the Varroa mites are distributed in the beehive and how they affect the colony, have already been validated in initial experimental tests. The model also considers the role of the beekeeper. “This is because they actively intervene with the colony, collect the honey, give the bees sugar water, fight pests and disease vectors, and thus ensure the colony’s survival,” says Dr Preuss.

Beehave Model

The mixture and the amount of nutrients influence the strength and welfare of the bee colony. Some residues of crop protection products can also enter the beehive in food the bees gather. These products will be excreted through different mechanisms.

Bee Icon

The BEEHAVE* program is available free of charge and is constantly being developed further by bee researchers around the world. Kerstin Hörig, a PhD student in Dr Preuss’s former team at the German RWTH Aachen University, is making the virtual bee world more and more realistic as part of her doctoral thesis. She focuses on the ways in which crop protection products affect the colony. “We have to include feedback mechanisms that play an important role in biological systems such as bee colonies,” explains Dr Preuss.

“Many factors mutually affect each other – such as food availability and the beehive’s susceptibility to disease – and can be represented best using computer models.”

There can also be a few surprises along the way: During a simulation study, Dr Preuss did not let any of the forager bees return to the virtual hive for five consecutive days. Yet the negative effects on the colony were minimal: “This is only possible in an extremely favorable environment, with plenty of nectar and pollen available,” says the biologist. “This keeps the colony healthy and resilient. It can build up stocks, which helps make up for losses.”

The researchers are currently assessing the data collected from monitoring studies and experiments with a view to better understanding the biology of bees and offering them even more protection in future. They have to filter out the relevant information from the findings, evaluate it and develop a set of measures. “We want to understand the impact of environmental factors and make recommendations on how to prevent potential negative effects of plant protection products under realistic conditions,” says Dr Preuss.



Simulation programs, such as BEEHAVE*, analyze and evaluate the risks to bee populations posed by a variety of different factors in different scenarios. On this basis, the complexity of the factors at play and their combined effects can be better understood and, thus, environmental risk assessment can be amended and optimized. There are comparable mathematical models that simulate the effects of plant protection products and environmental factors on organisms such as water fleas and mice.

in numbers


Milligrams of nectar per bee/trip


Nectar is a source of carbohydrates and energy for bees. Its availability depends on time of year and weather. On each trip, foraging bees transport some 30 milligrams of nectar in their “honey stomachs”. Every year, a colony gathers about 120 kilos of nectar in the hive and turns it into around 25 kilos of honey.


Kilos of nectar per colony/year


Pollen is a source of protein. Bees transport it in the pollen baskets on their hind legs. Every year, the worker bees transport between 20 and 25 kilos of pollen and use it to make “bee bread” which is fed to the brood. Over the season, collection and consumption of pollen roughly balance out, so stocks are usually stable at about one kilogram.

20 TO 25

Kilos of pollen per colony/year


* Becher MA, Grimm V, Thorbek P, Horn J, Kennedy PJ, and Osborne JL, BEEHAVE: a systems model of honeybee colony dynamics and foraging to explore multifactorial causes of colony failure. J. Appl. Ecol., 51(2): 470-482 (2014).

© The University of Exeter, Matthias Becher 2013

Complete Story

From mathematical models to bee-friendly plant protection There is safety in numbers
Back To Top