The Madingley Model


Ecologists, biologists and environmental scientists at Microsoft Research and the United Nations Environment Programme World Conservation Monitoring Centre (UNEP-WCMC), have spent 3 years developing the world’s first ‘General Ecosystem Model’, or GEM, a model that simulates all life on Earth (land and oceans).  The model is the first to couple all of the key biological processes that underpin the life cycle and behaviour of all of the millions of trillions of the planet’s organisms – photosynthesis, feeding, metabolism, reproduction, dispersal, and death – in order to attempt to capture how such processes lead to the structure and function of whole ecosystems. In this way, the model addresses the central questions in ecology: how do interactions among plants and animals lead to the ecosystems that we see around us, how do these ecosystems vary across the world, what will happen to these ecosystems in the future in response to various human pressures, and how can we mitigate or reverse any damage?

Dubbed the ‘Madingley Model’, after the village in Cambridgeshire, UK, where the scientists first hatched the idea, the model is now able to reproduce characteristics of the ecosystems that we see in the real world, from the small-scale (how long do small mammals live?),  to the large (do more productive ecosystems support more herbivores?). The model also makes new, testable predictions about how ecological interactions among individual organisms shape the natural world that surrounds us.

The Madingley Model could help address key environmental issues, in particular, the effects of human pressures such as habitat loss or the introduction of invasive species on the health and functioning of ecosystems, and the ‘services’ those ecosystems provide for humans (for example, stable supplies of clean water, pollination of crops, or fisheries). The model offers decision-makers a tool to explore the potential effects of their decisions on the environment, in a computer, before the decisions are rolled out in the real world. In this light, the Madingley Model was cited as an exemplary development at the plenary meeting of the Intergovernmental Platform on Biodiversity and Ecosystem Services (IPBES) in 2013.

The model brings together decades of previous work in theoretical and empirical ecology. The current version is a first working version which will hopefully encourage other scientists to become involved in developing this, or analogous, global models of life. We are releasing the model as open source code, allowing anyone to inspect the current version of the model or develop it further.
In some ways, what we have begun to do is build the equivalent of the climate models that are used to predict the future of the earth’s atmosphere and oceans, but for ecosystems. We still have some way to go before we reach the complexity and sophistication of such climate models, which have been in development for decades, but we hope that by engaging the broader ecological community we can make rapid progress.


What, in a nutshell, is ‘The Madingley Model’?

A huge computer simulation of all life on Earth.

What was Microsoft’s contribution to the work?

Microsoft is very much not just technical / computer support on this project! For 8 years now, Microsoft Research has had its own group of ecologists, who carry out and lead novel ecological science such as this. However, like most scientists, the Microsoft ecologists are highly collaborative, and need to work with other scientists and in partnership with other organizations.

Why is Microsoft interested in ecology?

The world is facing many pressing environmental issues. Solving these issues requires new ecology and environmental science, and this science depends on advances in computational methods, software, data gathering, modelling and technology in general. As the world’s largest software company, it is natural that Microsoft would be heavily involved in environmental science.

What is UNEP-WCMC?

The mission of UNEP-WCMC, based in Cambridge, UK, is to provide the world with the information it needs to assess issues and biodiversity and conservation worldwide. UNEP-WCMC employs over 100 scientists who assemble gather data, and other sources of information, into uniquely valuable products relating to conservation and biodiversity, including but not limited to land-use change, invasive species, trade in wildlife, carbon, fisheries, pollution, and agriculture.

Why is UNEP-WCMC interested in modelling all life on Earth?

UNEP-WCMC believes that the Madingley Model project has the potential to create valuable new information relating to conservation and biodiversity worldwide. UNEP-WCMC has been an equal partner with Microsoft Research throughout the project, including co-funding the project, and providing various expertise.

What does ‘open sourcing’ the model mean?

Open source is a mechanism for sharing computer code in way that allows others to inspect, and modify and build upon that code.

Why are you open sourcing the Madingley Model?

The project aims to create an asset that is of value globally, to all humankind, and that can be develop and improved upon by many scientists over coming years and decades. So it is natural that we should open source it.

What have you found so far?

To date, the main result is that the model, which is based on individual-scale ecological processes (like finding and eating food, giving birth to juveniles) reproduces many of the larger-scale patterns that we can see in ecosystems, such as geographical gradients in animal biomass, or trophic pyramids. This is an exciting ‘proof of principle’ that this kind of modelling is actually possible. However, we have already begun to examine what happens when you hit the model with simulated human pressures, such as deforestation and hunting of wildlife.

What do other scientists think of the model?

The reaction to the model has been largely positive, but the model is somewhat controversial. Many ecologists consider that nature is simply too complex to model in this way. Others believe that the outputs from the models are too uncertain to be the basis of important decisions, such as where to place protected areas or how best to run fisheries. But the creators of the Madingley Model believe that ecologists must at least try to model ecosystems, if we are to understand and ultimately save them.

How will the model be used?

There are so many potential uses of the model, that is impossible to predict which will become most important. Because the model if a kind of ‘virtual biosphere’, then it allows us to simulate almost any scenario. From predicting the effects of invasive rats on islands; to Amazon deforestation; to pollution; to overfishing; to climate change; to the removal of keystone species; to ‘rewilding’ scenarios that attempt to restore ecosystem function; the model can be expected to become a key tool that people everywhere use to guide their decisions on how to interact with the environment. Last, but not least, the model may be crucial in guiding science, e.g. which processes to study, in which ecosystems, which data to gather, which experiments to do etc. The model is likely to throw up new scientific questions that were not evident before it was built.