Projects

CEES stands for the Computational Ecology and Environmental Science group at Microsoft Research Cambridge. If you haven't done so already, it would probably make sense to read about the overall goals of the group before going on...

We carry out, lead, collaborate on, or are otherwise involved with a wide variety of research projects, most of which are listed below in alphabetical order -- so don't say you weren't warned! However, we aim to integrate the results of our many projectrs into two flagship projects in modelling the global carbon-climate feedback, and modelling global ecosystem function.

 

  • Distribution Modeller: Environmental Modelling at the Speed of Thought
    Distribution Modeller (temporary name only!) is CEES' end-to-end browser tool that lets the researcher to rapidly import data, supplement that data with environmental info from FetchClimate, specify an arbitrary model by point and click or in code, parameterize the model against the data using Filzbach, make and visualize predictions with a full propagation of parameter uncertainty – then package and share everytihng, in a way that is inspectable, repeatable, and modifiable.
  • 2020 Science
    The University College London and University of Oxford have recently received funding from the EPSRC Cross-Disciplinary Interfaces Programme (2020 Science: Mathematical and Computational Modelling of Complex Natural Systems) to collaborate with Microsoft Research Cambridge on a programme of research that will involve up to 17 post-doctoral Research Associates over a five year period.
  • Conservation at Microsoft
    We develop and accelerate better, predictive, conservation science, tools and technologies in areas of societal importance. We aim to provide scientific support for effective environmental solutions for key decision makers, from the boardroom to governments makers. We are committed to leveraging the unique position our group occupies to influence how individuals and nations approach and tackle issues such as natural resource scarcity and biodiversity loss.
  • Species Distribution Modelling
    Species Distribution Modelling (SDM) aims to explain why species occur where they do, and why they do not occur anywhere else. For instance, why does an oak tree not occur further south in hotter and dryer regions, and why it may not occur further north in colder and wetter regions? This sort of information is amongst the most fundamental of all ecological knowledge and is of great societal importance. Distribution data can feed into almost all other biodiversity models, and can inform adaptive
  • Mapping Threats to Biodiversity
    Science, Policy, and Tools & Technology drive the Conservation@Microsoft Research Unit. This unique project combines those pieces to provide fresh insight into the relationship between species, their environment and the impact that human activity has on them.
  • Filzbach
    Filzbach is a flexible, fast, robust, parameter estimation engine that allows you to parameterize arbitrary, non-linear models, of the kind that are necessary in biological sciences, against multiple, heterogeneous data sets. Filzbach allows for Bayesian parameter estimation, maximum likelihood analysis, priors, latents, hierarchies, error propagation and model selection, often with just a few lines of code.
  • Global Ecosystem Function
    An expanding population, with expanding resource use per capita, is resulting in an alarming loss and degradation of ecosystems. In order to balance the need for increased food, timber and textiles production, with industrial use of natural resources, with the healthy functioning of natural, semi-natural and artificial ecosystems, we require predictive, global models of the response of ecosystems to various human activities.
  • The probabilistic terrestrial carbon model: towards actionable predictions
    Climate change is the greatest global challenge of the 21st century. Models that reliably forecast future climates associated with different policy scenarios are urgently needed. This project has developed a new model for the key source of uncertainty in earth system models: the terrestrial carbon climate feedback, using a methodology to account for all known sources of uncertainty and enable robust estimates of the confidence that can be placed in predictions and objective model refinement.
  • Predictive modelling of tropical deforestation
    This project aims to generate a predictive model of tropical deforestation, able to predict the rate of deforestation in different regions of different countries under various economic and policy scenarios.
  • Understanding and Predicting Tree Mortality
    Tree mortality is central to an understanding of almost anything to do with forests, and yet we currently know little about the nature or magnitude of variation in tree mortality, how it depends on species, climate, or extreme events, or how best to include it in simulation models. This project aims to combine new models, with large forest inventory databases, to enable reliable predictions of tree mortality at scales from the individual tree, through the forest stand, to the globe.
  • A Network Approach to Understanding and Managing Meadows in Yosemite National Park
    The meadows in Yosemite National Park form a complex network of interconnected habitat patches. We are using network analysis techniques to understand the network's structure, to model species movement through the network and to model effects of human use on the network.
  • Forest Dynamics
    Forests contain two thirds of terrestrial biodiversity and store as much carbon and is currently in the atmosphere. We are combining new abstractions of forests, with various sources of data, via Bayesian statistics, to produce useful, predictive models of forest dynamics.
  • Tools to aid in teaching ecology
    Improving the communication of scientific principles and techniques is a widely desired goal in schools, universities and academic institutions. One of the key challenges in achieving this is in developing communication methods that are understandable and memorable to a varied audience. This project will explore the potential benefits of using small computational “toys” to demonstrate key ideas in scientific communication.
  • Predicting disease dynamics in wildlife
    Datasets on disease spread in human and wildlife populations are increasing in size and variety. This project will address how can we best understand and predict host-pathogen dynamics in wildlife populations and whether we can predict the risk posed to human populations from zoonoses through the study of their dynamics in their wildlife reservoirs.
  • Understanding species population dynamics
    Wild species populations are known to exhibit a wide range of spatiotemporal dynamics. We are using mathematical modelling to investigate why biological populations show a variety of spatiotemporal dynamics, more commonly associated with chemical and physical systems, such as travelling waves, spirals and chaos. We aim to make our state of the art mathematical techniques understandable and available to biologists generally.
  • Graphical tools for text analysis
    Despite the considerable effort authors put into arranging the written word, the structure produced is inevitably a long line. The structure contained within this line is only usually revealed by a summary of content and an index. Contents point to subsections of the line that may have some similarity in subject and indices the positions of examples or similar topics within the whole line. This project explores the relationship between the two and structural forms within the text. I am investiga
  • Range deformation: consequences of range shifts during climate change
    As climate changes populations’ relationship with environmental variables alter, driving changes in populations' spatial distributions (Range Deformation). The changes in spatial and within population structure generate alterations in populations’ evolvability, producing evolutionary changes feedback into the population structure.
  • Next Generation Bioclimate Modelling
    Predictions of populations responses to climate change have largely centred around bioclimate models. Bioclimate modelling has afforded conservation biology a means to rapidly and extensively assess the distribution of climate change’s threat to populations. However there are numerous conflicts between the assumptions and biological reality. Parameter estimation techniques may allow ecological reality to enter climate change predictions.
  • Lambda-Omega Equations Simulator
    This software tool allows the user to view animations of the dynamics illustrated in the paper "Absolute stability of wavetrains can explain spatiotemporal dynamics in reaction-diffusion systems of lambda-omega type". Smith, Sherratt & Rademacher, submitted.
  • Density-Dependent Dispersal in Cyclic Populations Simulator
    This software tool allows the user to view animations of the dynamics illustrated in the paper “The effects of density-dependent dispersal on the spatiotemporal dynamics of cyclic populations.” by Smith, Sherratt and Lambin, that appeared in Journal of Theoretical Biology (doi: 10.1016/j.jtbi.2008.05.034).
  • Environmental Scenario Search Engine
    Data mining tools to explore exponentially growing archives of environmental sciences.
  • Using 25 Years of Infra-red Satellite Data to Derive a New Global Fire Model
    Forests harbour around 60% of the world’s biodiversity and around half of its terrestrial carbon, so there is an urgent need to predict how forests will respond to increased atmospheric CO2, logging and land-use change. This project will collate millions of pre-existing field measurements of trees from national forest inventories into a coherent, user-friendly database and use this data in the development and parameterization of models fire at global scales.
  • Tools for the Analysis, Interpretation, and Visualization of Biodiversity Data from Mexican Forest
    Evaluating and applying emerging ecological niche modelling techniques to study the changing distribution of Mexican cloud forests.
  • Toolbox for Spatial Analysis of Invasive Species Spread
    Invasive species are causing significant economic and environmental damages worldwide. This project will develop a tool to calculate the rate of spatial spread of an invasive species though habitat, and to determine the factors that determine that rate.
  • Structure and Dynamics of Complex Ecological Networks
    Computational approaches to studying food webs, networks of who eats whom in an ecosystem.
  • Data-constrained Modelling of Plant Growth
    Plants are complex reactive systems. They process water and nutrients, fix carbon dioxide (CO2) into new plant material, decide where to allocate this new carbon (e.g. leaves, roots, stems), and decide when to flower and produce seeds.
  • Computational Tools for Biodiversity Science
    A multi-institution consortium addressing the scientific and technological needs of biodiversity scientists and policy makers.
  • Autonomous Monitoring of Vulnerable Habitats
    Automatic monitoring of the effects of changing environmental conditions on the ecology and behaviour of the Manx Shearwater.
  • Cambridge University Herbarium Digitisation
    Digitization and online publication of the most scientifically and historically significant parts of the Cambridge University Herbarium’s collection.