Request for Proposals: Deforestation Early Warning System with WWF

The World Wide Fund for Nature (WWF) is seeking proposals from potential tech partners interested in joining them to develop an Early Warning System to predict the risk of deforestation using machine learning and big data. Submissions are due September 22nd, 2019.

Date published: 2019/08/08

If we are to predict illegal deforestation before it happens, we will be able to act timely and avoid (further) destruction. Existing deforestation models mainly focus on event warnings, resulting in alerts of deforestation after this has already started. Building an advanced risk model combining big data, such as satellite imagery and human activity, will result in better predictive capabilities.

Alerts from this Early Warning System (EWS) will enable us and landscape actors, such as governments, communities and the private sector, to intervene before or in the early stage of unwanted deforestation. Important to address are the negative impacts of the possible interventions to prevent illegal deforestation – economic & social impacts – on the local communities.

Request for Proposals

We invite you to submit a proposal to become a tech partner in the Early Warning System program which entails the development of a machine learning model to predict the risk of deforestation using big data.

What will you be working with/on?

You will be working on the further improvement and development of the medium-term forecasting model, developed by BCG (current tech partner), by adding new data and scaling up the geographic area (Borneo and Sumatra at first).

You will be building a user interface that visualizes the model outputs. Hereby, based on user board feedback and interaction, prioritize deforestation predictions to make these more actionable for land managers.

You will be working with open-source software scripts that have been developed so far, in Python and bash, preferably through cloud computing technologies for scalability.

You will be challenged to evaluate and integrate different machine learning methods over time as the geographic scope expands or/and the input data changes or increases (both in volume or data type).

You would potentially be working with organizations or NGOs that are working on deforestation detection systems, like Global Forest Watch, to integrate the outputs of both systems if this is of interest to both systems.

In a later stage of the program, you will develop and refine the short-term predictive model, also complementing this model with new input data.

Interested?

Submission

Submit your Application Form and supporting documents to this address.
Please keep an eye on this website for updates and possible changes in the RfP!

Deadline for submission: September 22nd, 2019, before 23:00 CEST.

Join the Machine & Deep Learning group to connect with others working in this space.

Continue the discussion… Machine and Deep Learning