Data Science for Biodiversity

Author(s): Tadeusz Bara-Slupski

Originally published on Towards AI.

Conservation of plant, animal and forest genetic resources

The climate change effort is becoming a multi-trillion industry. In such a landscape characterized by a multitude of different projects, it is difficult to discern what is truly going to carry the most significant benefit in the fight for our planet.

One way to go about this is to resort to specific heuristics, and for a long time, mine has been consulting with people who get their hands dirty on the ground. They tend to be the most sensitive to the intricacies of their field and often the most committed to their line of work. Many of the efforts we need to launch to adapt to this changing world require decades-long commitment. I had the privilege to meet such people at the recent GenRes Bridge conference in Tuusula, Finland.

The conference brought together almost 70 stakeholders in the area of genetic resources conservation spanning three domains: plants, animals, and forests. While complex, their work boils down to preserving biodiversity and ensuring food security for generations to come. This is particularly salient in the age of climate change. Many species we currently grow and harvest or plant to establish new forest habitats may not be resilient to the changing ambient temperatures or precipitation. Therefore, we need to model these changes, predict the nature of future ecosystems, and match them with appropriate plant varieties so that our conservation efforts now do not prove futile when their fruits are most needed. We also need to store these varieties in safe vaults in case of natural catastrophes or war.

The organizations in this area encompass a variety of actors from farmers’ associations, through gene banks (like the gene bank in Norwegian permafrost you probably heard about) to multinational institutions, which collect and curate data, as well as develop strategy and policy. In Europe, we have three leading organizations dealing with genetic resources: ECPGR (plants), ERFP (animals), and EUFORGEN, along with EFI (forests).

The GenRes Bridge initiative aims at bringing their three domains together to develop a common strategy and generate higher lobbying power in the EU. To garner greater interest among the general public and EU officials, the organizations need to make better use of their data — both for storytelling and creating higher, cross-disciplinary impact with their work. Appsilon was invited by the European Forest Institute to present to the biodiversity community what kind of services we can offer to support them.

Engaging with the biodiversity community as a data science consultancy

The four-day-long conference was challenging because of the complexity of the subject matter and fast-paced workshops and discussions. As a representative of a data science company with little background in genetic resources preservation, I was doing my best to catch up with my understanding of the issues to fully participate in the discussions. Of course, I did not have much to offer in terms of the substantive matters of improving the coordination of the different domains. Still, I brought an outsider perspective, which was valued on multiple occasions.

Crucially, however, I could offer and educate the participants about the potential data science, improved analytics, and visualizations could bring to their work. It was not an easy task because the participating organizations are generally under-resourced and cannot devote much attention to data-related matters. The resources they have been spent on doing important conservation work. Still, I was very encouraged, because once the word got out about Appsilon’s AI for Good initiative amongst the participants, I was approached by different stakeholders with questions and interest. I was also granted the opportunity to provide a short presentation to the participants about our offer.

There exists a significant demand for data science services in this community. I firmly believe it is representative of many such communities in the environmental sustainability and climate change areas. However, it is also clear that we are at an early stage of raising awareness amongst such actors about the potential data science could bring to their work. Most of the organizations require essential database maintenance services and support in interlinking the dispersed data to create reference points for information produced as part of conservation efforts. They rarely have the time to imagine more advanced applications of machine learning and artificial intelligence and need support in this area.

Going forward and lessons learned

Straight out of the conference, we started a project with one of the stakeholders. Botanic Gardens Conservation International gathers, curates, and disseminates plant resources data from all of the botanic gardens in the world. This allows researchers, conservationists, and other organizations to understand better the global resources of plant genetic material, which allows for better planning of conservation efforts. We will help them with developing an automated validator for data input and clean their databases of inconsistent data such as repeated or incomplete entries.

While we are happy that we will get to contribute to the community at this early stage, we are very eager to engage in broader education efforts. Our goal is to build a global AI for Good movement, which is inclusive and welcoming of organizations in all areas of expertise. To this end, we realized that such organizations require some basic understanding of the opportunities that are currently arising. They need examples of how DS/ML/AI can be applied. Therefore, we are presently planning a webinar for the biodiversity community, which should help fill this gap and pave the way for new and exciting projects.

Overall, I was inspired and humbled by the efforts and knowledge of all the actors attending the conference. I was also very excited to see the demand and interest in data science services. I believe we gathered invaluable insights, and I strongly encourage other data science practitioners to attend such domain-specific conferences to network and better understand the practical reality of the fight for our planet.

Thanks for reading. Follow me on Twitter at @tbaraslupski

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI