Tim Rademacher wants to lead a data revolution to understand trees and climate

Tim Rademacher believes that  forests carry more information than we currently make use of. This Research Pitching Competition 2019 finalist from Harvard University and the Northern Arizona University wants to know more about wood and share that knowledge with the world.

Finalistit_Rademacher

Wood is at the core of your research idea. Why is that?

My idea is about understanding the most important factor when it comes to climate change, and that is wood and how it’s formed. Even though wood and forests are right now quite well covered in the media and the public conversation – for example the fire in the Amazon rainforest – it’s still remarkable how little we know about how exactly wood forms and grows, and how this is influenced by climate.

But the information is out there. When wood is formed, current environmental conditions are embedded in its structure. Wood contains a great amount of history and information about past climate conditions. If we could harness all that data in our research it could, for example, help us create more accurate and reliable climate histories and projections. 

So you want to create an open platform with all the possible information about wood. How are you going to do that?

The datasets used in research right now are too small to really understand the growth process of wood. I want to create larger datasets by providing processing and analysis services to the research community through a platform that collects all this information together.The difficulty in making larger datasets is that there are millions of wood samples with many different features. Measuring all these features manually would be super expensive and almost impossible, that’s why the only way to achieve this is to create an automated process for digitising and then analysing the information from the wood samples. My team and I have already created an open access online platform, called the Tree Ring Image Analysis and Dataset, TRIAD. The purpose of TRIAD is to make the digitisation of millions of samples possible, provide free tools for the analysis of new samples, and enable collaborators to easily share the resulting wealth of data.

How do you feel about pitching?

I have pitched before, but not on stage. I think  pitching is somewhat similar to teaching and definitely a good skill to have in the academic world. 

Do you see some value in bringing research to unexpected arenas, for example Slush?

I think it’s really important to bring research to the startup and tech world, because contact creates opportunities for collaboration. My best ideas usually come when I have talked to people who do something totally different. Spaces where people talk about their ideas are crucial to fertilise innovation.

Tim’s wild idea:

Trees grow deeply rooted in their environment. When wood is formed, current environmental conditions are embedded in its structure, turning wood over time into an encrypted memory of the past with decades to millennia of data. Despite billions of these woody memory sticks standing in forests around the world and millions of them already prepared and locked away in scientific collections, our understanding of environmental influences on growth is typically based on very small sample sizes (n < 50). Sample sizes are limited due to the largely manual nature of sample analysis and the lack of digitisation of raw data. Unfortunately, automation requires large data sets to train and evaluate algorithms, such as tree ring detection. Existing samples ought to be digitised to enable the automation of sample analysis and facilitate data sharing. We started building an open access online platform, called the Tree Ring Image Analysis and Dataset (TRIAD). TRIAD intends to make the digitisation of millions of samples that are currently lying idle in scientific collections possible, provide free state-of-the-art tools to streamline the analysis of new samples, and enable collaborators to easily share the resulting wealth of data. With every upload users will grow the public database, which in turn will help to refine tools and enable studies at previously unimaginable scales, leading to breakthroughs in climate reconstructions and improved estimations of carbon sequestration by trees.

Do you want to know more about Tim’s research idea? Get in touch with us: me@skolar.fi