Database led to breakthrough in forest research

Sweden has systematically inventoried its forests for more than 100 years. Every year, data on the state and growth of Swedish forests is gathered through the Riksskogstaxeringen NFI. This is a valuable resource for researchers, especially considering the long time series of permanent plots, which enable researchers to study changes over time.  

Swedish research led the way

Inventories such as Riksskogstaxeringen exist in many countries across the world. How the data is used has changed over time. A peer-review article by a group of Swedish ecologists would prove especially important for the development of ecological research.

In 2013, Higher levels of multiple ecosystem services are found in forests with more tree species was published in Nature communications. Writing the article, researchers used survey data in a way that was groundbreaking at the time.

- That article was a bombshell. It gave us insights about how “failing” to rejuvenate and manage a forest might bring about more species by natural rejuvenation. Which is exactly what happens in Swedish forests, says Bertil Westerlund, researcher with SLU Riksskogstaxeringen and responsible for the Swedish data deliveries to GFBI, Global forest biodiversity initiative.

The fact that Higher levels… made an impact was partly because Riksskogstaxeringen had been using permanent sample plots for decades, partly because of technical data development:

- It was a whole new way of working. We did trials with new models that we then applied to larger areas. Based on that we were able to draw conclusions. This was possible thanks to the permanent plots we had been using since 1983. Also, there had been massive development in research as far as modelling and data technology. All this meant that we could use our survey data in a new way, says Bertil Westerlund.

Merging on a global scale

Survey data of the Riksskogstaxeringen type had previously been used in statistics and traditional forest research. But not much in ecological research. It was now possible to do this, making the material much more useful. After a group of researchers realised that a large, common database could assist in biodiversity research, GFBI was founded:

- This collection of data has helped us to understand the ecological mechanisms at play in different regions, and to draw conclusions we otherwise couldn’t have. And it’s all thanks to a major global collaborative effort, with experts from around the world, says Tom Crowther, one of the founders of GFBI and professor with the department of environmental systems science at ETC Zürich.

Crowther and his colleagues were originally looking to do a global analysis of trees. To do this, they contacted a large number of institutions all over the world, to get hold of as much inventory data as possible. Once they had data equal to hundreds of thousands of points, they were able to publish a study showing that the Earth is home to a little over three trillion (3,000,000,000,000) trees. A global analysis of this kind had never been made before although obviously, the approximate number on a national level was known. 

- After this initial collection of data, I was contacted by Jingjing Liang at Purdue college of agriculture. She too had built a large data set, to explore the relationship between biodiversity and carbon sequestration in forests. We combined our data sets and thus were able to prove that biodiversity contributes to increased carbon sequestration, says Tom Crowther.

At Jingjing Liang’s initiative, the group was formalised as GFBI, Global forest biodiversity initiative. The concept was then developed further on the web-based platform Science-i.

Swedish expertise

Swedish contributions to GFBI have been made by Riksskogstaxeringen and its permanent sample plots, which constitute around 60 percent of annual sample data. Tree species and biomass growth has been reported, i.e. changes in biomass growth in Swedish forests.

- We have delivered data but we also have expertise on this issue since we have produced survey data for so long, and learned how to use it. For example, we have helped answer questions about how to use survey data to compile the database, and how to balance data deliveries so that it works with what you are looking to achieve, says Bertil Westerlund at Riksskogstaxeringen.

Despite only using a select number of variables – tree species and biomass growth – data needs to be harmonised to be compatible across borders.

- Inventories don’t use the same sample plot size, or the same number of sample plots per square unit. Therefore, plots must be balanced along with sample intensity, meaning number of sample plots per hectare. You could say it’s all one big harmonisation job, says Bertil Westerlund.

The most influential studies Bertil Westerlund has contributed to are Climatic controls of decomposition drive the global biogeography of forest-tree symbioses (Nature 2019), on mykorrhiza, The number of tree species on Earth (Proceedings of the National Academy of Sciences, 2022) and Co-limitation towards lower latitudes shapes global forest diversity gradients (Nature ecology & evolution, 2022), on tree diversity. Many other studies have also been published based on the same data, but SLU decides on a case-by-case basis whether to appear as co-authors or not.

Bertil Westerlund says the greatest advantage of GFBI’s big, global data sets is the ability to generalise beyond a local or regional perspective. Says Bertil:

- In research on forest biology and forest ecology, we typically evaluated to try and find causation. For example the interplay between tree and fungi, tree leaf properties, tree species diversity and its effect on biomass production. The results have only been valid for a local or regional area, however. What GFBI has enabled us to do, is to generalise beyond these causal contexts, on larger areas such as biospheres and the entire Earth’s ecosystems.