SLU's knowledge bank

Open data and the Swedish NFI 

Last changed: 21 February 2023
Well maintained spruce forest.

The Swedish national forest inventory – Riksskogstaxeringen – with its uniquely long time series, has developed its own way of working with open data, combining openness with confidentiality. 

2023 marks the 100th anniversary of the Swedish NFI Riksskogstaxeringen (henceforth NFI) which, along with its Finnish and Norwegian counterparts, produces data on the status and growth of our forests, as well as offering a vast archive of data.

Every year, inventories are performed in 5 key areas: area conditions, standing volume and tree biomass, annual volume increment, vegetation and site conditions, and forest damage. 

Two thirds of the samples are permanent areas while the rest are temporary.  

Growing interest 

More and more researchers, journalists and private corporations are using NFI data to make new analyses and develop solutions and products with societal benefit. In the last 15 years the number of requests and tailor-made data orders has grown significantly – from 54 a year to 133. Jonas Fridman, senior environment analyst at SLU, thinks this is in part due to forests and forestry being a hot topic currently.   

– The increase we’re seeing is probably a combination of a growing interest in our statistics and our data due to this being a hot topic, and that we’ve become better at registering our requests, including small ones from journalists and the likes, says Jonas Fridman.  

Previously, data requests came mainly from researchers at SLU and others with special knowledge of the NFI.  

– Another reason we’re seeing an uptick in interest are marketing work being done to spread of awareness of the NFI to relevant groups outside of SLU. More and more researchers also appreciate the value of systematically collected data and long time series, says Anna-Lena Axelsson, who co-ordinates the work with open forest data at SLU.  

Open on request 

The NFI publishes Skogsdata (Forest data) annually, with current statistics and a thematic section focusing on a specific forest-related issue. Data from the NFI has been shared for decades, by methods relevant at the time. 1992 was the first time data was shared digitally, on diskettes or cd. In 2000, the NFI’s own website was launched. Taxwebb for aggregated statistics came along in 2005, and PxWeb (micro data) in 2016.  

– We chose to publish data from our temporary plots and the reason for that was mainly the requests for reference data from international remote sensing researchers, says Jonas Fridman.  

Micro data or statistics? 

Roughly speaking, NFI users can be split up into two groups: researchers looking for raw micro data for their research, and other users primarily looking for statistics, meaning aggregated data. Graphs and tables for example.  

Users can download the data and/or statistics they need from the NFI’s interactive web services. Many also request help from an NFI analyst. A user will then typically send a requests per e-mail or over the phone, followed by a dialogue with the NFI on what’s needed. Relevant data is then delivered. This could be anything from a simple compilation of statistics to a large research project.

Sometimes there’s a continued dialogue during the course of the project, to add additional data and improve the finished manuscript. In some cases NFI experts will function as co-authors. When it comes to requests from researchers, the main users are modellers and remote sensing analysts, and biologists with a conservation profile.  

Accessing statistics and micro data  

All NFI statistics are openly available from the statistical archive, using PX-web. In the interactive tool Taxwebb, users can make their own, simple estimations with an easy-to-use “drag and drop” type method.  

Certain micro data, like tree rings, data from temporary plots and data from the first NFI inventory are available for download as open data, through the website. Micro data is primarily used by researchers while statistics and Taxwebb content is used by the broader public as well as various professionals in the forest industry.

There is ongoing work on producing new open micro data. One example is data on ground vegetation, being made available through Artportalen (Species portal) and Swedish Biodiversity Data Infrastructure (SBDI) 

Simultaneously open and closed  

The philosophy of the NFI is a combination of open and “closed” data. This is due to two things. 

Firstly, roughly half of the plots used are permanent, and re-used year after year. Publishing exact coordinates would mean exposing these plots. Therefore, it is not possible for the NFI to offer this data openly. Exact coordinates can however be revealed, provided that the user describes how they will be used, and signs a non-disclosure agreement.  

Secondly, there is a cost associated with open data. Both in terms of actual money and working hours:  

– If we were to offer all our data openly, it would require a lot of descriptions on how the data fits together. That in turn would take an enormous effort that we don’t think justifies the effort. That’s why [the NFI] has chosen to take requests and manually extract what the user wants, says Jonas Fridman.