Forest data acquisition in Nordic countries - roadmap for the future Annika Kangas, Rasmus Astrup, Johannes Breidenbach, Jonas Fridman, Terje Gobakken, Kari T. Korhonen, Matti Maltamo, Mats Nilsson, Thomas Nord-Larsen, Erik Næsset & Håkan Olsson 18.9.2017 9.12.2018

Forest resources assessments are needed to ensure sustainability The sustainability of utilizing forest resources was discussed already at 18th and 19th centuries Visual assessments of forest resources were carried out to address fears concerning sustainability The debate resulted in starting national sampling-based forest inventory Norway 1919, Finland 1921, Sweden 1923 In Denmark the forest resources assessments were carried out with questionnaires until 2002 9.12.2018

Current NFI designs Method used Number of strata Cycle   Method used Number of strata Cycle N of plots per year Use of dGNSS Proportion of permanent clusters Denmark Cluster sampling - 5 years 8600 33% Finland Stratified cluster sampling 6 Panel system of 5 years 15000 Since 2014 100 % 60% Norway Stratified systematic sampling 4 4400 Currently 50% 100% Sweden 5 10000 From 2017 onwards 65% 9.12.2018

Use of remote sensing in NFI is limited Currently remote sensing information is not used either in design nor estimation of the official regional results Exception small-area results (e.g. municipality level results in Finland) Instead, remote sensing is utilized in forest resources map production Satellite images Airborne laser scanning (ALS) Denmark, Sweden Digital aerial photogrammetry (DAP) Norway Maps publicly available for service providers 9.12.2018

Remote sensing material used Publishing times Coverage Resolution   Name of the product Remote sensing material used Publishing times Coverage Resolution Field data Availability for public Denmark Land use/land cover map Landsat 1990, 2000, 2011 100% 30m × 30 m None No Forest resource map ALS 2006-2007 25m × 25 m NFI plots No (Yes for research purposes) 2014-2015 Yes Finland MS-NFI map Landsat (Spot, Sentinel) 2015, 2013, 2011 16 m × 16 m Yes (older maps exist but not published) FMI forest resources map Parts of country yearly Full coverage 2020 Special set of plots No (legislation to open it under work) Norway SR16 Forest resources map DAP / ALS 2017 Trøndelag region available SATSKOG 2007 Sweden SLU forest map 2000, 2005, 2010 100 % Skogliga grunddata 2016 100% (Except alpine areas) 12.5m × 12.5m National Corine 10m × 10 m-25m × 25 m 9.12.2018

Remote sensing will be used more in future Local pivotal method (LPM) uses auxiliary information for obtaining a balanced sample design taken into use in Sweden tested in Finland New estimation methods are being tested and introduced to operational NFI model-assisted estimation post-stratification small area estimation 9.12.2018

Forest management inventory FMI   Method used Who pays Subsidies available Coordination Denmark Visual assessment of aerial photos in combination with field measurements Forest owner no Finland Area-based laser scanning  (Government) Visual assessment in the field (Planning organization) Government (computationally recommended treatments) Forest owner (plan) yes (data collection) no (plan) Norway Visual assessment of aerial photos Area-based laser scanning yes (currently about 57%) (by the region forester) Sweden 9.12.2018

Users of the FMI data FMI data traditionally collected for the decision making of forest owner Harvest scheduling Silvicultural decisions Forest industry and forest service providers have increasing needs for the FMI data Timber trade Wood procurement Planning for end-use of timber for different products Forest administration as well Monitoring cuttings, regeneration, natural hazards Inspections of forest operations 9.12.2018

Each decision problem has unique requirements for the contents and accuracy of the data Monitoring cuttings Timeliness critical → automatic detection from satellite images needed Timber trade Species-level timber assortments estimates and quality assessments critical → field information so far most accurate Planning end-use of timber and optimal bucking Accurate diameter distribution estimates critical → field information so far most accurate No single data acquisition method can fulfill all requirements 9.12.2018

Synergies between NFI and FMI possible NFI and FMI are designed for different purposes NFI for national / regional statistics and decision making FMI for local (stand-level / estate level) decision making In the future the scopes are getting closer due to utilization of remote sensing Remote sensing and NFI data can be combined to produce accurate information also at stand level Remote sensing used for collecting FMI data for larger regions and with shorter intervals 9.12.2018

Synergies require research on optimal approach Are locally selected plots needed for FMI in addition to NFI plots? Does the accuracy of ALS and DAP campaigns reduce due to sensor / weather condition / forest characteristics variation when data is collected from larger areas? What is the accuracy required for the produced data to be useful for the original and new purposes? CARISMA network is working on answering these questions 9.12.2018