Predicting snow damage in conifer forests using a mechanistic snow damage model and high-resolution snow accumulation data

ABSTRACT

Forest damage caused by heavy wet snow accumulation in the canopy is the second most important abiotic forest disturbance agent in Nordic conifer stands after wind. The extent and frequency of snow damage in the future climate in the Nordic region is a major uncertainty. Few mechanistic models of snow damage risk to trees exist that could support forest management scenario analysis and decision making. We propose a snow damage risk model consisting of a numerical weather prediction-based snow accumulation model for forest canopies and a mechanistic critical snow load model. Snow damage probability predictions were validated on snow breakage data from the winters of 2016 and 2018 covering 3.5 million individual trees in south-eastern Norway derived from pre- and post-damage aerial laser scanning campaigns. The proposed model demonstrated satisfactory damage and no-damage class separation with an AUC of 0.72 and 0.77 in Norway spruce and Scots pine, respectively, and an F1 score of 0.7 in conifers taller than 10 m that suffered moderate stem breakage. The model achieved a classification accuracy that is comparable to that of statistical models but is simpler and requires fewer inputs.

KEYWORDS:

• Snow damage
• mechanistic model
• critical snow load
• snow accumulation

Acknowledgements

The authors are grateful to Christer Sandum and Fritzøe Skoger AS for making available the ALS and forest inventory data and to Eva Solbjørg Flo Heggem for pre-processing the forest inventory data.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author, PZ, upon reasonable request. Restrictions may apply to the availability of the source data, which were used for this study and are available from the corresponding author with the permission of Fritzøe Skoger AS.

Additional information

Funding

This work was supported by the Research Council of Norway under Grant 301745 as part of the MARCSMAN (Maximizing the Resilience and Carbon Sequestration in Managed Norway Spruce Forests) project.