Epsilon Archive for Student Projects

Abstract

Forest fire is a major natural disturbance that influence forest ecosystem structure and composition at both the stand and landscape level. In boreal Fennoscandia fire frequency has been very low over the last c.150 years due to effective fire suppression, although occasionally high-intensity fires escape initial attack and grow big. In the Swedish forest landscape mires of various types cover on average 15% of the land surface and are often relied upon as fire breaks during wildfire incidents. However, it is known that mires sometime actually burn, but so far there has not been any attempt to elucidate fuel characteristics of different type of mires and their potential as barriers or avenues for fire.

This study makes use of a large forest fire that occurred in the county of Västmanland in 2014 to investigate to what extent fire affect different mire types, and to analyse the fuel characteristics of each mire type.

The investigation focused on the three main mire types present in the area: Open mires (dominated by graminoids), shrub mires (dominated by dwarf-shrubs) and pine bogs (canopy of pine trees and dwarf-shrubs in the field layer). I used aerial photo interpretation of pre-fire and post-fire IRF photos to map mire types and the extent of non-burnt mire patches within the burnt area. Further, extensive field sampling was conducted both within and outside the burnt area: 77 sampling plots were surveyed by recording vegetation characteristics and estimating fire intensity. Also, live vegetation and litter was destructively sampled in unburnt mires to calculate fuel dry mass on an area basis.

Pine bogs has the highest surface fuel quantities of the mire types investigated and has the highest proportion of burnt area. Within the perimeter of the burnt area (13100 ha), almost 100% of the pine bog area burnt. Shrub mires with slightly lower fuel quantities burnt at 89,94% with the frequent present of smaller patches untouched by the fire. Open mires burned to a lower proportion (72,91%) than the other two types, particular in their edges. Presumably, the sphagnum mosses surface in their central parts was still wet despite the long drought, and that the graminoid litter was too sparse to carry the fire. Considering fire propagation at the landscape level, only the very large open mires were able to have an effect.

In order to predict the potential fire propagation in different mire types, additional fuel sampling and fire spread experiments would be needed. In comparison with closed forest, the surface vegetation of mires, and thus their fuel type, can more easily be mapped using remote sensing. Hopefully, detailed maps predicting fire spread potential of wetlands can be produced in the future, as a tool for operational fire suppression efforts.