Fire History Sampling Strategy of Fire Intervals Associated with Mixed- to Full-Severity Fires in Southern Alberta, Canada

We propose a series of methods to conduct a field-based fire history in landscapes regulated by mixed- to high-severity fires, where repeated lethal fires create a scarcity of trees bearing more than a single scar and pose a challenge for calculating fire return intervals (FRIs). In the context of applied forest and fire management, we use a case study based on six sampling units across the Montane, Subalpine, and Upper Foothills natural subregions of the southern Canadian Rockies of Alberta. The methods use historical aerial photography and a targeted paired-plot sampling strategy along fire boundaries and edges of unburned island remnants. This process allows for the calculation of point FRIs even when fire scars are absent. The Kaplan-Meier nonparametric likelihood estimator is used for the fire frequency analysis. A total of 814 sampling sites were visited, from which 3,123 tree cross-sections containing 522 fire scars were collected. The period of data analysis ranged from the oldest stands sampled in the unit (1535-1770) to 1948, before effective fire suppression. Spatial variability in the probability median FRIs between natural regions was identified. It ranged from 26 to 35 years in the Montane and was 39 years for the single Upper Foothills unit sampled, whereas the two Subalpine units sampled recorded values of 65 and 85 years. We concluded that the approach was sound for landscape-scale study areas regulated by large fires with significant tree mortality and with an average fire interval greater than 20 years.