Role of fire severity in controlling patterns of stand dominance following wildfire in boreal forests

Abstract

Global trends of climate warming have been particularly pronounced in northern latitudes, and have been linked to an intensification of the fire regime in Arctic and boreal ecosystems. Increases in fire frequency, extent, and severity that have been observed over the past several decades are expected to continue under a warming climate. Severe fires can drastically reduce or remove the deep organic layers that accumulate in mature black spruce forests. Extensive studies in the boreal forests of interior Alaska and Canada have shown that parts of the landscape that undergo severe burning provide favorable seedbeds for the recruitment of deciduous tree seedlings, and thereby reduce the relative abundance of coniferous seedling recruitment in these areas shortly after fire. The persistence of deciduous species such as aspen beyond the seedling recruitment and establishment stage is as yet relatively unknown. To address this knowledge gap, I asked the question: is increased deciduous recruitment observed in severely burned areas transient, or does it result in persistent changes in stand composition later in succession? I examined changes in relative dominance patterns of aspen and black spruce that had occurred between 8 and 14 years post-fire along an organic layer depth gradient within a single burn. I found that patterns of relative species dominance established shortly after fire persisted into the second decade of succession, resulting in productive aspendominated stands in severely burned areas with shallow organic layers, and black spruce dominated stands in lightly burned areas with deep organic layers. These patterns of stand dominance in relation to post-fire organic layer depth were also observed in several other burns in the region. Therefore, deep burning fires are likely to result in a persistent shift from black spruce to aspen dominance in severely burned parts of the boreal forest. In order to understand how variation in organic layer depth is driving these alternate successional pathways, I measured nutrient uptake rates of aspen and spruce in severely and lightly burned sites within a single burn. I also examined relationships between post-fire organic layer depth and a suite of soil variables, and evaluated the relative importance of these soil variables in explaining variation in stand level aspen biomass, spruce biomass, and the relative dominance of aspen vs. spruce. I found that variations in post-fire organic layer depth result in contrasting soil environments, with soils in shallow organic layer sites being warmer, drier, and more alkaline than soils in deep organic layer sites. Variations in aspen biomass and aspen: spruce biomass were largely being driven by substrate conditions, whereas stand level spruce biomass was less sensitive to these same variations in soil conditions. Nutrient uptake rates of both aspen and spruce were higher in severely burned areas with shallow organic layers, but the differences between species were magnified by stand biomass patterns in relation to post-fire organic layer depth. My results suggest that the positive effects of soil conditions associated with mineral soil substrates extend well beyond the initial seedling recruitment phase, and may continue to influence aspen growth rates into the second decade of succession resulting in the differential patterns of biomass accumulation and stand dominance in relation to post-fire organic layer depth. With the predicted increase in fire severity and shortening of the fire cycle, the proportion of aspen dominated stands on the landscape is likely to increase, which will incur substantial changes in ecosystem function (e.g., land-atmosphere energy exchange, C and N storage, nutrient cycling, net primary productivity, and wildlife habitat quality) compared to the current forests dominated by conifers.