Evaluating the relationship between wildfire extent and nitrogen dry deposition in a boreal forest in interior Alaska

Alaska wildfires may play an important role in nitrogen (N) dry deposition in Alaskan boreal forests. Here we used annual N dry deposition data measured by CASTNET at Denali National Park (DEN417) during 1999-2013, to evaluate the relationships between wildfire extent and N dry deposition in Alaska. We established six potential factors for multiple regression analysis, including burned area within 100 km of DEN417 (BA(100km)) and in other distant parts of Alaska (BA(AK)), the sum of indexes of North Atlantic Oscillation and Arctic Oscillation (OI), number of days with negative OI (OIday), precipitation (PRCP), and number of days with PRCP (PRCPday). Multiple regression analysis was conducted for both time scales, annual (using only annual values of factors) and six-month (using annual values of BA(AK) and BA(100km), and fire and non-fire seasons' values of other four factors) time scales. Together, BA(AK), BA(100km), and OIday, along with PRCPday in the case of the six-month scale, explained more than 92% of the interannual variation in N dry deposition. The influence of BA100km on N dry deposition was ten-fold greater than from BA(AK); the qualitative contribution was almost zero, however, due to the small BA(100km). BA(AK) was the leading explanatory factor, with a 15 +/- 14% contribution. We further calculated N dry deposition during 1950-2013 using the obtained regression equation and long-term records for the factors. The N dry deposition calculated for 1950-2013 revealed that an increased occurrence of wildfires during the 2000s led to the maximum N dry deposition exhibited during this decade. As a result, the effect of BA(AK) on N dry deposition remains sufficiently large, even when large possible uncertainties (>40%) in the measurement of N dry deposition are taken into account for the multiple regression analysis. (C) 2016 Elsevier B.V. and NIPR. All rights reserved.