Tree regeneration response to a shifting soil nutrient economy depends on mycorrhizal association and age

Abstract

Atmospheric nitrogen (N) deposition has led to an increase in N cycling and N availability. This increase in inorganic N is likely to impact forest ecosystems, although the responses remain uncertain. Most tree species are associated with one of two mycorrhiza types – arbuscular mycorrhizae (AM) or ectomycorrhiza (ECM). Due to functional differences in their ability to access soil nutrients, we might expect that increased N cycling and availability of inorganic N would benefit AM associated species, with negative or neutral impact for ECM associated species. This study addresses how the abundance of the regeneration layer responds to those shifting soil conditions, based on their mycorrhizal association. We used a long-term experiment located in a temperate deciduous forest, where the native acidic soils are low in nutrients. Soil treatments began in 2009, by adding lime and/or phosphate to raise pH and increase the availability of N and P. All trees ≥ 6.0 cm in 2010 were tagged and have been monitored with annual censuses. To quantity the density of the regeneration layer, seedlings and saplings were recorded in the control and limed plots in 2019. Trees that had recruited into the canopy (DBH ≥ 6.0 cm) were measured in 2020 in all treatment plots. Seedlings older than one year responded to the treatments as predicted, as AM seedlings increased by 42 % in the lime treatment (1.53 ± 0.38 individuals per m2 in control, to 2.17 ± 0.56 in lime) and ECM seedlings decreased by 49 % in response to liming (0.61 ± 0.14 individuals per m2 in control, to 0.31 ± 0.04 in lime). AM saplings also responded positively to liming, increasing 254 % from control to lime (from 0.013 ± 0.003 to 0.046 ± 0.018 individuals per m2), while ECM sapling abundance was neutral (0.063 ± 0.015 individuals per m2 in control, and 0.054 ± 0.009 in lime). The highest number of ECM recruits was found in the control plots (34.4 %), followed by phosphate (25.8 %), lime + phosphate (21.5 %), then lime (18.3 %). For AM associated tree species, the number of trees recruiting into the canopy was substantially higher in the phosphate treatment (39.4 %), with lower numbers in the other three treatments (26.8 % in control, 19.7 % in lime, and 14.1 % in lime + phosphate). Overall, we found a positive regeneration response in AM associated species to increasing nutrient availability which has important consequences for species composition in forests experiencing chronic N deposition.