Background/Question/Methods Nitrogen (N) inputs to forests have increased substantially due to human activities, and contrary to expectations N retention is high even where plant N sinks are saturated. This includes old-growth forests, which are expected to have steady-state N pools with a low capacity for new N retention. In many forests, soil organic matter (SOM) is the dominant sink for N inputs and the rapid transfer of inorganic N into SOM has motivated research to identify fast turnover pools that may be sinks for N.  We tested an alternative hypothesis, that high ecosystem N retention results rapid (minutes to months) immobilization of inorganic N in stable SOM (turnover times > 1 yr). We conducted two tests of this hypothesis. First, in the lab, we added ¹⁵NH₄⁺, ¹⁵NO₂^–, and ¹⁵NO₃^– to live and sterile O-horizon soil from old-growth conifer and hardwood stands of Pennsylvania, and recovered ¹⁵N after 15 minutes, 1 day, and 21 days of incubation. Second, we sprayed ¹⁵NO₄¹⁵NO₃ onto field plots of young (<10 years) and old (100 years) stands of Pennsylvania hardwoods. Organic and mineral soils were collected over time from 25 minutes to 1 year, after which stable and labile pools were separated.
Results/Conclusions In the laboratory experiment, high NO₂^– retention occurred within 15 minutes via abiotic pathways, while NO₃^– retention was not observed. Microbial immobilization of NH₄⁺ was observed in conifer soils by 21 days. In the field experiment, total ¹⁵N recovery in the O-horizon ranged from ~30% (young forests) to ~60% (old forests). Of this recovered N, ~40% was in the stable pool within 25 minutes, and ~75% was in the stable pool by one year in both young and old forests. These results support the hypothesis that inorganic N immobilization in stable SOM occurs much more rapidly than the pool decomposes, and that N form, plant species composition, and successional status influence these rates. Additionally, because our later-successional forests had larger SOM pools, expressing our results on an aerial basis suggests that forests that are often expected to have steady-state N pools have a greater N retention capacity than young forests with aggrading plant N pools.