Litter N retention over winter for a low and a high phenolic species in the alpine tundra

Abstract

Although soils are often the largest sink for N in terrestrial ecosystems, litter can immobilize N during the first year of decay and could affect N retention in seasonally snow-covered ecosystems that have high N losses during snowmelt, a time when soil N retention and plant uptake can be low, and litter N retention could be high. In alpine ecosystems, the annual loss of 0.5 g N m−2 is dominated by a yearly peak in N loss during snowmelt when frozen soils and declining winter microbial populations decrease soil N retention. Here, we aimed to quantify litter N immobilization over winter and identify the potential for high phenolic accumulation in an alpine plant to increase the capacity of litter to immobilize N. A one-year litter decomposition experiment was carried out using litter of an alpine species high in phenolics, Acomastylis rossii (R. Br.) Greene, and an alpine species with similar litter chemistry but low in phenolics, Deschampsia caespitosa (L.) Beauv. We also doubled litter production to characterize the capacity of litter to immobilize N when substrate is not limiting. D. caespitosa and A. rossii litter immobilized 0.1 ± 0.1 and 0.3 ± 0.1 g N m−2, respectively, over winter. Doubling litter production led to the immobilization of 1.2 ± 0.2 g N m−2 in A. rossii litter, but did not affect immobilization in D. caespitosa litter. Higher N immobilization in A. rossii litter than in D. caespitosa litter was consistent with our expectation of phenolic control; but the lack of an order of magnitude difference in litter N immobilization at the natural rate of litter production was not consistent with the 15-fold greater concentration of phenolics in A. rossii litter. Yet, A.␣rossii litter also lost significantly more mass over winter than D. caespitosa litter. By decomposing pre-senescent litter, the difference between the species in litter decay rates was eliminated, and A. rossii tissues retained an order of magnitude more N than D. caespitosa tissues. Litter mass loss explained over 50% of the variation in litter N retention, and this relationship implies that A. rossii litter needs to lose two times more mass than D. caespitosa litter prior to the initiation of litter N loss. These results indicate that litter N immobilization under the snow contributes to ecosystem N retention in the alpine tundra and strongly suggest that high phenolic accumulation in A. rossii increases the capacity of litter to immobilize N.