Recovery of plant diversity following N cessation: effects of recruitment, litter, and elevated N cycling

Plant species richness has declined and composition shifted in response to elevated atmospheric deposition of biologically active nitrogen over much of the industrialized world. Litter thickness, litter nitrogen (N) content, and soil N mineralization rates often remain elevated long after inputs cease, clouding the prospects that plant community diversity and composition would recover should N inputs be reduced. Here we determined how N cycling, litter accumulation, and recruitment limitation influenced community recovery following cessation of long‐term N inputs to prairie‐like grasslands. We alleviated each of these potential inhibitors through a two‐year full‐factorial experiment involving organic carbon addition, litter removal, and seed addition. Seed addition had the largest effect on increasing seedling and species numbers and may be necessary to overcome long‐term burial of seeds of target perennial grassland species. Litter removal increased light availability and bare sites for colonization, though it had little effect on reducing the biomass of competing neighbors or altering extractable soil N. Nonetheless, these positive influences were enough to lead to small increases in species richness within one year. We found that, although C addition quickly altered many factors assumed favorable for the target community (decreased N availability and biomass of nearby competitors, increased light and site availability), these changes were insufficient to positively impact species richness or seedling numbers over the experimental duration. However, only carbon addition had species‐specific effects on the existing plant community, suggesting that its apparent limited utility may be more a result of slow recovery under ambient recruitment rather than from a lack of a restorative effect. There were dramatic interactions among treatments, with the positive effects of litter removal largely negated by carbon addition, and the positive effects of seed addition generally amplified by litter removal. It remains unclear whether each mechanism explored here will induce community recovery, but over different temporal scales. Long‐term monitoring will help resolve these remaining questions. Regardless, our results suggest that reversal of species loss and compositional shifts from N deposition in prairies may be more inhibited by habitat fragmentation, recruitment limitation, and long‐term suppression of fire than from continued effects of elevated N.