Aboveground net primary production (ANPP) in drylands is tightly linked to mean annual precipitation amount, where higher precipitation leads to higher production, while droughts yield opposite trends. In tandem with soil water availability are biogeochemical mechanisms that play a crucial role in dryland productivity; after water, nitrogen (N) availability is the most important limiting factor for ANPP. However, after 13 years of rainfall manipulation (20%, ambient, and 180%) at the Jornada Basin LTER (NM, USA), there was no effect of N fertilization (+10 g N m-2 yr-1) on ANPP in any of the precipitation treatments. We tested the hypothesis that there were no effects of fertilization because there was a shift in the N source. Unfertilized plots may access N supplied via N-fixation, a process that is reduced under high N-availability conditions. We analyzed the δ15N ratios of leaves from Bouteloua eriopoda and Prosopis glandulosa and soil from five depths across all water and N treatments in 2011, 2012, and 2018. Our objective was to link N pools with their sinks to better understand nutrient cycling in a semi-arid grassland under extreme precipitation conditions.
Results/Conclusions
Soil δ15N ratios were enriched at all depths (>5‰) among all water and fertilization treatments compared to air, and this enrichment was more pronounced in surface layers (0-10 cm) when water availability was low and plant uptake improbable. Fertilization had an effect on B. eriopoda foliar δ15N ratios; we saw an increase from ranges of -2‰ to 2‰ in unfertilized plots to 3‰ to 5‰ in fertilized plots across all water treatments, indicating their sources of N to be derived from free-living soil microbial N-fixation and fertilizer, respectively. There were no distinguishable differences in δ15N ratios for P. glandulosa leaves (3.5‰ to 4.5‰) among either water or fertilization treatments, despite P. glandulosa being an N-fixing shrub. The enriched δ15N values are likely due to within-plant fractionation processes and do not act as a proxy from N-fixation rates alone. Foliar %N increases for both B. eriopoda and P. glandulosa, but the strength of this response depends on the year and treatment. Our results suggest that the flexibility of N sources and the rapidly shifting paths of the N cycle may explain the lack of ANPP response to both water and N availability.