Isotopic discrimination during long-term decomposition in an arid land ecosystem

Discrimination in carbon and nitrogen isotopes of decomposing plant litter in the northern Chihuahuan Desert was determined for a 5-year period. Factors influencing isotopic change were assessed from inter-species comparisons of litter chemistry, mass loss patterns, and isotope values of associated soil. Average delta N-15(litter) values of buried roots increased 1.2 and 2.6 parts per thousand for Big Blue Stern (Schizachyrium gerardi, grass) and Varital (Drypetes glauca, hardwood) during the study, respectively. Small but inconsistent variations were observed for Slash Pine (Pinus elliotii, conifer) roots. Average delta N-15(litter) values of wooden dowels from Ramin (Gonystlylus bancanus, hardwood) increased ca. 2.0 parts per thousand during years 1-4, and then decreased slightly during year 5. Changes in delta N-15(litter) were independent of N content, and may reflect microbial fractionation or preferential retention of N-15 enriched substrates. Surprisingly, there was no clear relationship between litter N dynamics and C/N ratios. There were no discernable changes in delta C-13(litter) values for Gonystlylus bancanus and Finns elliotii. Average delta C-13(litter) values for Schizachyrium gerardi decreased similar to2.0 parts per thousand during years 0-2 and then increased slightly. In contrast, average delta C-13(litter) values for Drypetes glauca increased similar to0.5 parts per thousand from years 0-1 then remained relatively constant until decreasing slightly in year 5. delta C-13(litter) discrimination may have been masked by interfering delta C-13 fractionations or feedbacks between decomposers and litter chemistry. Our data indicate that isotopic discrimination is characteristic of early litter decay stages. These results may highlight aspects of isotope discrimination and nutrient cycling unique to arid land environments. Additional studies will be needed to confirm this. (C) 2001 Elsevier Science Ltd. All rights reserved.