Estimates of N2-fixation from variation in the natural abundance of 15N in Sonoran desert ecosystems.

The 15N abundance of tissues of five Prosopis specimens at our primary study site (a Prosopis woodland at Harper's Well in the Sonoran desert of Southern California) was determined over two growing seasons 1980 and 1981. The 15N abundance of soil and of tissues of presumed non-N2-fixing (control) plants was also measured. Prosopis tissues were significantly lower in 15N than either soil N or corresponding tissues of presumed non-N2-fixing plants which derive their N entirely from soil. Soil N was also significantly higher in 15N than atmospheric N2. We conclude that it is feasible to use variations in the natural abundance of 15N as an index of N2-fixation in this kind of ecosystem, and that N2-fixation is of considerable importance to Prosopis growing at this site.We also determined the 15N abundance of leaf tissue of presumed N2-fixing and control plants growing at the same site at six additional sites (five in the Sonoran desert of southern California and one in Baja California, Mexico near the town of Catavina). Four of these additional sites were dominated by Prosopis and two were mixed communities. There were statistically significant differences between the 15N abundances of the pooled legume population and control plants at all sites, although not every legume specimen exhibited this difference. From 15N abundance data we estimated the fractional contribution of biologically fixed N to the N economy of desert legumes. We concluded that N2-fixation is very important to Prosopis at six of seven sites in the Sonoran Desert. At the site where Prosopis did not appear to be fixing N2, N2-fixation was important only for legumes of the sub-family Papilionoideae, Lupinus, Dalea, Astragalus and Lotus.