Suitability and use of the N-isotope dilution method to estimate nitrogen ®xation by actinorhizal shrubs

Nitrogen ®xation rates were estimated by the N-isotope dilution method for Ceanothus velutinus and Purshia tridentata in the understory of central Oregon ponderosa pine forests. Field rates were measured in small pole-, large pole-, and sawtimbersized pine stands using two shrubs (Arctostaphylos patula, Ribes cereum) and one graminoid (Carex rossii) as non-®xing reference species. Shrub cover ranged from 23 to 59% for Ceanothus and from 9 to 15% for Purshia. Foliage samples were collected monthly following two applications of N ammonium sulfate (5 kg N haÿ1 at 10 atom% N) to compare N uptake patterns and determine the fraction of N derived from ®xation (Ndff). Several violations of the underlying assumption of the isotope dilution method that N ®xing and reference plants have access to similar pools of soil N were identi®ed. These included non-uniform distribution of N in the soil pro®le and dissimilarities in rooting volumes, N uptake patterns, and plant ages between the reference and N ®xing plants. Despite these compromises, there were only minor differences in percentage of Ndff when calculated independently using each reference species. Low N uptake by Ceanothus and Purshia evidently negated the requirement that N ®xing and reference plants have comparable spatial and temporal access to soil N. Both Ceanothus and Purshia were highly effective, obtaining more than 80% of their N from ®xation regardless of pine stand, shrub cover, or shrub age. Annual N ®xation by Ceanothus ranged from 4 to 15 kg haÿ1 yearÿ1. This rate is lower than previously reported for Ceanothus in other ecosystems in the western United States, yet is suf®cient to replace N losses from disturbances such as prescribed ®re. Purshia ®xed considerably less N, ca. 1 kg haÿ1 yearÿ1, roughly equivalent to the rate of atmospheric N deposition in central Oregon. # 2000 Elsevier Science B.V. All rights reserved.