Possible ideas on carbon and nitrogen trophic fractionation of food chains: a new aspect of food‐chain stable isotope analysis in Lake Biwa, Lake Baikal, and the Mongolian grasslands

<jats:title>Abstract</jats:title> <jats:p> Trophic fractionation of carbon and nitrogen isotopes (Δδ <jats:sup>13</jats:sup> C, Δδ <jats:sup>15</jats:sup> N) was examined using previously complied databases for food chains in Lake Biwa, Lake Baikal, and Mongolian grassland. The following two features were clarified: (1) For each ecosystem, the ratios of trophic fractionation of carbon and nitrogen isotopes (Δδ <jats:sup>15</jats:sup> N/Δδ <jats:sup>13</jats:sup> C) throughout food chain could be obtained as the slope of linear regression line on the δ <jats:sup>15</jats:sup> N–δ <jats:sup>13</jats:sup> C plot. (2) Further, analysis of covariance (ANCOVA) revealed the slopes on δ <jats:sup>15</jats:sup> N–δ <jats:sup>13</jats:sup> C were not significantly different among these various ecosystems and allowed us to have the regression by setting δ <jats:sup>15</jats:sup> N as the response variable: δ <jats:sup>15</jats:sup> N = 1.61 δ <jats:sup>13</jats:sup> C + [ecosystem specific constant] with standard errors of [±0.41] and [±9.7] for the slope and the intercept, respectively. It was suggested that the slope of the regression (or the ratio Δδ <jats:sup>15</jats:sup> N/Δδ <jats:sup>13</jats:sup> C) could be applicable to more complicated food webs in case nitrogen and carbon isotope ratios of primary producers can be assumed constant in space and time within the ecosystems. The results from simple linear regression analyses coincided well with the ANCOVA results for these ecosystems, although there was some discrepancy between the results of the two statistical analyses. Possible factors that govern the linear relationship between δ <jats:sup>15</jats:sup> N and δ <jats:sup>13</jats:sup> C along a food chain are discussed together with a new scope for the stable isotope food chain analyses. </jats:p>