Apportioning sources of organic matter in streambed sediments: An integrated molecular and compound-specific stable isotope approach

We present a novel application for quantitatively apportioning sources of organic matter in streambed sediments via a coupled molecular and compound-specific isotope analysis (CSIA) of long-chain leaf wax n-alkane biomarkers using a Bayesian mixing model. Leaf wax extracts of 13 plant species were collected from across two environments (aquatic and terrestrial) and four plant functional types (trees, herbaceous perennials, and C-3 and C-4 graminoids) from the agricultural River Wensum catchment, UK. Seven isotopic (delta C-13(27), delta C-13(29), delta C-13(31), delta C-13(27-31), delta H-2(27), delta H-2(29), and delta H-2(27-29)) and two n-alkane ratio (average chain length (ACL), carbon preference index (CPI)) finger prints were derived, which successfully differentiated 93% of individual plant specimens by plant functional type. The delta H-2 values were the strongest discriminators of plants originating from different functional groups, with trees (delta H-2(27-29) = -208 parts per thousand to -164 parts per thousand) and C-3 graminoids (delta H-2(27-29) = -259 parts per thousand to -221 parts per thousand) providing the largest contrasts. The delta C-13 values provided strong discrimination between C-3 (delta C-13(27-31) = -37.5 parts per thousand to -33.8 parts per thousand) and C-4 (delta C-13(27-31) = -23.5 parts per thousand to -23.1 parts per thousand) plants, but neither delta C-13 nor delta H-2 values could uniquely differentiate aquatic and terrestrial species, emphasizing a stronger plant physiological/biochemical rather than environmental control over isotopic differences. ACL and CPI complemented isotopic discrimination, with significantly longer chain lengths recorded for trees and terrestrial plants compared with herbaceous perennials and aquatic species, respectively. Application of a comprehensive Bayesian mixing model for 18 streambed sediments collected between September 2013 and March 2014 revealed considerable temporal variability in the apportionment of organic matter sources. Median organic matter contributions ranged from 22% to 52% for trees, 29% to 50% for herbaceous perennials, 17% to 34% for C-3 graminoids and 3% to 7% for C-4 graminoids. The results presented here clearly demonstrate the effectiveness of an integrated molecular and stable isotope analysis for quantitatively apportioning, with uncertainty, plant-specific organic matter contributions to streambed sediments via a Bayesian mixing model approach. (C) 2015 Elsevier B.V. All rights reserved.