Quantifying ratios of suspended sediment sources in forested headwater streams following timber-harvesting operations

Historically, timber-harvesting has increased fine sediment inputs to streams due to increased hillslope and streambank erosion and mass wasting along roads. However, under modern best management practices, the relative importance and variability of these sources is poorly understood. We present preliminary results from an ongoing study investigating the primary sources of suspended sediment in Oregon Coast Range streams influenced by timber harvesting. We instrumented two catchments, Enos Creek (harvested 2016) and Scheele Creek (reference) in fall 2016. Phillips samplers (5-6 per catchment) have been deployed longitudinally down the streams to enable robust characterization of suspended sediments—the collected samples integrate the chemical signatures of upstream sediment exports. We will collect samples monthly over 2 wet seasons and return to the laboratory to analyze the sediment using source fingerprinting approaches. The fingerprinting technique compares the chemical properties of stream sediment samples with the chemical properties of potential source areas, including 1) roads, 2) stream banks, and 3) hillslopes. To design a robust model for sediment-source identification, different types of chemical data are required—we will analyze sediment samples using a combination of: a) stable isotopes and C/N ratios (i.e., δ15N, δ13C, and C/N), b) geochemistry (Fe, K, and Ca), and c) radiogenic isotopes (137Cs and 210Pb). At the harvested site, the C/N ratios of the streambanks (17.9 ± 3.8) and the hillslopes (26.4 ± 4.8) are significantly different from one another (p = .016). C/N ratios of the suspended sediment (20.5 ± 2.0) are intermediate values between potential endmembers and behave conservatively with transport. The C/N ratios of the suspended sediment appear unaffected by roads (18.9 ± 8.7) along specific sections of the stream, suggesting that roads are not a primary sediment contributor. Under this assumption, the suspended sediment is, on average, comprised of 69.5% streambank sediments and 30.5% hillslope sediments. Additional analyses are required (and in progress) to support these implications and to further interpret the importance and variability of suspended sediment sources through both space (from head to outlet) and time.