Sources of sediment-bound organic matter infiltrating spawning gravels during the incubation and emergence life stages of salmonids

The biodegradation of organic matter ingressing spawning gravels in rivers exerts an oxygen demand which is believed to contribute to detrimental impacts on aquatic ecology including salmonids. Catchment management strategies therefore require reliable information on the key sources of sediment-bound organic matter. Accordingly, a novel source fingerprinting procedure based on analyses of bulk stable 13C and 15N isotope values and organic molecular structures detected using near infrared reflectance (NIR) spectroscopy was tested for assessing the primary sources of sediment-bound organic matter infiltrating artificial Atlantic salmon spawning redds in five rivers across England and Wales. Statistically-verified source fingerprints were identified using a combination of the Kruskal–Wallis H-test, principal component analysis and GA-driven discriminant function analysis. Interstitial sediment samples were obtained from artificial redds using retrievable basket traps inserted at the start of the salmonid spawning season and extracted subsequently in conjunction with critical juvenile phases (eyeing, hatch, emergence, late spawning) of fish development associated with incubation and emergence. Over the duration of these four basket extractions, the overall relative frequency-weighted average median source contributions to the interstitial sediment-bound organic matter sampled in the study rivers ranged between 26% (full uncertainty range 0–100%) and 44% (full uncertainty range 0–100%) for farm yard manures/slurries, 11% (full uncertainty range 0–75%) and 48% (full uncertainty range 0–99%) for damaged road verges, 16% (full uncertainty range (0–78%) and 52% (full uncertainty range (0–100%) for decaying instream vegetation and 4% (full uncertainty range 0–31%) and 10% (full uncertainty range (0–44%) for human septic waste. The results of mass conservation tests suggest that the procedure combining bulk 13C and 15N isotope values and NIR spectroscopy data on organic molecular structures is sensitive to the risks of significant non-conservative tracer behaviour in the fluvial environment and will therefore not necessarily work at all in-channel sites in all catchments.