Tracing Sediment Sources Using Mid‐infrared Spectroscopy in Arvorezinha Catchment, Southern Brazil

Abstract

The widespread adoption of the sediment fingerprinting approach to guide catchment management has been limited by the cost and the difficulty to prepare and process samples for geochemical and radionuclide analyses. Spectral properties have recently been shown to provide a rapid and cost‐efficient alternative for this purpose. The current research objective was (i) to quantify the sediment source contributions in a 1∙19‐km2 rural catchment of Southern Brazil by using mid‐infrared (MIR) spectroscopy and (ii) to compare these results with those obtained with geochemical approach and near‐infrared and ultraviolet–visible spectroscopy methods. The sediment sources to discriminate were cropland surface (n = 20), unpaved roads (n = 10) and stream channel banks (n = 10). Twenty‐nine suspended sediment samples were collected at the catchment outlet during nine significant flood events. The sources could be distinguished by MIR spectroscopy. Cropland and channel bank sources mainly differed in their clay mineral contents, but their similar organic matter content complicated the MIR‐model predictions. Unpaved road contributions were discriminated from the other sources by their lower organic carbon content. When the results of the current research based on MIR spectroscopy are compared with those obtained using other sediment fingerprinting approaches, based on geochemistry and near‐infrared and ultraviolet–visible spectroscopy, an overestimation of channel banks contribution and an underestimation of cropland and unpaved road contributions is found. These results suggest that MIR spectroscopy can provide a useful tool that is non‐destructive, rapid and cheap for tracing sediment sources in rural catchments and for guiding the implementation of soil and water conservation measures. Copyright © 2016 John Wiley & Sons, Ltd.

Provenance discrimination of fine sediments by mid‐infrared spectroscopy: Calibration and application to fluvial palaeo‐environmental reconstruction

“…River systems are influenced by climate, tectonic and anthropogenic impacts (Macklin & Lewin, ) which cause fluctuations in fluvial morpho‐dynamics and sediment composition (Hoffmann et al ., , ; Erkens et al ., ). Determining sediment provenance is thus an important tool to understand the evolution of fluvial environments at short timescales (Poulenard et al ., , ; Evrard et al ., ; Palazón et al ., ; Tiecher et al ., ) middle timescales (for example, millennia, Holocene period, Owens et al ., ; Foster et al ., , ; Hatfield & Maher, ) and long timescales (for example, Quaternary and pre‐Quaternary periods, Hagedorn & Boenigk, ; Suresh et al ., ; Mange & Otvos, ). Characterizing sediment sources is crucial to determine time‐transgressive changes in sediment provenance from sedimentary deposits (Sircombe, ).…”

Using source‐specific models to test the impact of sediment source classification on sediment fingerprinting

“…Based on the DRIFTS spectra of the sediment source samples, 17 characteristic absorption peaks were identified that are typical for soil samples (Figure ; Tiecher et al, ). The 3,695 and 3,620 cm −1 peaks are characteristic for aluminosilicates (kaolinite and micas), which are typically present in clays (Parikh, Goyne, Margenot, Mukome, & Calderón, ; Poulenard et al, ; Tiecher et al, ; Yang & Mouazen, ).…”

“…Based on the DRIFTS spectra of the sediment source samples, 17 characteristic absorption peaks were identified that are typical for soil samples (Figure ; Tiecher et al, ). The 3,695 and 3,620 cm −1 peaks are characteristic for aluminosilicates (kaolinite and micas), which are typically present in clays (Parikh, Goyne, Margenot, Mukome, & Calderón, ; Poulenard et al, ; Tiecher et al, ; Yang & Mouazen, ). The peaks around 2,920 and 2,850 cm −1 are generally attributed to organic matter (Poulenard et al, ; Reeves III, ; Tiecher et al, ), whereas the peak at 2,505 cm −1 is attributed to carbonates (inorganic carbon; Poulenard et al, ; Reeves III, Mccarty, & Reeves, ; Viscarra Rossel et al, ).…”

“…The 1,990; 1,870; and 1,785 cm −1 peaks are generally related to quartz (Qz), and 1,630 cm −1 to Qz and clay minerals. The peaks around 1,530 and 1,360 cm −1 are attributed to Qz and organic matter, whereas 1,160 cm −1 relates to clay minerals and organic matter (Tiecher et al, ). Finally, the 1,115 to 698 cm −1 peaks are attributed to the combination of clay minerals and Qz (Ge, Thomasson, & Morgan, ; Parikh et al, ; Reeves III, ; Viscarra Rossel, Walvoort, McBratney, Janik, & Skjemstad, ).…”

Temporal variation in suspended sediment transport: linking sediment sources and hydro‐meteorological drivers

“…Franz et al ., ; Palazón et al ., ; Chen et al ., ; Vale et al ., ; Tiecher et al ., ), which has evolved significantly over the past decades as a method to estimate sediment source contributions directly from river sediment (Davis and Fox, ; Mukundan et al ., ; Owens et al ., ; Pulley and Collins, ). Particularly interesting from a monitoring perspective is the recent development of methods based on infrared spectrometry (Poulenard et al ., ; Cooper et al ., ; Tiecher et al ., ), which are more time‐ and cost‐effective compared to traditional geochemical sediment fingerprinting methods (Collins et al ., ). These advancements enable sediment fingerprinting to be easier applied at a finer temporal frequency (i.e.…”