Carbon isotope geochemistry of the Santa Clara River

The Santa Clara River is a prototypical small mountainous river, with a headwater height greater than 1000 m and a basin area smaller than 10,000 m 2. Although individual small mountainous rivers export trivial amounts of sediment and carbon to the ocean, as a group these rivers may export a major fraction (as much as 50%) of the total global river sediment flux [Milliman and Syvitski, 1992], making their geochemistry relevant the study of the ocean's carbon cycle. In addition, many small rivers export sediment in a few high flux events, causing massive, sporadic discharge of carbon onto coastal shelves, discharge conditions very different from those of large rivers. This class of rivers is an end‐member of the river‐ocean carbon exchange system,. opposite the Earth's largest river, the Amazon. The carbon mass and isotopic properties of the Santa Clara River are significantly different from previously studied large rivers. During the 1997–1998 winter, all Santa Clara carbon pools were old, with flux‐weighted average Δl4C values of−428±76‰ for particulate organic carbon, −73±31‰ for dissolved organic carbon, and−644±58‰ for black carbon. The age of exported carbon is primarily due to the deep erosion of old soils and not to inclusion of fossil fuel carbon. Additionally, the δ13C signatures of exported carbon pools were high relative to terrestrial carbon, bearing a signature quite similar to marine carbon (average particulate organic carbon (POC) δ13C = −22.2±0.8‰). The Santa Clara's estuary is small and drains onto the narrow eastern Pacific coastal margin, exporting this old soil organic matter directly into the ocean. If the Santa Clara export patterns are representative of this class of rivers, they may be a significant source of refractory terrestrial carbon to the ocean.

The carbon isotope geochemistry of the Santa Clara River is relevant to understanding the ocean's particulate and dissolved organic carbon cycles (DOC and POC) and the controls on ocean sedimentary organic carbon sequestration. The burial of POC in oceans accounts for the presence of 02 in the Earth's atmosphere and represents the final sink removing carbon (C) from rapid circulation between the atmosphere, biosphere, and the oceans [Berner, 1989;Hedges and Keil, 1995]. The oceanic DOC pool contains approximately as much carbon as the atmospheric CO2 pool and the -6000 year 14C age of deep Pacific DOC reflects slow exchange with the biosphere [Williams and Druffel, 1987]. As significant as these forms of carbon are, complete characterization of their sources remains elusive. Rivers export enough DOC to account for the ocean's entire DOC pool [Williams and Druffel, 1987] and enough POC to account for all of the carbon stored annually in ocean sediments [Hedges and Keil, 1995;Meybeck, 1982].
However, studies of tracers characteristic of organic matter derived from large rivers (like the Amazon) suggest that only small amounts of riverine 407 . '  It is also clear from studies of large rivers that the fates of DOC and POC are intertwined through suspended-dissolved carbon interconversion at the river-ocean margin. Keilet al. [1997] have shown that suspended mineral material experiences replacement of 813C-depleted terrestrial carbon with 8]3C-enriched marine material in transition from rivers to continental shelves. Estuarine and oceanic adsorption and desorption of mineral-associated C is influenced by the mineralogy of associated particles, the decompositional history of the organic matter, and the local estuarine dynamics [Hedges and Keil, 1999]. Tracer studies based on the physical, isotopic, and biomarker properties of very large rivers like the Amazon show that these adsorption/desorption processes work to enhance remineralization of terrestrial organic matter and inhibit its transport to the open ocean. However, this may not be the whole story: the physical and isotopic properties of OM exported by the Santa Clara are quite different from those of the Amazon, and combined, small mountainous rivers export much more of the global river sediment flux than the Amazon [Milliman and Syvitski, 1992]. Small river geochemistry is an important missing component in our understanding of the role of rivers in the ocean's carbon cycle.
The sediment transport processes at the mouth of the Santa Clara River are substantially different from those of larger rivers in terms of estuarine processes, shelf deposition, and the coupling between sediment discharge and coastal storage. Because the Santa Clara flows intermittently, it has created only a small, seasonal estuary. During the flood events responsible for over 90% of the Santa Clara's sediment export, flow is massive and fast, delivering terrestrial organic matter directly to the Pacific Ocean without the benefit of the estuarine processing typical of larger rivers like the Amazon.
The eastern rim of the Pacific lacks the broad, shallow shelf present in both the north and south Atlantic. Broad continental shelves such as those of the northwest Atlantic trap terrestrial sediments, allowing coastal reworking of organic matter and impeding sediment delivery to the deep ocean. However, because the Eastern Pacific shelf is much narrower, sediments borne by rivers like the Santa Clara are more likely to survive coastal processing and reach the deep ocean. Additionally, the Santa Clara's discharge events are caused by California's winter storms which concurrently impact the Pacific shelf, connecting river discharge to the offshore storms which mobilize coastal sediments Sherwood et al., 1994]. The narrowness of the Pacific margin and the coupling between river discharge and coastal storms creates conditions ideal for the transport of river organic matter to the open ocean.
The 1997-1998 E1 Nifio winter rains caused flooding in the south-central California area comparable only to the 1969 E1 Nifio, which caused the largest floods recorded in the preceding 50 years. The extraordinary 1997-1998 winter flooding created the opportunity to study the geochemistry of a number of high-flux events from the Santa Clara in one year. Plate 1 (top) shows the Santa Clara under low-flow conditions and Plate 1 (bottom) shows the Santa Clara during one medium flow event (February 14, 1998).
Along with POC and DOC, we measured the concentration and isotopic signature of black carbon (BC) in the particulate carbon exported by the Santa Clara. BC is a class of solid, refractory carbon compounds produced during biomass burning and fossil fuel combustion (for information on the BC cycle, see Kuhlbusch [1998]), making BC a tracer of terrestrial organic carbon. Where BC analyses have been performed, BC makes up a measurable component of C in most sediments [e.g., Griffin and Goldberg, 1975;Gustafsson and Gschwend, 1998;Herring, 1985;Masiello and Druffel, 1998;Smith et al., 1973;Verardo, 1997], suggesting that terrestrial organic matter is a measureable component of sequestered carbon in at least abyssal ocean sediments. Radiocarbon measurements of BC extracted from deep-sea sediments point to rivers and/or the ocean's DOC pool as potential conduits transporting this terrestrial material to the deep ocean [Masiello and Druffel, 1998].

Methods and Research Site
Through the 1997-1998 season we measured the concentration, /5•3C, and A14C of organic carbon in three pools: DOC, POC, and black carbon in POC. Additionally, samples were collected for salinity to verify that there was no intrusion of ocean water during high-flow samples (see data in Table 1). We collected two low-flow samples, one on  Here total = total material in sample, susp = suspended particulates in sample, and sink = sinking particulates in sample.
Filters were acidified to pH 2 with 5% H3PO4 for 24 hours, dried, and then cornbusted with prefired CuO and Ag foil in quartz tubes ar 850øC for 2 hours [Druffel et al., 1992]  Although the U.S. Geological Survey (USGS) has discontinued direct measurements of the sediment exported by the Santa Clara, measured water discharge is still continuously monitored. It is possible to estimate the total suspended sediment load from water discharge by using a rating curve, an empirical relationship developed from previous USGS measurements. From 1969 to 1976, the USGS measured the high flow suspended sediment concentration at USGS station 11114000 at Montalvo, California (Montalvo is located-2 km upstream from our sampling site and receives >99% of the Santa Clara's sediment load). Using this data, Brownlie and Taylor [1981] Table 2) and the data range (highest value -lowest value for high flow events, reported in Table 2

Liu, 1996]. The Lanyang Hsi A•4C values approach -875%0
(12.6% modern, or-16,600 ]4C years) and as much as 70% of the POC from this system may be eroded bedrock, much of it mobilized due to recent road construction [Kao and Liu, 1996].
Using the total volume of water exported by the Santa Clara in 1997-1998 and mean weighted average values (Table 2) Santa Clara flood events typically last -3 days. Because the residence time of sediment in the river is so short, it is unlikely that in-river biological processes (such as degradation or photosynthesis) change the isotopic signature of the transported carbon. Most likely, the isotopic signatures of carbon pools reflect their source: eroded soils, resuspended sediments, and bedrock t?om the Santa Clara drainage basin. The lack of river-estuary processing and the highly erosive nature of this class of rivers are the likely causes of the difference between the age of Santa Clara carbon pools and that of larger rivers.
The standard deviation of Santa Clara A•4C POC data (+ 76%0)is far greater than that attributable to analytical    [Meybeck, 1982], suggesting that small rivers are not a major direct source of DOC to the ocean, although DOC-POC interconversion may occur in the estuary and coastal regions.

uncertainty (+ 6%0). Changes in
Previous studies suggest that transport to the open ocean is a major fate of sediments discharged from small mountainous rivers, studies such as the STRATAFORM study on the Eel River. The Eel is also a small, mountainous river discharging onto the narrow, active North American western margin, and it has been shown that on century timescales, less than 20% of the Eel's terrestrial sediment load is trapped on the upper slope, and at least 60% is transported beyond the shelf-slope break [Sommerfield and Nittrouer, 1999]. The long-range transport of old, 15•3C enriched terrestrial organic carbon from small mountainous rivers is consistent with the results of Bauer and Druffel [1998], who showed that continental margins (including the Eastern North Pacific) are a source of •4C-depleted DOC and POC to the deep ocean.
Not only does this class of rivers have the potential to transport carbon effectively off-shore, but it is also the smaller clay and silt-sized minerals that are preferentially exported [Sommerfi'eld and Nittrouer, 1999]. This is significant because some of the biochemical markers used to detect terrestrial organic matter in sediments, such as lignins [Gogi Prahl et al., 1994] are poorly retained on clays. This suggests that small, episodic rivers like the Santa Clara may be exporting C that is difficult to detect in the ocean by some forms of biomarker analyses. Some biochemical markers, such as n-alkanes, have been shown to be transported with silt and clay-sized particles [Prahl et al., 1994], and measurements of these markers (along with lignins) in rivers would add substantially to our understanding of the processes which move carbon from soils to the deep ocean via continental margins.
The isotopic signatures of the Santa Clara's carbon pools point toward a different understanding of the fate of terrestrial carbon in the open ocean. The highly erosive conditions and effective coastal dispersion characteristic of rivers on active margins, coupled with the •4C age and refractivity of C observed in the Santa Clara suggest that more information on the geochemistry of small mountainous rivers may be needed to understand the role of rivers in the ocean's carbon cycle. Further measurements will be necessary to understand the fate of old carbon exported by this class of rivers.