Particulate organic carbon mobilisation and export from temperate forested uplands

The transfer of organic carbon stored in continental biomass to geological
storage is an important pathway in the global carbon cycle, with the
potential to sequester significant amounts of carbon dioxide. Despite an
initial focus on active mountain belts as prime locations for such erosion,
temperate forested uplands also have a significant role to play. This thesis
investigates the origins, mobilisation and export of particulate organic
carbon (POC) in the headwaters of two representative temperate areas,
the Swiss Prealps and Western Oregon, and addresses the significance of
the results in a global context.
Broadly, organic carbon concentration as a percentage of the suspended
load is inversely correlated with clastic yield. Mean values in natural
catchments range from 1.5% in Switzerland to 10% in Oregon. By chemically
fingerprinting this POC and major organic carbon stores within
each catchment, using carbon and nitrogen elemental and stable isotopic
compositions, its provenance is determined. By monitoring its changing
concentration and composition over a range of discharge, the processes
by which it is mobilised are elucidated. In Switzerland, additional methods
including radiocarbon analysis, Raman spectroscopy and biomarker
geochemistry add further insights into sources and pathways.
Riverine POC in Switzerland derives from binary mixing between bedrock
and modern biomass with a soil-like composition, with little direct input
of plant matter. The hillslope and actively incising channel are strongly
coupled, allowing overland flow to deliver biogenic material directly to
the stream beyond a moderate discharge threshold. At this point, the
broad trend is reversed; the proportion of organic carbon now increases
with discharge and suspended sediment concentration. At higher
flows, more biomass is mobilised and the fraction of modern organic carbon in
the suspended load reaches 0.70, increased from 0.30 during background
conditions. In Oregon, little fossil organic carbon enters the suspended
load even where it is present. Instead, riverine POC derives from mixing
between soil-like material and foliage from a variety of plants. Overland
flow rarely develops and hillslopes are isolated from channels. Material
comes instead from the channel itself and immediately adjacent areas.
There is no systematic switch to POC addition; instead, continued dilution
by clastic material is observed as discharge increases.
Significant amounts of non-fossil organic carbon are thus mobilised in both
areas without the need for extreme events such as landsliding. Precipitation
is key: as soon as the rain stops, biomass supply ceases and fossil
carbon again dominates in Switzerland, while Oregon streams run clear
once more.
Relationships with discharge are used where possible to calculate longterm
export fluxes of total and non-fossil POC. In the most active Swiss
catchment, rating curves are integrated over 29-year discharge records,
giving fluxes of 23 + or -6 t km^-2 yr^-1 and 14 + or -5 t km^-2 yr^-1 respectively.
On the order of 6 t km^-2 yr^-1 of total POC are exported from the Oregon Cascades, of which about 100% is non-fossil. These represent near-end members of POC export in temperate forested uplands, with the other catchments forming a continuum between them. One Oregon catchment subjected to intensive recent logging shows dramatically different behaviour, marked by high clastic yield and a much-reduced fraction of modern organic carbon.
Ecosystem biology is shown to be the principal control on POC export
style, with lithology having a lesser influence.
Yields of non-fossil POC from the temperate forested uplands studied
are comparable to those from active mountain belts, yet the processes
responsible for them are much more widely applicable. Their collective
contribution to global land-ocean POC discharge may be greater than
previously thought, and their role in the carbon cycle -- including potential Earth -- climate feedbacks-- more significant.