Modelling DOC production, transport and retention in forest soils, exploiting recent 14C enrichment at Oak Ridge, USA

The incorporation of the carbon of plant litter into forest soils, and how long it spends in different soil layers (horizons), are important factors in the global carbon cycle, and therefore in the rate and extent of climate change. In many soils, downward movement of dissolved organic carbon (DOC) in percolating water is the main means by which carbon is transferred to deeper soil layers, where long-term carbon storage can occur. Moreover, DOC is an important transporter of metals, nutrients and organic contaminants in soils, and from soils to water. DyDOC is a process-based model designed to simulate the soil carbon cycle, including DOC generation and transport. This project aims to improve the model by calibrating it on a unique data set that is being collected following the localised enrichment of atmospheric CO2, and hence forest vegetation, due to the release of carbon-14 from an incinerator close to the Oak Ridge Reservation in Tennessee USA. The data collection is being carried out by American colleagues in a project called EBIS (Enriched Background Isotope Study). Carbon-14 is a natural radioactive isotope, well-known for its application in 'carbon dating', for samples of ages in the range 100 to tens of thousands of years. However, a second way in which it can be used in ecosystem research into the carbon cycle is through the enrichment of atmospheric CO2 by human activities. Until recently, this only meant 'bomb carbon', enrichment caused by atmospheric weapons testing in the middle of the 20th Century, which approximately doubled the worldwide 14C level. The 'bomb-labelled' carbon can be traced in different soil pools and, through the use of an appropriate model, its rate of incorporation and loss can be deduced. This has already been used to calibrate DyDOC at sites in Europe. But now a further localised label has been generated by the emission of carbon-14 from the incinerator at Oak Ridge during 1999, and, thanks to the swift actions of local scientists in setting up a monitoring programme and experimental studies in EBIS, much more detailed isotope data are consequently available for modelling. The work proposed here is to make use of the EBIS data to calibrate DyDOC much more extensively and precisely than would otherwise be possible. The work in the project will firstly comprise the collation of the EBIS data on soil carbon 14C, the amounts of carbon in different soil pools, and DOC concentrations, together with other local data (soil depths, rainfall, temperature, litterfall etc) characterising the Oak Ridge forests. We will then run the model, adjusting its parameters in order to match as closely as possible the measured values. Additional model runs will be done to see how sensitive the outputs are to changes in the parameter values. The calibrated model will be used to forecast (a) the evolution of the carbon-14 labelling of soil carbon pools and DOC, in readiness for further testing as more data are collected in the future, and (b) the dependence of soil carbon cycling and DOC formation and transport on external environmental factors such as temperature, rainfall and litter inputs. The results should provide useful information to assess the possible responses of forest soils to future variations in climate and other environmental conditions. This 'one-off' application of the model to a unique research site will provide information that can be used to limit the ranges of parameter values permissible in model applications to other sites, and therefore ultimately to more reliable regional-scale forecasting of forest soil carbon behaviour.