Research Programs

Peter Thornton
FY2006 activities

TIIMES Theme:
BGS & BEACHON

Peter Thornton - CGD - TIIMES

Carbon-nitrogen interactions in CLM

Comparison of βL between C-only and C-N model configurations. a) Trends in βL over the period 2000 – 2100, for C-only simulations with increasing CO2 (Ca1 and Cb1), C-N simulation with increasing CO2 (CN1, Cn1x), and C-N simulation with increasing CO2 and mineral nitrogen deposition (CN3). Symbols indicate values for eleven C-only models participating in C4MIP (diamonds) and the mean of the C4MIP models (square). C4MIP results from Friedlingstein et al. [2006]. b) Trends in ratio of bL between C-N and C-only model configurations (CN1:Ca1, CN1x:Ca1, and CN3: Ca1) over the period 2000-2100. See text for discussion of CN1x experiment.

This project involves the application of the recently completed CLM3-CN model to evaluate the influence of interactions between terrestrial carbon and nitrogen cycles on the aspects of the global carbon cycle that are most critical to global-scale carbon-climate feedbacks. The main activity in 2006 has been completion of a series of global simulations in which CLM3-CN is forced by transient signals for historical and potential future CO₂ concentrations and rates of mineral nitrogen deposition, over the period 1850-2100. The objectives were to determine the sensitivity of the model to increases in CO₂ , and also the sensitivity to variation in temperature and precipitation, and to compare these sensitivities to a model without C-N interactions.

The study resulted in the following conclusions:

1. The sensitivity to increasing CO₂ concentration for the C-N model is about four times smaller than for the C-only model, while this sensitivity for the C-only model was in the same range as found for an analysis of nine other C-only models. So the inclusion of C-N coupling has a dramatic influence on one of the major feedback mechanisms in the coupled carbon-climate system, in the direction of a weaker negative feedback of terrestrial ecosystems on CO₂ concentration.
2. The magnitude of sensitivities to variation in temperature and precipitation were both reduced by introduction of the C-N coupling, but the signs are the same for C-N and C-only modes: a positive feedback (release of carbon from land) for increasing temperature, and a negative feedback (uptake on land) for increasing precipitation.
3. Under a transient scenario of increasing CO₂ concentration, the sensitivity of the land biosphere to temperature and precipitation variation increases for the C-only model, but decreases when the C-N coupling mechanisms are introduced.

The spatial patterns of these feedback mechanisms are complex, and the fact that there are likely to be competing positive and negative feedbacks makes it difficult to determine a priori what the influence of introducing the C-N mechanisms would be in a fully-coupled climate-carbon cycle simulation. A manuscript describing these results is in preparation for Global Biogeochemical Cycles.

Grid-enabled compute platform for high resolution surface weather interpolation and terrestrial biogeochemistry

Interpolated precipitation over parts of North America.

This was a technology development project to produce a user interface to two widely used models: Daymet (for interpolation and extrapolation of surface weather observations to a high-resolution grid) and Biome-BGC (for terrestrial carbon, nitrogen, water, and energy cycle modeling). The project involved close collaboration between CGD, SCD, and the Computer Science Dept. at CU, to design, implement, and test all the necessary technology components. New components include a web-based user interface (portal) to the science code, a Grid-Compute enabled interface to a remote computational cluster at CU, and modifications to the core science routines to convert to netCDF I/O. The final system was tested during the summer 2006 on an operational problem: generation of daily gridded surface weather fields over Canada at 10 km resolution, and integration of these results with 1km gridded surface weather fields over the U.S. A manuscript describing this implementation of the prototype system is in preparation.