Inclusion of CO2 fluxes in a coupled mesoscale land surface and atmospheric model

Markus Uebel, Prabhakar Shrestha, Mauro Sulis and Andreas Bott

University of Bonn, Meteorological Institute, email: muebel@uni-bonn.de

1) Motivation

An essential part of numerical weather prediction models is the accurate simulation

of the interaction of the land surface with the lower atmosphere. We use a fully

coupled model system (COSMO-CLM-ParFlow) that comprehensively calculates

the exchange processes between the soil, the vegetation and the atmosphere.

Field measurements on the regional scale indicate distinct spatio-temporal

heterogeneities in the distribution of atmospheric carbon dioxide (CO2). This

variability of CO2 induces a direct response on the stomatal resistance of plants

resulting in a modified transpiration that influences the near surface atmospheric

moisture. Thus, for a consistent modeling of latent and sensible heat fluxes the net

CO2 flux between the land surface and the atmosphere will be included in our

coupled model system. This flux consists of the photosynthesis rate A as an atmos-

pheric sink as well as plant and soil respiration, Rplant and Rsoil as sources for the

atmospheric CO2 concentration (see Fig. 1). Now, the atmospheric CO2 is no longer

a constant value but a prognostic variable used for calculating canopy processes.

Figure 1: CO2 fluxes between the land surface and the atmosphere

2) Coupling of CO2

The atmospheric CO2 distribution

is initialized in the atmospheric

model COSMO and sent to the

external coupler OASIS which

organizes the downscaling to the

finer grid resolution of the

Community Land Model (CLM)

and unit conversions of all

variables and fluxes that are

exchanged between the models.

The CLM receives the CO2

partial pressure and uses this for

the calculation of the CO2 fluxes

(see Fig. 1). The determined net

CO2 flux is sent to the OASIS

coupler to be upscaled to the

atmospheric grid. The atmospheric

CO2 content is updated by this net

CO2 flux and the COSMO model

performes the atmospheric

transport. This coupling cycle is

repeated every coupling time step.

The coupling frequency can also

be varied with the OASIS coupler. Figure 2: Coupling cycle of CO2 with the OASIS coupler

for the coupling option COSMO – CLM

3) Model simulations

The selected day (8th of May 2008) was dominated by fair weather conditions with unattenuated

solar radiation over the western part of Germany. All model simulations with the coupled model

system COSMO – CLM were initialized with 390 ppmv of CO2 in all atmospheric model layers:

Reference run: no coupling of CO2 fluxes (standard model coupling)

Turbulence run: coupling of photosynthesis rate A and vertical turbulent mixing of CO2 in

the COSMO model (no advection)

analysis of the direct effects of plant type depending photosynthesis rates

Real run: coupling of photosynthesis rate A, influence of all transport processes on CO2

analysis of expected 3D-patterns of CO2 in the atmosphere

4) Atmospheric CO2 transport

The atmospheric model of our coupled model system is the non-hydrostatic, incompressible local

weather prediction model COSMO provided by the German Meteorological Service (DWD).

With this model we perform mesoscale weather forecasts with a grid resolution of 1km. For the

atmospheric transport of the specific CO2 content qCO2 a passive tracer has been included.

5) Canopy processes

6) Summary and outlook

The simulations show that the coupling of the photosynthesis rate A with the COSMO model results

in a reduction of the near surface atmospheric CO2 content and in a diurnal vertical variation of the

CO2 distribution due to vertical turbulence and advection with the wind field. This decrease of

atmospheric CO2 leads to a reduced photosynthesis rate and as a direct consequence to an increase

of plant transpiration compared with the reference run (Fig. 7 and 8). Plants react on the modified

availability of CO2 with opening their stomata to take up more CO2 for producing photosynthesis.

In our future research the plant respiration as well as the soil respiration will be included in our

model system to calculate the accurate net CO2 flux which will be exchanged between the CLM

and the COSMO model. We will perform sensitivity studies both for idealized cases and for

different real weather conditions. Different atmospheric CO2 distributions will be used in order to

assess the response of the stomatal resistance and the moisture and heat fluxes on these variations.

Modifications in the atmospheric moisture distribution and their possible impact on the simulation

results will be analyzed. Further, we will validate our model results with field and remote sensing

measurements performed by other working groups of the TR32 project.

The photosynthesis rate A as well as the plant

transpiration TP is controlled by the stomatal

resistance rst which describes the permeability

of the stomata of leaves:

ei* water vapor saturation pressure inside the leaf

b minimum stomatal conductance

Standard model coupling:

atmospheric pressure Patm(lat,lon,t)

water vapor pressure at leaf surface es (lat,lon,t)

CO2 partial pressure at leaf surface cs constant (every time step)

Coupling of CO2 fluxes:

atmospheric pressure Patm(lat,lon,t)

water vapor pressure at leaf surface es (lat,lon,t)

CO2 partial pressure at leaf surface cs (lat,lon,t)

prognostic

(atmospheric forcing variables)

prognostic

(atmospheric forcing variables)

Figure 6: Transpiration and CO2 uptake through stomata

(leaf cross section)

Figure 7: Difference of photosynthesis rate [μmol(CO2)m-2s-1]

between standard coupling and CO2 coupling

(reference – real): 05/08/2008, 09 UTC

Figure 8: Difference of transpiration [Wm-2] between standard

coupling and CO2 coupling (reference – real):

05/08/2008, 09 UTC

After the coupling with the CLM the tracer becomes active (influence on other variables).

local tendency advection sources/sinks subgrid scale processes

Figure 3: Specific CO2 content qCO2 [g kg-1] in the lowermost model layer of the COSMO model: 05/08/2008, 09 UTC

(a) turbulence run, (b) real run (red cross: position of the vertical profile, dashed line: position of the cross-section)

Figure 5: Vertical profiles of specific CO2 content qCO2 [g kg-1]

in the PBL at different times (real run): 05/08/2008

Figure 4: Vertical cross-section of the specific CO2 content qCO2

[g kg-1] in the PBL (real run): 05/08/2008, 10 UTC

Budget

equation

The spatial patterns in Fig. 3a) are mainly the result of different photosynthesis rates depending on

the plant type whereas in b) also advective transport occurs. The red regions in a) are urban areas

or bare soil without photosynthesis so that the initial concentration remains.

(a) (b)

Acknowledgements: The authors thank the German Meteorological Service (DWD) for providing the COSMO model

and the analysis data needed for driving the model. We gratefully acknowledge financial support

from SFB/TR32 “Patterns in Soil-Vegetation-Atmosphere Systems: Monitoring, Modelling and Data Assimilation” funded by the Deutsche Forschungsgemeinschaft (DFG). www.tr32.de

References: Baldauf, M. et al. (2011). “Operational convective-scale numerical weather prediction with the COSMO model: Description and sensitivities”. In: Mon. Wea. Rev., 139, pp. 3887-3905

Collatz, G.J. et al. (1991). “Physiological and environmental regulation of stomatal conductance, photosynthesis and transpiration: a model that includes a laminar boundary layer”. In: Agric. For. Meteorol. 54, pp. 107 –136.

Dolman, A.J. et al. (2006). “The CarboEurope regional experiment strategy”. In: University of Wollongong Research Online.

Shrestha P. et al. (2012). „Development of a Scale-Consistent Soil-Vegetation-Atmosphere Modeling System Using COSMO, Community Land Model and ParFlow“ (to be submitted)

TR32