Explicit Coupling of Vegetation Responses and Shallow Cumulus accounting for Direct and Diffuse Radiation

Boundary-layer clouds and the terrestrial surface are closely linked since they both depend on and affect the radiation balance, carbon and water budgets and thermodynamic profiles of the boundary layer. Here, we focus on the interactions between shallow cumulus and the vegetated surface at short spatio-temporal scales. Clouds alter the surface conditions by shading areas. The spatial scale of the shaded surface ranges from meters to kilometers; the temporal scale from seconds to minutes. Under clouds, the direct component of the solar radiation is reduced, which is partially compensated by diffuse radiation, resulting in direct and diffuse components of shortwave radiation. This alters the amount and ratio between diffuse and direct Photosynthetically Active Radiation (PAR) at the canopy and strongly modifies the surface canopy temperature and leaf-to-air vapour pressure deficit. Consequently, this leads to changes in the carbon and water exchange between vegetation and atmosphere at diurnal scales, with corresponding modifications to the energy partitioning between sensible and latent heat flux. In turn, the change in the surface energy balance affects the atmospheric turbulence and therefore the heat and moisture transport in the lower atmosphere, which closes the loop by affecting the boundary-layer cloud formation and development.

By coupling a Large Eddy Simulation (LES), where we resolve the turbulent motions of the boundary layer, with a land surface model, which calculates stomatal conductance and carbon net absorption rate, we study the radiation-driven changes at the surface and their effect on surface fluxes and boundary layer evolution. We explicitly simulate boundary-layer clouds formed during the day above a vegetated surface and calculate the amount of diffuse and direct radiation under clouds using the delta-Eddington approach. Our land-surface model, accounting for in-canopy interactions between leaves and radiation, calculates the radiation profiles inside the canopy for both direct and diffuse radiation at three canopy heights and as a function of solar angle at sunlit and shaded leaves. We obtain stomatal conductance and net carbon absorption values using a physiological plant model, and upscale these results to canopy level by means of Gaussian integration before calculating the canopy resistance. Using this, the diurnal variability of photosynthesis, the surface energy balance, atmospheric turbulence and the characteristics of boundary-layer clouds during a daily cycle are examined.

In order to tackle the complexity of the investigated interactions, we design three experiments that conserve the incoming solar energy and determine the various responses of clouds and vegetation through the treatment of direct and diffuse radiation. In the reference experiment the partitioning between direct and diffuse radiation due to clouds is calculated using the delta-Eddington approach. In the two additional numerical experiments, we apply the same amount of radiation at the surface, but assume it to be fully direct or fully diffuse under clouds. To better isolate the effects of clouds at the surface we apply conditional averaging of the surface properties for clear sky and under different bins of cloud thickness. In addition, we conditionally average atmospheric properties at parts of the cloud (general cloud, cloud core, cloud updraft) to investigate how the partitioning between diffuse and direct radiation at the surface affects the water content and dynamics of these clouds. We find differences in photosynthesis rates below clouds as high as 20% for the same amount of radiation depending on the direct/diffuse partition for vegetation. Since the surface energy balance depends on canopy resistance, we find shade-driven dynamic heterogeneities in surface fluxes, which are also sensitive to the direct/diffuse radiation partition. We analyse the contributions to surface latent heat flux and net carbon absorption rate to determine the most relevant factors under different cloud regimes. Preliminary results show two terms driving the variations in plant net carbon absorption under a cloud: stomatal conductance and the carbon dioxide concentration gradient between leaf and atmosphere, which are affected by atmospheric variables such as radiation and vapour pressure deficit. Furthermore, results suggest the presence of two cloud-vegetation interaction regimes separated by a threshold in optical thickness of around 7, being 5 a characteristic value for shallow clouds. For a lower optical thickness, vegetation shows higher photosynthesis rates and latent heat fluxes than under clear sky conditions, even if total radiation is lower. For higher cloud optical thickness values, photosynthesis rates decline and all surface fluxes decrease due to reduced radiation intensities at the surface. Sensible heat flux always decreases under a cloud. Finally, our results suggest that direct/diffuse radiation partitioning at the surface increases the water content and development of clouds compared to a situation in which all radiation would be direct.