Spring 2020 broke sunshine duration records across Western Europe. The Netherlands recorded the highest surface irradiance since 1928, exceeding the previous extreme of 2011 by 13%, and the diffuse fraction of the irradiance measured a record low percentage (38%). The coinciding irradiance extreme and a reduction in anthropogenic pollution due to COVID-19 measures triggered the hypothesis that cleaner-than-usual air contributed to the record. Based on analyses of ground-based and satellite observations and experiments with a radiative transfer model, we estimate a 1.3% (2.3 W m−2) increase in surface irradiance with respect to the 2010–2019 mean due to a low median aerosol optical depth, and a 17.6% (30.7 W m−2) increase due to several exceptionally dry days and a very low cloud fraction overall. Our analyses show that the reduced aerosols and contrails due to the COVID-19 measures are far less important in the irradiance record than the dry and particularly cloud-free weather.

The radiative transfer equations are well known, but radiation parametrizations in atmospheric models are computationally expensive. A promising tool for accelerating parametrizations is the use of machine learning techniques. In this study, we develop a machine learning-based parametrization for the gaseous optical properties by training neural networks to emulate a modern radiation parametrization (RRTMGP). To minimize computa- tional costs, we reduce the range of atmospheric conditions for which the neural networks are applicable and use machine-specific optimized BLAS functions to accelerate matrix computations. To generate training data, we use a set of randomly perturbed atmospheric profiles and calculate optical properties using RRTMGP. Predicted optical properties are highly accurate and the resulting radiative fluxes have average errors within 0.5 W m −2 compared to RRTMGP. Our neural network-based gas optics parametrization is up to four times faster than RRTMGP, depending on the size of the neural networks. We further test the trade-off between speed and accuracy by training neural networks for the narrow range of atmospheric conditions of a single large-eddy simulation, so smaller and therefore faster networks can achieve a desired accuracy. We conclude that our machine learning-based parametrization can speed-up radiative transfer computations while retaining high accuracy. This article is part of the theme issue ‘Machine learning for weather and climate modelling’.

One-dimensional radiative transfer solvers are computationally much more efficient than full three-dimensional radiative transfer solvers but do not account for the horizontal propagation of radiation and thus produce unrealistic surface irradiance fields in models that resolve clouds. Here, we study the impact of using a 3-D radiative transfer solver on the direct and diffuse solar irradiance beneath clouds and the subsequent effect on the surface fluxes. We couple a relatively fast 3-D radiative transfer approximation (TenStream solver) to the Dutch Atmosphere Large-Eddy Simulation (DALES) model and perform simulations of a convective boundary layer over grassland with either 1-D or 3-D radiative transfer. Based on a single case study, simulations with 3-D radiative transfer develop larger and thicker clouds, which we attribute mainly to the displaced clouds shadows. With increasing cloud thickness, the surface fluxes decrease in cloud shadows with both radiation schemes but increase beneath clouds with 3-D radiative transfer. We find that with 3-D radiative transfer, the horizontal length scales dominating the spatial variability of the surface fluxes are over twice as large as with 1-D radiative transfer. The liquid water path and vertical wind velocity in the boundary layer are also dominated by larger length scales, suggesting that 3-D radiative transfer may lead to larger convective thermals. Our case study demonstrates that 3-D radiative effects can significantly impact dynamic heterogeneities induced by cloud shading. This may change our view on the coupling between boundary-layer clouds and the surface and should be further tested for generalizability in future studies. Plain Language Summary Solar radiation warms the surface and provides energy for evaporation and biological processes, resulting in the release of heat and moisture to the atmosphere. This upward transport of warm and moist air eventually leads to the formation of clouds, which then alter the spatial distribution of solar radiation at the surface by partly reflecting and absorbing the incoming sunlight. Most previous studies that simulated these complex interactions between clouds, solar radiation, and the surface used 1-D radiation models. These are faster than 3-D radiation models but produce unrealistic surface solar radiation fields by only considering the vertical propagation of radiation. In this study, we use a relatively fast 3-D radiation model to simulate the formation of clouds and the surface heat and moisture fluxes. In our simulations, 3-D radiation results in thicker and wider clouds than 1-D radiation, predominantly because clouds no longer shade the surface beneath them when radiation propagates under an angle. Unlike in simulations with 1-D radiation, we find higher surface fluxes below clouds than under clear-sky and higher surface fluxes with increasing cloud thickness in simulations with 3-D radiation. Our results show that 3-D radiation may strongly impact the coupling between clouds and the land surface.

Solar radiation enters the atmosphere following the direct beam of the sun, and can be absorbed or scattered in any direction by gas molecules, aerosols, cloud droplets, and the surface. This 3D nature of radiation affects the spatial structure of atmospheric radiative heating and global horizontal irradiance (GHI). Clouds intercept more radiation at their sides as the solar zenith angle increases, which enhances the size of cloud shadows (side illumination; Hogan & Shonk, 2013), but may also result in more diffuse irradiance (side escape; Hogan & Shonk, 2013). Additionally, clouds can intercept radiation reflected by the surface or by neighboring clouds (entrapment;Hogan et al., 2019). These solar radiation-cloud interactions create complex surface patterns with cloud shadows and regions where GHI exceeds clear-sky radiation.Ideally, one would capture these 3D effects in the radiation computations of cloud-resolving simulations. However, the actual status quo in weather and climate models is the use of relatively efficient two-stream methods that solve radiation in the vertical direction only (Cahalan et al., 2005;Meador & Weaver, 1980). These approaches are computationally affordable, but lack the aforementioned 3D radiative effects.The development of the TenStream solver (Jakub & Mayer, 2015), which reduces the photon propagation to a limited set of directions, allowed for detailed studies into 3D radiative effects in a coupled cloud-resolving model. These studies Veerman et al., 2020) revealed clear impacts of 3D radiative effects on cloud development, mainly driven by GHI patterns that strongly deviate from those in simulations with two-stream solvers. However, while emphasizing the importance of surface-radiation feedbacks relative to in-cloud processes, the GHI patterns produced by the TenStream solver still deviate from those produced by ray tracing (Jakub & Mayer, 2015, their Figure 8).Monte Carlo ray tracing is widely regarded as the most accurate technique to solve radiative transfer in 3D and can produce patterns in GHI that closely resemble field observations (Gristey et al., 2020). Furthermore, ray tracing is often used as reference for the development of 3D radiative transfer approximations (Hogan et al., 2016;

Vegetation and atmosphere processes are coupled through a myriad of interactions linking plant transpiration, carbon dioxide assimilation, turbulent transport of moisture, heat and atmospheric constituents, aerosol formation, moist convection, and precipitation. Advances in our understanding are hampered by discipline barriers and challenges in understanding the role of small spatiotemporal scales. In this perspective, we propose to study the atmosphere-ecosystem interaction as a continuum by integrating leaf to regional scales (multiscale) and integrating biochemical and physical processes (multiprocesses). The challenges ahead are (1) How do clouds and canopies affect the transferring and in-canopy penetration of radiation, thereby impacting photosynthesis and biogenic chemical transformations? (2) How is the This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.