Using ARM data to evaluate the dependence of surface downward longwave radiation on near-surface temperature and water vapour path, in both ARM observations and the Met Office NWP model.
| Henderson, Peter | Environmental Systems Science Centre |
| Slingo, Anthony | Environmental Systems Science Centre |
In this work, we continue our comparisons between ARM data and simulations from the UK Met Office Numerical Weather Prediction (NWP) model. The present analysis focuses on the variables that control the downwelling longwave radiation at the surface, in particular the column water vapour and near-surface temperature. Water vapour is both a source and a sink of latent heat, is an active modulator of atmospheric radiative transfer and therefore influences both the general circulation and the global energy balance. It is the dominant greenhouse gas, with a strong feedback on future climate change, and must therefore be represented correctly in models. We consider the functional relationships between the down-welling broadband longwave radiative flux at the surface, the total column water vapour path and the near-surface temperature, across the diverse range of conditions encountered at all the central facilities of the ARM program. The internal consistency of the model is evaluated by comparing with the observed relationships and implications for the modelling of surface radiation fluxes are demonstrated. We also evaluate simple parametrizations that have been developed to estimate the downwelling flux using readily available surface measurements as input, beginning with Angstrom (1918) and more recently Prata (1996). The relative decoupling of the top of atmosphere and surface longwave fluxes makes inference of the surface values from satellite observations difficult at best, so there remains a very real need to exploit such surface based approaches. ARM retrievals of the column water vapour from passive microwave radiometers are used to compare the Prata parameterisation against more physically based calculations using the Edwards-Slingo radiation code, forced both with the observed fields and with the predictions from the NWP model. The work is currently being extended to consider the additional influence of cloud on the fluxes, as identified by a broadband cloud-mask similar to that developed by Long and Ackerman.
