Land–Atmosphere Coupling Indicators in Subseasonal Atmospheric Models

The influence of the land surface wetness state on precipitation is known to vary across the world. In areas where there exists such a strong coupling between land and atmosphere, soil moisture information can be used to increase the predictability of precipitation on several timescale, varying from the daily to the seasonal timescale.

Whether there is a strong coupling between land surface and precipitation depends on how the surface energy balance affects turbulent atmospheric boundary layer processes and cloud formation. If the atmosphere is too dry, no precipitation will occur regardless of land surface conditions. If the atmosphere is too moist, precipitation will always occur regardless of land surface conditions. Moreover, there should be some conditional instability the to allow convective precipitation to occur.

Several indicators (CTI-HIlow (Findell et al. (2003)), dΘ/dEF (de Ridder, 1998)) exist to assess the atmospheric conditions with the aim of determining the coupling strength. Previous studies have analysed these indicators based on idealised one-dimensional atmospheric models, as well as in climate models.

Here, we will study them based on (sub-)seasonal atmospheric simulations (S2S-project archive) to determine the relevance of these indicators up to the seasonal timescale. We specifically analyse the precipitation difference between ensemble members and regress these differences against the different land surface initialisations of the ensemble members. This regression is used to determine how sensitive precipitation is to land surface conditions for each model globally. Finally, we relate this model sensitivity to the coupling strength based on the land-atmosphere coupling indicators.

Initial results show that coupling indicators have the potential to diagnose coupling, but may have to be calibrated for each model in the S2S-database. A global analysis of land surface wetness-precipitation coupling in the S2S-database, as well as the robustness of several land-atmosphere coupling indicators, will be presented.