Atmospheric deep convection
Atmospheric deep convection means thermally direct circulations that result from the action of gravity upon an unstable vertical distribution of mass. Deep convection is the cause of practically all precipitation in the tropics and most of the warm season precipitation in midlatitudes. Despite significant research, our basic understanding of what controls deep convection is still incomplete. This means that representing deep convection in numerical models (from weather prediction models to climate models) is difficult and such models often fail to accurately simulate the occurrence and characteristics of deep convection. Our group is currently analysing observations of deep convection in the tropics to better understand the mechanisms that control deep convection.
Atmospheric aerosols, ice crystals and microphysical processes
Solar radiation drives the atmospheric circulation and thereby weather and climate. Atmospheric aerosols and ice clouds play an important role in Earth's climate system by affecting the radiative transfer . Since even small changes in planetary radiation balance can cause changes in the climate, it is important to know how singe- and multiple-scattering properties of aerosols and ice crystals influence solar radiation. In our group shortwave radiation in the presence of ice crystals and aerosols is modelled using both column and 3-D radiative transfer models. In particular, the sensitivity of the radiation to uncertainties in the concentrations and shapes of ice crystals is being quantified with help of airborne microphysical data.
Atmospheric aerosols also act as cloud condensation nucleai and thus affect cloud properties. Aerosol-cloud interactions are considered as the main factor of uncertainty in the current understanding of the forcers of the climate change. By altering cloud microphysics aerosols can change the radiative balance as well as the hydrological cycle. We have recently been interested especially on the effect of aerosols on upper tropospheric humidity via changes in the microphysics of deep convective clouds.
Dynamics of mid-latitude weather systems
Extra-tropical cyclones are the dominant cause of weather in the mid-latitudes and can lead to heavy precipitation and damaging winds. Therefore it is vital that the dynamics of these systems are well understood so that accurate forecasts can be made. Furthermore it is important that we understand how extra-tropical cyclones are likely to change as our climate warms. Relevant questions that our group are addressing include: How will the amount and intensity of precipitation associated with extra-tropical cyclones change in the future? How will the number, intensity and location of extra-tropical cyclones change in the future? What is the role of changing sea-surface temperatures and sea-ice concentration on the characteristics of extra-tropical cyclones? We are addressing these questions using reanalysis data and a hierarchy of numerical modelling experiments. Idealised experiments are conducted with the Weather Research and Forecasting (WRF) model and more realistic experiments are conducted with OpenIFS. Diagnostic tools such as the omega equation and the height tendency equation are used for interpreting the model results.
Climate Change Projections
Changes in the atmospheric composition, particularly increases in carbon dioxide and other greenhouse gases, are altering the global climate. For helping the adaptation to the forthcoming changes, realistic scenarios of future climate together with information on their uncertainty are needed. Therefore, we are interested in both the inter-comparison of climate changes between different models and on the bias correction problem (how to best combine the information available from models and observations?). The effect of climate change on snow and snowfall climate in northern Europe is also a topic of both scientific and practical interest.
Atmospheric models are our main tools to simulate and predict the complex flow dynamics of the atmosphere. These models are derived from the continuous thermo-dynamical equations by means of discretisation which introduces unavoidable uncertainties - a model error - to these tools. We want to develop and apply methods and approaches to expose model error, and develop model error representations to data assimilation methods and simulation models. We have successfully applied approaches such as
- algorithmic methods to optimize model parameters and estimate model parameter error covariances for stochastic model error representations
- structures preserving data compression methods to enable extremely high-dimensional time-series analysis and evaluation of deficiencies in climate models' representation of low-frequency variability
- diagnosis of data assimilation system performance regarding information content of the so-called residual sequences
- impact evaluation of stochastic parametrizations for the simulations accuracy of the climate mean and its variability