zum Inhalt springen

Methodological Research Areas

Microwave remote sensing

Microwave radiation of the atmosphere is a treasure for atmospheric scientists. Compared to other spectral regions, mircowave sensors can penetrate all clouds and still deliver valuable information about them! Remote sensing based on microwaves can be can be used quantify profiles of atmospheric temperature, water vapor, clouds and precipitation. Cloud radars transmit microwave radiation to obtain vertical profiles from cloud and precipitation particles, while microwave radiometers measure the emission from atmospheric constituents such as oxygen, water vapor and hydrometeors. Instruments can operate from surface, air-borne and satellite platforms. Our research group

  • has obtained unique expertise in the development, installation and operation of microwave radiometers on all three platforms.
  • is setting up a European center for microwave radiometer best practices.
  • exploits microwave radiometers for filling gaps in current measurements systems and for answering research questions related to clouds and climate.   

Radiative transfer simulations

Without simulations of what a satellite or a ground-based remote sensing instrument would “see” for a given atmospheric situation, we would not be able to develop methods to derive atmospheric parameters such as temperature, water vapor content, cloud water amount. Such simulations are called “radiative transfer” and require a physical model of how atmospheric radiation interacts with the atmosphere (i.e. gases, clouds, precipitation, etc..) and the Earth surface. Our research group

  • investigates absorption by atmospheric gases and the scattering of cloud and precipitation particles in the microwave and submillimeter spectral regions.
  • applies radiative transfer models in different spectral regions, i.e. microwave, infrared.
  • couples radiative transfer models as a so called forward operator to climate and weather prediction models.

Atmospheric retrieval algorithm development

Remote sensing instruments, be they in space or on the surface of the Earth, do not deliver the desired atmospheric parameters (such as temperature, water vapor content, cloud liquid amount, etc..) directly. The measurements of the instruments (mostly in form of some kind of radiation measure) need to be converted to the desired meteorological variable, i.e. retrieved. The retrieval process is usually not straight-forward because often many atmospheric states can represent one and the same measurement - an under-determined problem. Solving this problem requires a mathematical algorithm, that should be consistent with certain physical constraints and/or take into account some prior knowledge (i.e. a climatology of the variable at stake). The research group

  • has developed a broad range of atmospheric retrieval algorithms (multi-variate regression, neural networks, variational approaches) to derive atmospheric quantities.
  • has pioneered synergetic retrieval techniques which combine of different sensors making use of their complementarity.
  • distributes these retrieval algorithms to interested scientists and users.


Reanalyses represent the optimal weather reconstruction of the past. They provide time series of the atmospheric state, e.g. temperature, water vapor, wind, on a three dimensional grid and are therefore frequently used to investigate climate change processes. They are built by an optimal combination of observations and an atmospheric model taking into account the respective uncertainties. Global reanalysis have resolutions of about 80 km and most go back unl the start of the satellite area. The research group

  • has been involved in the development of high resolution regional reanalyses (COSMO-REA6 with 6 km and COSMO-REA2 with 2 km resolution) in the framework of the Hans-Ertel Centre for Weather Research
  • uses reanalyses to assess the availability of renewable energy sources.
  • investigates the quality of various reanalysis in the Arctic and the Atacam Desert (where only a sparse number of observations are included) for answering climate-related questions