Kneifel, Stefan (2011). Characterization of snowfall using ground-based passive and active remote sensors. PhD thesis, Universität zu Köln.

diss_kneifel_2011.pdf - Updated Version

Download (6MB)


Snowfall is a key quantity in the global hydrological cycle and has an impact on the global energy budget as well. In sub-polar and polar latitudes, snowfall is the predominant type of precipitation and rainfall is often initiated via the ice phase. Currently, the spatial distribution of snowfall is poorly captured by numerical weather prediction and climate models. In order to evaluate the models and to improve our understanding of snowfall microphysics, global observations of snowfall are needed. This can only be obtained by space-borne active and passive remote sensors. In order to be able to penetrate even thick snow clouds, sensors operating in the microwave frequency region are favoured. The challenge for snowfall retrieval development lies first in the complexity of snowfall microphysics and its interactions with liquid cloud water. Secondly, comprehensive knowledge is needed about the interaction of electromagnetic radiation with snowfall in order to finally relate the radiative signatures to physical quantities. A general advantage of ground-based observations is that simultaneous measurements of in-situ and remote sensing instruments can be obtained. Such a six-month dataset was collected within this thesis at an alpine site. The instrumentation included passive microwave radiometers that covered the frequency range from 22 up to \unit[150]{GHz} as well as two radar systems operating at 24.1 and 35.5 GHz. These data were complemented by optical disdrometer, ceilometer and various standard meteorological measurements. State-of-the-art single scattering databases for pristine ice crystals and complex snow aggregates were used within this thesis to investigate the sensitivity of ground--based passive and active remote sensors to various snowfall parameters such as vertical snow and liquid water distribution, snow particle habit, snow size distribution and ground surface properties. The comparison of simulations with measurements within a distinct case study revealed that snow particle scattering can be measured with ground--based passive microwave sensors at frequencies higher than 90 GHz. Sensitivity experiments further revealed that ground-based sensors have clear advantages over nadir measuring instruments due to a stronger snow scattering signal and lower sensitivity to variable ground surface emissivity. However, passive sensors were also found to be highly sensitive to liquid cloud water that was frequently observed during the entire campaign. The simulations indicate that the uncertainties of sizes distribution and snow particle habit are not distinguishable with a passive-only approach. In addition to passive microwave observations, data from a low-end radar system that is commonly used for rainfall were investigated for its capabilities to observe snowfall. For this, a snowfall specific data processing algorithm was developed and the re-processed data were compared to collocated measurements of a high-end cloud radar. If the focus can be narrowed down to medium and strong snowfall within the lowest 2-3 km height, the reflectivity and fall velocity measurements of the low-end system agree well with the cloud radar. The cloud radar dataset was used to estimate the uncertainty of retrieved snowfall rate and snow accumulation of the low-end system. Besides the intrinsic uncertainties of single-frequency radar retrievals the estimates of total snow accumulation by the low-end system lay within 7% compared to the cloud radar estimates. In a more general approach, the potential of multi-frequency radar systems for derivation of snow size distribution parameters and particle habit were investigated within a theoretical simulation study. Various single-scattering databases were combined to test the validity of dual-frequency approaches when applied to non-spheroid particle habits. It was found that the dual-frequency technique is dependent on particle habit. It could be shown that a rough distinction of snow particle habits can be achieved by a combination of three frequencies. The method was additionally tested with respect to signal attenuation and maximum particle size. The results obtained by observations and simulations within this thesis strongly suggest the further development of simultaneous ground-based in-situ and remote sensing observations of snowfall. Extending the sensitivity studies of this study will help to define the most suitable set of sensors for future studies. A combination of these measurements with a further development of single-scattering databases will potentially help to improve our understanding of snowfall microphysics.

Item Type: Thesis (PhD thesis)
Translated title:
Charkterisierung von Schneefall mit Hilfe von bodengebundenen passiven und aktiven Fernerkundungssensoren.German
CreatorsEmailORCIDORCID Put Code
Kneifel, Stefanskneifel@meteo.uni-koeln.deUNSPECIFIEDUNSPECIFIED
URN: urn:nbn:de:hbz:38-45665
Date: October 2011
Language: English
Faculty: Faculty of Mathematics and Natural Sciences
Divisions: Faculty of Mathematics and Natural Sciences > Department of Geosciences > Institute for Geophysics and Meteorology
Subjects: Natural sciences and mathematics
Earth sciences
Uncontrolled Keywords:
remote sensing of snowfall, ice particle scattering, radar remote sensingEnglish
Date of oral exam: 21 November 2011
NameAcademic Title
Crewell, SusanneProf. Dr.
Simmer, ClemensProf. Dr.
Illingworth, AnthonyProf. Dr.
Projects: TOSCA
Refereed: Yes


Downloads per month over past year


Actions (login required)

View Item View Item