Original paper

Capabilities and uncertainties of aircraft measurements for the validation of satellite precipitation products – a virtual case study

Lammert, Andrea; Ament, Felix

Meteorologische Zeitschrift Vol. 24 No. 5 (2015), p. 495 - 502

16 references

published: Aug 31, 2015
published online: Jul 16, 2015
manuscript accepted: Apr 23, 2015
manuscript revision received: Apr 22, 2015
manuscript revision requested: Mar 11, 2015
manuscript received: Dec 19, 2014

DOI: 10.1127/metz/2015/0663

BibTeX file


Open Access (paper can be downloaded for free)

Download paper for free


Remote sensing sensors on board of research aircraft provide detailed measurements of clouds and precipitation which can be used as reference data to validate satellite products. Such satellite derived precipitation data using passive microwave radiometers with a resolution of typically 50 × 50km2 stands against high spatial and temporal resolved airborne measurements, but only along a chosen line. This paper focuses on analysis on the uncertainty arising from the different spatial resolution and coverage. Therefore we use a perfect model approach, with a high resolved forecast model yielding perfect virtual aircraft and satellite observations. The mean precipitation and standard deviation per satellite box were estimated with a Gaussian approach. The comparison of the mean values shows a high correlation of 0.92, but a very wide spread. As criterion to define good agreement between satellite mean and reference, we choose a deviation of one standard deviation of the virtual aircraft as threshold. Considering flight tracks in the range of 50 km (one overflight), the perfect agreement of satellite and aircraft observations is only detected in 65 % of the cases. To increase this low reliability the precipitation distributions of the virtual aircraft were fitted by a gamma density function. Using the same quality criterion, the usage of gamma density fit yields an improvement of the Aircraft reliability up to 80 %.