Towards an analysis tool for ceiling and visibility

Accurate estimates of ceiling (CEIL) and visibility (VIS) are of great importance for safety and effectiveness in aviation. Low ceiling and/or poor visibility conditions are common causes or contributing factors in weather-related aviation accidents, and they can severely reduce the capacity of an airport. A more effective use of all relevant available meteorological data could have a positive impact on decision-making processes in these situations.

The aim of this work is to develop a tool generating quality-controlled gridded analyses of ceiling and visibility based on the integration of various kinds of observational data. The system is developed for a region around Austria (Central Europe) covering the Eastern Alps and its surroundings, but it is transferable to other parts of the world. Automatic measurements of cloud base height and visibility are currently in the process of replacing human observations at many standard meteorological stations in Austria and constitute the basic data for the analysis. Additional data is used in both major parts of the analysis tool: in the quality control as well as in the analysis itself for downscaling. Preliminary results of the ongoing development and testing based on a one-year data set will be shown.

At the Department of Meteorology and Geophysics at the University of Vienna, the Vienna Enhanced Resolution Analysis (VERA) system has successfully been developed and operated for standard meteorological parameters during the last 20 years. Even if not all elements of VERA are appropriate for CEIL/VIS, it provides a sound framework for the CEIL/VIS analysis. VERA is based on the separation of the analysis field into an “explained part” reconstructed by predefined known patterns called “Fingerprints” and a residual “unexplained part” subjected to a smoothness and a flatness condition. Current operational and test-phase Fingerprints include e.g. idealized typical orographically induced pressure or temperature perturbations, remote sensing data (radar) or simple empirical-statistical relationships to topography. The Fingerprint patterns provide subgrid-scale information (downscaling) and are only imposed on the analysis field if they are recognized in the observed values. Prior to the analysis, an automated quality control of input data based on data self-consistency (VERA-QC) is performed, including gross error recognition and bias correction.

For CEIL and VIS, spatial consistency checks like in the core VERA-QC are often not well applicable, since the scale of typical phenomena can be small compared to the average station distance. Instead, the CEIL/VIS quality control focuses on internal consistency checks between measurements at the same location. A set of simple relations between CEIL/VIS and additional meteorological parameters like, for example satellite data, lifting condensation level, relative humidity, radiosonde data, etc. is established. In combination, this set provides an estimate of the plausibility of a measurement upon which the decision is made for it to be discarded, corrected, flagged suspicious or kept as it is. Where both are available automated and manual observations are compared as well.

Quality-controlled CEIL and VIS measurements then enter the actual analysis which makes use of the VERA Fingerprint technique. A major class of Fingerprints consists of satellite derived products, e.g. Cloud Type or Cloud Top height by NWC SAF (Nowcasting Satellite Application Facilities) Meteosat Second Generation. Another promising Fingerprint approach is to quantitatively identify topographic features that favor low-level inversions (or an “inversion potential”) since those are often associated with low stratus. For the application in complex terrain it is essential to use high resolution topography data to define that kind of Fingerprint or other typical orography-related patterns like Foehn or barrage clouds. More Fingerprints are constructed using standard meteorological surface parameters, especially humidity. As a separate step in the analysis procedure, the NWC SAF satellite Cloud Mask in combination with ground observations is used to “clear” cloud-free regions.