A Concept for Sensor based error detection and quality assurance for a network of micrometeorological stations

Within the frame of a project aimed to determine the averaged fluxes of momentum, sensible heat and water vapour over an area corresponding to a grid cell of a numerical weather prediction model, the German Weather Service (DWD) operates a network of up to seven micrometeorological stations (so-called energy budget stations) in a 20km*20km area around the Meteorological Observatory Lindenberg. Measured parameters include profiles of wind speed, temperature and humidity, all components of the radiation budget (short-/longwave, PhAR, each up-/downward), radiative surface temperature, profiles of soil temperature and soil moisture, soil heat flux and turbulence parameters.

Some specific features and operation characteristics of this network are: - a high number of sensors, and physical and logical data channels, - the integration of direct flux measurement systems using both analog and serial ports, - the use of two independent radio transmission lines for data transfer and system control. The stations are operated over different types of the underlying surface during different seasons implying frequent re-installation. The high number of data channels and regular changes in the operating characteristics of the different stations create the risk of occurrence of different errors in system installation, sensor operation, signal conditioning, and data processing.

Considering this risk and the limited ressources of personnel and time to survey the system performance, a strategy is currently developed and implemented for a sensor based automatic error detection and quality control, making use of specific characteristics of the system and of the different sensors employed.

The presentation will focus on a brief description of the measurement system and on a discussion of the strategy for an automatic system of quality assurance and quality control.