ДО ПИТАННЯ УТОЧНЕННЯ КРИТЕРІЇВ ДОСТОВІРНОСТІ, ЯКІ ВИКОРИСТОВУЮТЬ ПРИ АВТОМАТИЗОВАНОМУ ПРОСТОРОВО-ЧАСОВОМУ КОНТРОЛІ ДАНИХ МЕТЕОРОЛОГІЧНИХ СПОСТЕРЕЖЕНЬ

Abstract

In order to ensure a reliability of data on a state of surface layer of atmosphere, data of meteorological observations conducted on the network of the Hydrometeorological Service of Ukraine are subject to regular multi-level technical and critical control. The entire set of errors in data of meteorological measurements, according to a nature of their manifestation, are divided into two groups: 1) random errors associated with instantaneous pulsations of measured meteorological valuables and their distortion during their further collection and processing; 2) systematic errors, which depend on defect in a devices design, their technical condition, as well as errors caused by violation of measurement technique and an influence of local features on measured meteorological valuables. In order to separate measurement errors from the manifestation of natural features of weather processes, all observational data coming from meteorological stations are subject to regular critical control, during which their reliability, that is, suitability for further use, is established. There are two main types of critical control of measurement data – intra-station and spatio-temporal. The basis of intra-station control is an analysis of values of interrelated meteorological values measured on one station. This is done at hydrometeorological stations with a help of special computer programs. Spatial-temporal control consists in comparing averaged over a certain period of time (decade, month) values of spatial fields of meteorological variables around a station, the data of which is checked/controlled. Local distortions of the field of meteorological values under an influence of local natural factors can be partially eliminated if we compare not values of meteorological values themselves, but their deviations from a climatic norm. Carrying out manual spatio-temporal control is a long and routine work, especially when it comes to analyzing a change in time of data observational inconsistencies. This made it necessary to conduct a study on improving the method of automated spatio-temporal data control of meteorological observations, and to present obtained results in a form of a normative document to be implemented in the activitiy of hydrometeorological organizations of the State Emergency Service of Ukraine. Such research was carried out at the Ukrainian Hydrometeorological Institute for 2020-2022. The purpose of the publication is to present scientific and methodological foundations of development and main provisions of the normative document “Clarification of reliability criteria used during automated spatio-temporal control of data from meteorological observations of stations. Methodical recommendations”, as well as the computer program “MeteoControl”, which forms a database of non-connections and calculates statistical characteristics, are needed for analyzing the array of non-connections and making a decision on the feasibility of changing the reliability criteria. The developed regulatory document provides a number of recommendations for specifying reliability criteria of meteorological observation data, in particular: 1) specification of a reliability criteria for all meteorological variables that are determined on a network of stations should be carried out at least once every 5 years; 2) before clarifying reliability criteria, it is necessary to assess an extent to which a current reliability criteria reveal a data, as well as to establish how many stations and how often in recent years have discrepancies that exceed a current reliability criterion; 3) before making a decision to change a reliability criterion for all stations, it is advisable to find out how, in connection with an introduction of new reliability criterion, a number of stations that will require additional analysis will change, or vice versa – whose data will be considered reliable despite to their dubious or erroneous values.