Analysing effects of meteorological variables on weather codes by logistic regression

Abstract

The qualitative parameters describing present weather are particularly difficult to automate. The weather types which create most of these difficulties are known, but little attention has been given to investigating the reasons for disagreements between the primary reference, the professional observer and an automated instrument. This paper provides a method –multiple logistic regression –to compare the WMO present weather codes detected by a professional observer and an automated system. A new approach is introduced to explain the errors relative to the official weather variables. Many weather periods have been analysed, but here results are presented for a snow period with slight and very slight precipitation. The best predictive variables were the dew point temperature, wind direction, relative humidity and visibility. The catalyst for this study was the need for better quality control and for tools to enable the development and the manufacture of better instruments and systems to detect present weather. Copyright © 2002 Royal Meteorological Society

Present weather: comparing human observations and one type of automated sensor

“…A particularly interesting possibility is that of being able to recognise differences or increased probabilities of differences from other meteorological variables. For example, the largest variation in the type of precipitation occurs close to 0 o C. Another example is the wind shading of the sensor sampling volume; this should be apparent from the prevailing wind speed and direction Goodison et al, 1994;Merenti-Välimäki & Laininen, 2002 In Figure 1 it is possible to see that on this day all sensors are slightly biased towards liquid precipitation, with two of them (FD2 and FD3) more likely to report liquid precipitation than the others. An estimate of the importance of coincidence at such determination is provided by Figure 1.…”

“…Until now not very much has been published on the reasons for disagreement. The work described here, and that of Merenti-Välimäki & Laininen (2002), attempts to identify and quantify the reasons for differences for one such PWS, the Vaisala FD12P, compared with the human observations carried out at the Jokioinen Observatory of the Finnish Meteorological Institute (location 60°49′N, 23°30′E) during the winters of 1996-97 and 1997-98. Four automatic PWSs of type Vaisala FD12P were installed on the precipitation intercomparison field of the observatory (Aaltonen et al, 1993;Elomaa et al, 1992;Elomaa, 1993), some 200 metres from the Observatory. All sensors were mounted at a 2-metre sampling height with booms in a 135-degree direction and receivers oriented north.…”

Predicting Weather Uncertainty with Deep Convnets

“…We use temperature prediction as an initial target. Based on previous exploratory work [18], we select a subset of Initial Parameters (IP) that have an influence on temperature: zonal and meridional wind, geopotential, temperature, relative humidity, and the fraction of cloud cover.…”

Automated present weather observations: a new concept for analysing the effects of meteorological variables

“…The difference between the nominal codes given by a professional observer and an automated instrument is referred to as the determination error (Merenti-Välimäki & Laininen 2002). These determination errors are binary (dichotomous) variables taking the values 0 and 1.…”