|
<br />.
<br />
<br />r..__'_'--F.lf~'"
<br />
<br />..",--
<br />
<br />682
<br />
<br />B. SEVRUK AND L ZAHLAVOVA
<br />
<br />of the orifice rim. The observational variables include mainly the height of the precipitation gauge orifice
<br />above ground, and the environmental ones include both meteorological variables, such as wind speed, form,
<br />and intensity of precipitation, and precipitation gauge site characteristics, such as the degree of protection
<br />of the gauge site against wind. The latter is especially important because precipitation gauges are installed
<br />outdoors under different conditions, in as yet non-standardized degrees of GSE, in cities. airfields, gardens,
<br />parks, woods, mountains, lakeshores, etc. The surrounding terrain and surrounding objects can provide
<br />different degrees of protection from the wind. The wind speed at the orifice level of a protected gauge can
<br />be reduced to one-third of the value of a nearby situated, but exposed, gauge. The resulting wind-induced
<br />error can differ considerably. As shown in Figure I, it amounts to 2 per cent for a protected site and to 5
<br />per cent for an exposed site in the summer at an altitude of 600 m a.s.1. in Switzerland. The respective
<br />figures for the winter are 7 and 18 per cent. At higher altitudes the differences are considerably greater
<br />(Figure I). Nevertheless, the GSE was not recorded in the past and is still not measured by many national
<br />meteorological services at present.
<br />Recommendations do exist for the installation of precipitation gauges with respect to the surrounding
<br />objects, but as pointed out by Sevruk (1973), they differ considerably according to the country and the
<br />author. In addition, they are rarely followed and are insufficient. The recommended ratio for the height of
<br />the objects and their distance to the gauge can range from 1 :0,5 to 1: 10 (Sevruk, 1973). Moreover, despite
<br />the missing standards of GSE, a generally accepted classification of GSE is not used. Sometimes the
<br />application of two extreme classes of GSE, such as exposed and protected, is recorded in the literature or
<br />in station history records, but only subjective and qualitative information is provided on the GSE, with no
<br />definition and no objective scale, including interim GSE classes such as partly exposed or partly protected.
<br />Three exceptions are worth noting: Korhonen (1944) used four classes of GSE to estimate the wind-induced
<br />error. Brown and Peck (1962) developed a qualitative classification system of GSE containing seven classes,
<br />from overprotected to very windy. It was based on the subjective description of the situation of the gauge
<br />site in a mountain area in relation to winds from directions associated with precipitation. A relative good
<br />relationship was evident between the GSE class and the difference of the winter precipitation catch and
<br />the water equivalent of snow cover surrounding the gauge. Shver (1965) used a system of six qualitative
<br />classes of GSE for the territory of the former USSR to convert the precipitation amount measured in an
<br />older model of national standard gauge to a new one. The system contains a short description of the general
<br />situation of the gauge site in relation to the surrounding topography, terrain, vegetation, and other objects.
<br />In addition, Moss (1960) dealt with the general problem of the effect of obstructions on trajectories of
<br />raindrops by considering various conditions of airflow over a precipitation gauge.
<br />The GSE problem arises from the fact that the available wind speed values are measured routinely by
<br />national meteorological services at heights well above the tops of the surrounding obstacles, that is, at
<br />
<br />SUMMER
<br />
<br />WINTER
<br />
<br />
<br />60
<br />loss
<br />[%]
<br />40
<br />
<br />20
<br />
<br />-.-,,,
<br />;;..~
<br />
<br />
<br />
<br />o
<br />
<br />600
<br />
<br />2000
<br />
<br />600 2000
<br />(m a. s. I.)
<br />
<br />Figure I. Comparison of the mean seasonal wind-induced error for open (hatched) and protected gauge sites at altitudes of 600 m
<br />a.s,1. and 2000 m a.s.1. in Switzerland for the Hellmann precipitation gauge. The summer season is from April to September and the
<br />winter season from October to March
<br />
<br />I~ "-,~".,,,:~,!~?t:~"!I"~\~~~
<br />
|