N
SUPPLEMENT
To the HONGKONG GOVERNMENT GAZETTE of 14th August, 1886.
GOVERNMENT NOTIFICATION.-No. 314.
The following report from the Director of the Observatory, on One Year's Observations of Thermometers exposed in Stevenson's screen, is published for general information.
By Command,
FREDERICK STEWART,
Acting Colonial Secretary.
:
Colonial Secretary's Office, Hongkong, 14th August, 1886.
ON ONE YEAR'S OBSERVATIONS OF THERMOMETERS EXPOSED IN
STEVENSON'S SCREEN.
It is known from papers read before the Royal Meteorological Society and from other sources, that minor alterations in the dimensions or form of Stevenson's screen have no important influence on the readings of the thermometers, although even in England the observations made in screens can scarcely
much smaller there · De taken as perfectly accurate. But at any rate the mean corrections must be
very
than in the torrid zone, and may perhaps amount to next to nothing where the screen is placed in a well exposed situation. But as the thermographic thermometers are exposed in screens fixed on the walls of the Observatories, the thermograms must at times require larger corrections, particularly in places, where the thermograph is not fixed in a position so favourable as e.g. at Kew. These circumstances · could be investigated by forming thermometric windroses, for whenever the screen is sheltered by the building the correction must increase, even when the thermograph is placed in a small detached hut.
A Stevenson's screen was erected here at the end of 1883, as explained in Observations and Resear- ches made in 1884, in a well exposed situation on the Observatory hill. It is sheltered by buildings or trees at a distance between NE and NW (from which quarter the wind seldom blows) and the view is unobstructed in any other quarter. The readings have been made at the same time as the accurate air temperatures were determined by aid of the rotating thermometer and monthly mean corrections have been obtained by comparison with the data published in the Monthly Weather Reports. These correc- tions may be applied to the observations made in wooden screens lately erected at the lighthouses, etc. in this neighbourhood.
The maximum temperature registered in Stevenson's screen is too high, especially during the months when the sky is generally clear. During the months of November and December the correction was further increased here owing to the screen being slightly sheltered from northerly winds. The correction to the temperature read off at 10 a. is larger than at 4 p. owing to the greater radiation which heats the screen, the highest black bulb temperature generally occurring at or shortly before noon. The minimum temperature is too low owing to the great radiating power of the white lead with which it is painted, which also is the cause of the heavy dew so frequently found on the screen. The tempera- ture at 10 p. is too low for the same reason.
The error of the temperatures registered during the day is however much greater than during the night, owing to the convection currents of heated air from the ground. Puffs or lulls in the wind cause the air particles, which are longer or shorter times in contact with heated objects (the ground, The vibrations seen on the the louvres of the screeen, etc.), to assume different temperatures.
thermograms during the day time are thus produced, though they depend also to some extent upon clouds, that diminish the radiation while passing over. There is always some uncertainty attached to the extreme temperatures, as the air is not quite uniformly hot, more especially is the maximum slightly uncertain. On a hot day objects are seen at any rate in a telescope somewhat unsteady through strata of air of different temperatures.
The thermometers used were those recommended by the Meteorological Office. Their bulbs are too small and thermometers with long cylindrical bulbs are more sensitive.
The absolute and relative humidity were obtained by aid of Blanford's tables, which are perhaps not strictly applicable as he fanned his thermometers, and it is known that anything that causes the layer of air attached to the damp bulb to be renewed, lowers the damp bulb temperature. In these observations the bulb was wetted by aid of a bottle of water permanently fixed beside the thermometer. If the bulb is wetted before every observation, it falls lower."
: