OCR Text |
Show SECTION 2- AIR TEMPERATURE An accurate temperature of the air is an elusive quantity. There are no instruments suitable to routine field use which measure this directly. All instruments in current use- thermometer, thermographs, temperature recorders- measure the temperature of a substance in the instrument, such as a mercury column, bimetallic element or gas in a closed system. Whether the temperature of this substance is the same as that of the surrounding air depends on the many complex factors affecting the heat balance of the sensor. Whether this surrounding air is at the same temperature as the free air in the vicinity is further a function of microclimate and environment. The finest precision thermometer is no better for measuring air temperature than the manner in which it is exposed. The heat balance of a temperature sensor depends on convec-tive and conductive heat exchange with the air, evaporation or condensation of moisture, and radiation, both visible and infrared. Ideally, it should depend only on heat exchange with the air; then the only errors would be due to lag in following rapid air temperature changes. Evaporation or condensation can cause very large errors, but these are usually transient. Temperature sensors are usually shielded from direct precipitation, but this cannot be eliminated entirely if a free flow of air is also allowed. The most persistent source of error is radiation. This can never be entirely eliminated, but it can be minimized by proper shielding. The most accurate air temperature measurements are made with sensors exposed to strong forced ventilation ( aspiration), so that the heat exchange with the air dominates the other exchanges. Normal climatologica1 ( and snow- avalanche) air temperature sensors are non- aspirated and depend only on natural circulation from the wind to assure heat exchange with the air. Temperature sensors should never be exposed to direct or reflected solar radiation. Further, they should be enclosed in a shield which in itself is reasonably close to the prevailing air temperature; otherwise infra- red heat exchange between sensor and shield wall will introduce error. The best shields are double-walled, the inner wall shielding the sensor and the outer wall shielding the inner one. A free circulation of air must be assured. The installation must be free from local influences of artificial heat sources or strong perturbations in the microclimate ( chimney, asphalt paving, etc.). Large air temperature variations occur close to the ground. These variations can be especially intense over a snow surface. The temperature sensor must be located above the major part of these variations. The minimum height above the ground is four feet; this is the standard for climatological observations. In 12 |