Some observations on flying locusts and atmospheric turbulence in eastern Africa

Flying swarms have been observed to vary in vertical extent from a few metres to several thousand metres, with the spacing of the locusts in them ranging from more than 10 locusts/m 3 down to the order of 0·001/m 3 . Some characteristic effects seen, from ground and air, are described in qualitative terms. Two well‐documented records of flying locusts in numbers at some 2,000 m above ground were both associated with lapse rates close to the dry adiabatic, from the ground up to at least the level of these locusts. On two other occasions detailed observations of vertical temperature distribution have been made in the immediate vicinity of large swarms flying up to 1,100–1,700 m above ground, and in both these cases the topmost locusts were within 150 m of the upper limit of superadiabatic or adiabatic lapse rates from the surface. Comparable observations on a swarm in which all flying locusts were below 6 m demonstrated isothermal conditions, with the air temperature at 50 m within 1/2°F of that at the surface. All adequately‐documented records of swarm displacement so far available have been directly down‐wind, at ground speeds which for the larger swarms studied have approximated to the speed of the corresponding wind. The smaller swarms, most of them less than 10 km 2 in extent, have shown lower flying heights, relative to the corresponding vertical extent of dry adiabatic lapse rates, together with ground speeds which were also low, in relation to the wind speeds concerned and to the corresponding performance of the larger swarms. There is some evidence of an association between rain and low flying heights; and consideration of heat exchange at the ground surface indicates that the energy normally available for convection, in the arid regions frequented by swarms, is likely to be drastically reduced after rain. Quantitative evidence on the strength and distribution of the vertical components of air movement likely to be encountered by flying locusts has been provided by pilot‐balloon and accelerometer data. Twin‐theodolite pilot‐balloon observations in central Somalia have shown that during much of the day some 10 per cent of uniformly‐distributed locusts in flight at about 20 m at any one time could be expected to experience up‐currents exceeding their gliding sinking‐speed. Gust‐spectra recorded by aircraft accelerometer within and around high‐flying swarms have demonstrated up‐gusts, exceeding a value equivalent to the sinking‐speed of a gliding locust, at about fifty points per km 2 among the higher‐flying locusts at any one time. The possible use of locusts as indicators of air movements is discussed, with a summary of the evidence available on the contributions made by the active behaviour of the locusts themselves to the effects observed. Attention is directed to the importance of gregarious behaviour in the continued cohesion of individual swarms, observed over many days and hundreds of kilometres, despite the disruptive effects both of atmospheric turbulence and of the apparently random orientation of the flying locusts themselves. The order of magnitude of these potentially disruptive effects is estimated for a particular case, and shown to be large compared with the variations actually observed in the extent of the swarm concerned.