Don't kill your allies: The impact of chemical and biological locust and grasshopper control on birds

Before the large-scale introduction of chemical insecticides after the Second World War (WW2), birds were considered worldwide as welcome natural allies to control acridids. For their role in acridid control they were worshipped in ancient times and in the first half of the 20th century protected in many countries worldwide. Until WW2, mechanical control e.g. by harvesting grasshoppers with ingenious dozers or digging trenches to interrupt marching hopper bands were chemically supplemented by massive use of arsenicals. The large scale introduction of Dieldrin in locust control around 1955 marked a turning point, as for the first time a persistent chemical became available to treat hopper bands which provided toxic residues killing hoppers emerging afterwards. The high toxicity for man and wildlife and the persistence of Dieldrin lead to a worldwide ban twenty years later, but in locust control it took until 1989 before its use eventually was discontinued. Meanwhile, organophosphorous insecticides had dominated the acridid arena since WW2 with varying contributions of carbamates, pyrethroids and some other chemical families. The interest in birds as natural allies had faded because of a widely held believe that the insecticides did a better job. In addition, the impact on birds of their widespread and large-scale use in Africa had neither been studied, nor assessed for its potential consequences for acridid control.

It was against this background that the studies described in this thesis were undertaken to compare the efficacy on acridids and the impact on birds of non-selective chemical and selective biological insecticides, to contribute to efficient locust and grasshopper management with the lowest environmental footprint. Two oganophosphorous insecticides, fenitrothion and chlorpyrifos, both commonly used in acridid control until the present day and the entomopathogenic fungus Metarhizium acridum were studied. Both organophoshates caused direct avian mortality (2-7%) and morbidity through anti-cholinesterase poisoning. They also reduced the availability of arthropod prey leading to reductions in bird densities, reduced body mass, nest desertion and debilitation of fledglings. Acridid numbers returned to pre-spray densities by 21 days post-spray but bird numbers remained 25-50% lower than controls. By using a novel approach, mimicking bird feeding behaviour by using fine tweezers to pick up affected locust nymphs, it was shown that nymphs accumulated residues by secondary uptake generally peaking from 3 h to 24 h post-spray. This would have been missed by the current approach of using pitfall traps, sweep nets or tethered insects which are unlikely to represent realistic residue levels. Therefore current regulatory frameworks likely underestimate risks of acridid control, and do neither take into account gorge feeding and depletion of insect prey by the use of non-selective chemical insecticides.  

M. acridum had a comparable but delayed impact on acridids which, however, lasted longer than that of the chemical insecticides studied. Where the chemical insecticides reduced acridid numbers within hours to days, the biological product took 7 to 12 days to reduce acridid numbers to the same extent. However, it did not impact bird densities but rather their numbers increased and extended the impact of M.acridum on acridids with several months. Birds preferred medium and large sized grasshoppers and female Desert Locusts. The latter were taken three times more than males by kestrels which were studied. No direct or indirect adverse side-effects were observed on non-target organisms including invertebrate and vertebrate locust predators such as ants, reptiles and birds. These substantial ecological advantages should also be considered when choosing between conventional chemical and biopesticide-based locust control.

More than 30 years since the development of M. acridum as a selective fungal insecticide for grasshoppers and locusts had started, potential users in Africa still prefer chemical insecticides as is shown by the ongoing outbreaks in the Greater Horn of Africa and beyond. This is partially due to misconceptions about its real costs by not taking into account the externalities associated with chemical insecticides and more importantly by a lack of scouting and early intervention in adjacent countries, during which locust swarms can build up, and a lack of planning to include the lag phase which is inherent to any bioinsecticide.

This thesis explains how depending on the different stages of acridid development, birds can play an important role in acridid control and under several conditions even could prevent grasshoppers from reaching Economic Thresholds or Desert Locusts from forming swarms, by combining natural predation with the use of M. acridum. The use of chemical insecticides can be reduced much further to narrow the window of application than currently is the case. The research presented in this thesis also outlines why fenitrothion and chlorpyrifos should be considered for removal from FAO's Pesticide Referee Group assessments because of their negative effects on natural predators of acridids.

In all scenarios there is a strong need for habitat restoration and grazing management as these factors would reduce opportunities for acridid development and increase bird numbers which have severely decreased over the past 50 years. The increasing place of biological locust and grasshopper control in China, both by using biopesticides and by attracting acridivorous birds by improving breeding conditions, is a good example of how to change the chemical oriented views prevailing up to the present in Desert Locust and Senegalese Grasshopper control. If this more comprehensive approach that will include the ecosystem services provided by avian allies will not be prepared in time and the necessary measures not executed, the next major upsurge will again trigger panic reactions resulting in massive use of chemical insecticides and potentially famine.