R. Peveling & J. Weyrich
Universität des Saarlandes.

Abstract

Field trials on the impact of locust control agents on non-target arthropods were carried out in Tamesna, southern Sahara, Republic of Niger, in September/October 1990. The study was conducted within the framework of the GTZ-Project 'integrated Biological Control of Grasshoppers and Locusts'. The effects of neem oil (0.1 % and 0.2% azadirachtin), Beauveria bassiana (Balsamo) Vuillemin and dieldrin on the tenebrionid beetle Pimelia angulata tschadensis Koch were studied in standardized in situ cage trials. Additionally, neem oil and B. bassiana were tested in small scale field trials in typical Schistocerca gregaria (Forskål) (Orthoptera: Acrididae) habitats (Schouwia thebaica vegetation islands). Dieldrin and neem (0.1 %) caused high mortalities in the cage trials. Neem (0.2%) induced an increase, dieldrin a decrease in activity of the beetles. Neither in the cage trials nor within additional laboratory tests could B. bassiana infections be observed in P. angulata tschadensis. Only topically inoculated Trachyderma hispida (Forskål) (Tenebrionidae) were infected. The field applications did not appear to cause any side-effects on the tenebrionid fauna, i.e., neither alteration of activity pattern nor lethal effects. The final results of these investigations (effects on other non-target arthropods) will be presented in subsequent publications. As key components of arid ecosystems Tenebrionidae could be useful indicator organisms to study ecotoxicological risks of insecticides prior to large scale field applications.

For decades locust control has relied almost exclusively on the application of chemical insecticides. However the increasing knowledge about their potential hazardous side-effects (related to their non-specificity and persistence) has recently promoted the search for environmentally sound alternatives and adoption of the conceptual approach of integrated pest control, which incorporates biological and biotechnological methods, has received increasing attention from experts.

The application of alternative locust control agents can by no means be considered a priori to be environmentally sound. As with chemical pesticides, they must undergo a careful and thorough evaluation of potential side effects prior to large scale field application.

This ecotoxicological risk assessment involves both laboratory and field trials, the former to work out principal physico-chemical and toxicological (or pathogenic) properties of the respective control agent, the latter being indispensable to study its field behaviour, and thus preferably carried out at an early stage of risk assessment programmes.

Because most alternative insecticides are not selective, adverse effects on non-target organisms are to be expected. The assessment of the impact on non-target organisms is one of the prime objectives of ecotoxicology. An in-depth risk assessment can only be achieved if non-target-specific ecological processes are understood. This is a formidable task in tropical and subtropical ecosystems, where many species have not yet been taxonomically described. For ecological monitoring of (large scale) conventional applications, when a prompt assessment is required, it is common practice to confine ecotoxicological studies to higher taxonomic categories (mostly order or family). This simplification ignores the fact that there are significant differences in ecology, behaviour and susceptibility to pesticides even between closely related taxa. As a result, severe effects at lower taxonomic levels might be camouflaged. The search for sensitive species as indicators of undesirable side-effects (e.g. incorrect applications) in further routine field applications of the same insecticide in the same area should therefore receive more attention Grant, 1989; Everts, 1990.

In the first phase of environmental impact studies, investigations should focus on identifying non-target species either directly `ecologically linked' to the target (competitors and natural enemies such as parasites or predators) or occupying key positions within the respective ecosystem (e.g. pollinators, reducers). However, it has to be emphasized that the importance of certain species may be underestimated or overlooked simply because the structure of the ecosystem, the underlying ecological processes and inter-specific relations are not sufficiently understood. Once identified, key species could undergo further standard laboratory testing (toxicity tests) in order to predict ecotoxicological risks of the application of a variety of agents in the same area. These indication methods should not be understood as substitutes for synecological approaches and will have to be worked out together with appropriate field scale assessments during field operations.

In the southern Saharan recession areas of the desert locust (Schistocerca gregaria (Forskål)), tenebrionid beetles are characteristic elements of the ground dwelling coenosis. Isolated plant and animal communities, low species density and diversity, and simple food webs are specific attributes of these ecosystems. Due to their high abundance and their omnivorous trophic function Tenebrionidae dominate the flow of organic matter. Moreover, they are direct food competitors of desert locusts and, as Popov (1980) demonstrated for the Senegalese grasshopper (Oedaleus sengalensis (Krauss) ), potential predators of locust eggs due to their partly hypogaeic way of life. Therefore, in the first phase of the study, tenebrionids were chosen for studying non-target effects of alternative acridicides. The following agents were tested:

1. two formulations of neem oil (seed extract of Azadirachta indica)
2. Beauveria bassiana (Balsamo) Vuillemin, an entomopathogenic fungus 3. dieldrin as a reference.

The potential of neem oil for the control of S. gregaria has so far been tested only in the laboratory. Schmutterer & Freres (1990) demonstrated its larvicidal effects. Furthermore, laboratory reared gregarious hoppers developed the typical anatomical and ethological characteristics of solitary locusts. B. bassiana was field tested against Nomadacris septemfasciata (Serville) as early as the 1930s in South Africa (Schaefer, 1936). Prior & Greathead (1989) recommended additional investigations of the potential of B. bassiana and other entomopathogens as alternative acridicides. Dieldrin used to be the most commonly applied insecticide during locust control operations. The environmental hazards of its use are still controversial (Müller, 1988; Everts, 1990) despite an almost universal ban on its use. Further field investigations of its ecotoxicological effects should therefore be carried out.

The studies were conducted in the Iguidi region about 15 km south-west of the CINA (Centre National de Lutte Antiacridienne) station 'In-Abangharhit' (south-east Tamesna, 18°03'N/6°02'E). The landscape can be described as more or less flat gravel and sand deserts, which are quite regularly separated by lines of dunes up to 20 m high. The vegetation consists of low-growing herbs, which are distributed in isolated patches between the dunes. Vegetation cover is dense along the dunes where patches of Schouwia thebaica (= purpurea)* (Cruciferae) up to 70 cm high (cover ~ 80% can often be found; smaller, less dense patches are found between the dunes. S. thebaica is dominant and constitutes the major food of Schistocerca gregaria in this area. Hemming & Symmons (1969) studied the germination and growth of this annual cruciferous plant. Other species such as Fagonia sp. and Tribulus ochroleucus (Zygophyllacease), Citrullus colocynthis (Cucurbitaceae), Aerva javanica (Amaranthaceae), Pergularia tomentosa (Asclepiadaceae) and Aristida sp. (Gramineae) are loosely scattered within the Schouwia stands. Similarly, the macrofauna is characterized by a small number of species, most of them birds such as Hirundo obsoleta buchanani (pale crag martin), Cursoprius cursor (creamcoloured courser) and Corvus ruficollis (brown-necked raven) or reptiles such as Acanthodactylus boscianus.

*Taxonomy in dispute (eds).

As an adaptation to the dry and hot conditions, most arthropod species are active during the night or at dusk. Tenebrionid beetles make up the largest part of the zoomass. Blaps bifurcata mirei Gredelli, Prionotheca coronata coronata (Olivier) and Pimelia subquadrata chudeaui Koch occur frequently, but are less abundant than Pimelia angulata tschadensis Koch and Trachyderma hispida (Forskål). As with S. gregaria, they feed mainly on Schouwia thebaica.

The cage trials were conducted in a plains area of about 5 ha mainly covered with S. thebaica. For the field applications three dense S thebaica fields of about 500 to 1500 m² were selected on the lee side of sand dunes. For logistical reasons, the trial sites were chosen near each other.

go to next section of this publication