Sensitive indicators of side-effects of pesticides on the epigeal fauna of Arable land

The main objective of the present study was to evaluate the possible impact of pesticides on epigeal arthropods in arable land. It was also envisaged to develop a predictive model for possible undesirable effects of pesticides on the epigeal arthropod fauna using an indicator species from the field. The strategy was the following. In the field, species were identified that were (1) sensitive to a number of pesticides, (2) abundant, (3) regular in time and space and (4) easy to sample and identify (Phase I). Two species, the erigonid spiders Oedothorax apicatus (Blackwall) and Erigone atra (Blackwall) were selected for toxicity studies in the laboratory using deltamethrin as a model compound (Phase II). The results of these studies were used to develop a predictive model which was tested under field conditions (Phase III).

Phase I: Initial Field Observations

Phase I was implemented in the South Flevoland Polder, in cooperation with the State Authority for the Development of the Lake IJssel Polders. Observations on the above-ground arthropods were conducted in 17 commercially farmed fields and in one experimental field, from March 1985 to November 1986, during two growing seasons and one winter. The arthropods were sampled by pitfall traps in oilseed rape fields which were treated with a number of insecticides and herbicides, and in cereal fields which were either treated with herbicides or weeded chemically against herbs, at recommended dosages. In the experimental field, dosages were varied and the effect of ploughing was observed. During the treatments in situ bioassays were carried out with captive spiders. From @ 5100 pitfall samples over 200,000 specimens were identified, belonging to 147 taxa. The observed short term effects were the following. A limited number of species were significantly affected by the treatments with insecticides. These were the carabids Clivina fossor after deltamethrin applications and Trechus quadristriatus after fenitrothion, and 7 spider species. The latter were reduced after treatments with either deltamethrin, fenitrothion or bromophos-ethyl. No effects were observed of herbicide treatments. It was also demonstrated, however, that the effect of mechanical disturbance by ploughing was comparable to the strongest effects of insecticides. It was concluded that an assemblage of spider species consisting of Oedothorax apicatus , Erigone atra , E. dentipalpis , Meioneta rurestris and Bathyphantes gracilis may be a sensitive indicator for early detection of side-effects of insecticides. This group represents 80-90% of the erigonid and linyphiid fauna of arable land and they are common throughout North-Western Europe. The most abundant species, E. atra and O.apicatus were selected for further study.

Possible long-term effects on the spider fauna have been studied in a transversal study in fields with different cultivation histories. The differences were related to the distance between fields rather than to cultural history. There was one exception: in cereal fields that have, since reclamation, been weeded by harrowing, the spider fauna was considerably poorer than in herbicide treated fields.

The effect of deltamethrin on the spiders was observed several times, under different weather conditions. It appeared that the toxicity was highest under hot and dry conditions and low under cold and wet circumstances. This was not in agreement with the known negative temperature-toxicity relationship of pyrethroids and the higher availablity of chemicals from wet surfaces. This relation, therefore, was studied in more detail in the following phase.

Phase II: Toxicity studies

The first step in the analytical phase was the development of new methods for rearing of and toxicity testing with the selected indicators. 0. apicatus females appeared to be the most appropriate species both for rearing and testing.

The second step concerned the availablity of deltamethrin to this species under varying environmental conditions. Three routes of exposure were examined: topical application (simulating exposure to spray); uptake via food, and exposure to contaminated surfaces (soil or glass filter). Part of the work was carried out with radio-labelled deltamethrin. It appeared that uptake of residues from a surface by the tarsi is the most important route. Field observations carried out with a fluorescent tracer showed that the direct exposure of the spiders and their prey species to the spray, was very limited. This indicated that in the field as well, direct exposure and oral uptake are of limited importance. The availability of the residue, however, strongly depends on the water content of the soil. A decrease of the soil moisture content to 66 % of the field capacity resulted in a drop of 75 % in availability.

The third step concerned the relation between environmental factors, i.e. climate and animal antagonists, and the toxicity of deltamethrin. Toxicity tests carried out under varying moisture and temperature conditions showed that at decreasing air humidity the toxicity increased, at all temperatures tested. The compound is most toxic at a combination of a low air humidity or the absence of drinking water and a high temperature. A negative temperature-effect relationship in the range 10 - 30 °C could be observed only at maximum air humidity. The behaviour of the spiders with regard to the moisture in the environment changed after exposure to sublethal dosages of deltamethrin. Spiders not visibly affected showed an increased preference for wet conditions. Affected spiders, on the other hand, showed a reduced capacity to select moist conditions. Their capacity to escape from predators was also reduced. Analysis of gut contents of the main predator, the carabid beetle P. cupreus , from the fields sampled in Phase I, showed an 8-fold increase in the consumption of linyphiids and erigonids after deltamethrin spraying. A few tests were carried out with both captive and reared spiders. There appeared to be a marked difference in sensitivity between these groups. This was primarily imputed to differences in selection mechanisms between laboratory and field.

With these results a regression equation was composed which predicts the toxicity of deltamethrin to O. apicatus females under various weather and soil moisture conditions.

Phase III. Validation under field conditions

Similarly to Phase 1, the field observations of Phase III were carried out in the South Flevoland Polder. A 75 ha field was divided into 27 test plots. These plots were treated three times at different dates, in three replicates, with three dosages of deltamethrin. The following observations were carried out: acute effects in in situ bio assays, activity-density of E. atra and O. apicatus ; deposition of deltamethrin on the soil, on the spiders, on carabid, and on collembola (the main food items of the spiders). Meteorological data were obtained from a near-by meteo station and moisture of the top soil was calculated from data on rainfall, evaporation, and characteristics of the soil and the crop. Only E. atra males were abundant enough at the first and the last spraying date to allow calculation of dose-response relationships. The angles of the regressions appeared to differ for both dates. However, comparison of the observed reduction in O.apicatus females and E. atra males with the toxic effect predicted for O. apicatus females, showed that the laboratory based model had a fair predictive value for the given situation. Apparently erigonid spiders are good indicator species for the assessment of possible undesirable effects of insecticides.