Effects of repeated insecticide pulses on macroinvertebrate drift in indoor stream mesocosms

Stream mesocosms were exposed to three successive (one week apart) 12h pulses of 12 µg/L of imidacloprid to assess macroinvertebrate drift. Pesticide contamination has been reported to result in macroinvertebrate drift, but pesticide pulses in the field usually co-occur with increases in flow velocity, turbidity, and water temperature, all of which can also cause drift. This study aimed to isolate the potential of imidacloprid exposure to initiate macroinvertebrate drift. Imidacloprid pulses were administered at night to simulate nighttime pesticide applications, which reduces immediate photolysis. The first insecticide pulse increased Baetis spp. drift by a factor of more than six, which declined to almost zero after the imidacloprid was flushed out. Corynoneura spp. showed a 17.7-factor drift increase in the first pulse and drift remained high through 12 h after the insecticide was flushed out. Other Orthocladiinae were also affected by the pulses but with a pattern showing effects after the second pulse had already been flushed from the mesocosm.  Tanypodinae did not show an increase in drift. Gammarus roeseli larger than 3.8mm in total length showed a slight increase in drift during the pulse series, and then declined to below the control level. Drift in G. roeseli smaller than 3.8mm in length increased by a factor of 19 after the pulses and then continued to increase to a factor of 37 the day after the pulse even after the imidacloprid had been flushed out. The drift seen in this study resembled passive drift, where the invertebrates seemed to lose their ability to avoid displacement in the stream rather than active drift. Since the current in the mesocosms was not particularly strong, the distance of drift observed may not be ecologically relevant. Their results showed that the impact of pesticides on drift is a function of pulse height and duration as well as the interval between pulses and the possibility of detoxification between them. The authors recommend that the potential of a chemical to induce drift should be considered for lotic surface water risk assessments, or accounted for with a safety factor.