Arch. Environ. Contam. Toxicol. 23, 391-409 (1992)
Archives
of
E nvironmental
c o nmlm tam ination and Iloxicology
© 1992Springer-VerlagNewYorkInc.
Fate and Effects of the Insecticide Dursban ® 4E in Indoor Elodea-Dominated and Macrophyte-Free Freshwater Model Ecosystems: II. Secondary Effects on Community Structure T. C. M. Brock*, M. van den Bogaert +, A. R. Bos +, S. W. F. van Breukelen +, R. Reiche +, J. Terwoert +, R. E. M. Suykerbuyk +, and R. M. M. Roijackers + *DLO the Winand Staring Centre for Integrated Land, Soil and Water Research, P.O. Box 125, 6700 AC Wageningen (correspondence address) and +Department of Nature Conservation, Wageningen Agricultural University, Ritzema Bosweg 32a, 6703 AZ Wageningen, The Netherlands
Abstract. Secondary effects of a single dose of the insecticide Dursban ® 4E (active ingredient chlorpyrifos) were studied in indoor experimental freshwater ecosystems intended to mimic drainage ditches. Two experiments were performed, one in which all model ecosystems were dominated by the macrophyte EIodea nuttallii, and one using systems devoid of macrophytes. In the Elodea-dominated and macrophyte-free model ecosystems, populations of primary producers, herbivores, carnivores and detritivores were indirectly affected via the loss of populations of Arthropoda as a direct result of insecticide application. However, the taxa in which secondary effects were observed differed considerably between these two types of model ecosystem. In macrophyte-dominated systems secondary effects were observed on populations of periphytic algae, the macrophyte Elodea nuttallii, the gastropod Bithynia tentaculata, Turbellaria, and sediment dwelling Oligochaeta. In open water systems it were populations of phytoplankton, the rotators Polyarthra and Asplanchna, bivalves (Sphaeriidae), Hirudinea, sediment dwelling Oligochaeta, and that of the isopod Proasellus coxalis in which secondary effects were observed. In aquatic ecosystems the presence or absence of a well-developed macrophyte vegetation may be a very important characteristic that determines the nature and route of secondary effects induced by pesticides. The differences in secondary effects observed between Elodea-dominated and macrophyte-free model ecosystems indicate that the system's structure and trophic dynamics should be taken into account when predicting ecological effects.
An undesirable side-effect of the use of pesticides in agriculture and horticulture is that these chemicals may enter freshwater ecosystems. Consequently, aquatic organisms may be affected in their survival, growth, or reproduction, due to direct toxicological effects of pesticide residues (the primary effects). Populations of organisms may also be affected in an indirect way when a reduction or elimination of pesticide-susceptible species results in a disturbance of biological interactions and processes.
Ecosystem changes that follow and result from primary effects are termed secondary effects (Hurlbert 1972). The present paper deals with secondary effects of the insecticide Dursban ® 4E (active ingredient chlorpyrifos; O,O-diethyl O-(3,5,6-trichloro-2-pyridyl phosphorothioate), on populations of animals and plants in two types of indoor experimental freshwater ecosystems. Both types of model ecosystem comprised biotic communities obtained from drainage ditches, but differed in the presence or absence of aquatic macrophytes. Macrophyte-dominated systems were treated with two doses of Dursban 4E, with the intention to obtain nominal chlorpyrifos concentrations of 5 (low dose) or 35 (high dose) Ixg/L. Macrophyte-free systems were only treated with the high dose. In view of single species toxicity tests performed in our laboratory (Van Wijngaarden et al., in prep.) and the large body of toxicity data on chlorpyrifos in the literature (Marshall and Roberts 1978; Odenkirchen and Eisler 1988) we postulated that, at the exposure levels used in our study, it would be the group of Arthropoda that predominantly suffered primary effects of the active ingredient chlorpyrifos. The fate of chlorpyrifos and its effect on aquatic arthropods in the indoor experimental freshwater ecosystems have been described in part I of the present series of papers (Brock et al. 1992). Some important results on responses of Arthropoda are summarized in Table 1. In both types of model ecosystem the application of a high dose of chlorpyrifos resulted, at least temporarily, in a substantial reduction in population densities of Cladocera, Copepoda, Amphipoda, Isopoda and Insecta. In the macrophyte-dominated systems that received a low dose the effects were less severe since no significant reductions in Copepoda could be demonstrated and the Cladocera and Isopoda showed a relatively rapid recovery. Most negative effects on the abundancy of Arthropoda could be explained from direct toxicity of chlorpyrifos. The first aim of the present paper is to elucidate shifts in the population densities of algae, macrophytes and invertebrates other than Arthropoda in the experimental freshwater ecosystems, resulting from the application of Dursban ® 4E. A second aim is to elucidate the impact of the presence or absence of macrophytes on the nature and route of secondary effects of
392
T . C . M . Brock et al.
Table 1. Response of various groups of Arthropoda to the application of the low (nominal 5 txg/L) or high (35 tzg/L) dose of chlorpyrifos in the model ecosystems (adapted from Brock et al. 1992). The data represent the mean relative numbers of arthropods in the treated systems, expressed as percentages of the control systems on each sampling date. An asterisk (*) indicates that treated systems differ significantly (p ~< 0.05; Mann-Whitney U test) from controls and a dot (°) is indicative of a trend of difference (0.05 < p 0.05). If these differences occurred on successive sampling dates this is indicated by underlining
"0 t-
10
,
,
-1
1
2
,
4
,
6
,
8
10
12
Week post insecticide application
3. Amounts of loosely attached periphytic algae on shoots of Elodea nuttallii, expressed as p.g periphyton chlorophyll-a per g dry weight macrophyte (mean ± s.d.; n = 4), in control (C), low-dose (L) and high-dose (H) model ecosystems of experiment I. An asterisk (*) indicates a significant difference (p ~< 0.05) from controls; a dot (,) indicates a trend of difference (0.05 < p ~< 0.10) Table
t-tg periphyton chlorophyll-a/g dry weight macrophyte Week after application
4
8
12
16
20
C
357 (127) 809" (372) 1265' (105)
382 (156) 670 (274) 769* (109)
1076 (408) 904 (367) 1592 (409)
644 (134) 611 (251) 1813 (1687)
1272 (654) 731 (666) 2624 (1774)
L H
mean (± s.d.) mean (± s.d.) mean (± s.d.)
the Turbellaria. In macrophyte-dominated systems the most abundant species of this group was Dugesia lugubris (Schmidt), while D. tigrina (Girard) and Bothromesostoma esseni M. Braun were codominant on several sampling dates. On several sampling dates macrophyte-dominated systems treated with Dursban 4E were found to contain significantly lower numbers of the total turbellarian population than control systems (Figures 7A and 7B). This effect was more pronounced in high-dose systems (weeks 8, 10, and 13) than in low-dose systems (week 13). In open water model ecosystems, nearly all individuals of the Turbellaria group were Dugesia tigrina, and a negative effect of insecticide application on densities of Tur-
bellaria could not be shown. Although not significant, mean numbers of turbellarians even were higher in treated than in control open water systems during the post-application period (Figure 7C). Other important carnivorous macro-invertebrates in our model ecosystems were Hirudinea, with Erpobdella octoculata (L.), Helobdella stagnalis (L.), and Glossiphonia heteroclita (L.) as the most important species from a quantitative point of view. In macrophyte-dominated systems densities of Hirudinea on artificial substrates were positively affected by the application of Dursban in weeks t and 2 post-application (Figures 8A and 8B). This effect contrasted with that observed in open
Secondary Effects of Dursban® on Community Structure
397
,;,;,; 1.2 ram) are able to suppress rotifers both by competition for algae as a shared food resource and by mechanical interference, as rotifers are swept into the branchial chambers of feeding cladocerans. In addition, it has been documented that carnivorous copepods (Lampert 1987) and Chaoborus larvae (Christoffersen 1990) can prey on rotifers. Several experimental ecosystem studies have documented that application of Dursban (Hurlbert etal. 1972; Lucassen and Leeuwangh, accepted) or other insecticides (Papst and Boyer 1980; Kaushik et al. 1985; Stephenson et al. 1986; Day et al. 1987; Helgen et al. 1988; Hanazato and Yasuno 1990) disrupted the zooplankton communities dominated by cladocerans and/or copepods and induced the abundant occurrence of rotifers.
Secondary Effects on Carnivores
application was remarkably different between the two types of model ecosystem. In the treated macrophyte-dominated systems the Turbellaria temporarily declined in numbers when compared with controls (Figure 7), while no negative effects on Hirudinea could be demonstrated. In contrast, Hirudinea in treated open water systems significantly declined in numbers (Figure 8), while no negative effects on the abundance of Turbellaria were observed. Several species of Turbellaria (Dugesia lugubris, D. tigrina) and Hirudinea (Erpobdella octoculata, Helobdella stagnalis) which were numerically important in our systems are reported to have greatly overlapping diets, consuming insects, crustaceans, oligochaetes and molluscs (Pickavance 1971; Schtirch and Walter 1978; Young 1981; Wrona et al. 1981; Dall 1983). Therefore, elimination or reduction of arthropods by the insecticide application most probably caused shifts in the diets of these predators, resulting in a more intense resource competition between Turbellaria and Hirudinea. It is possible that differences in foraging behaviour between these two groups of predators can explain the observed differences in their success between macrophyte-dominated and open water systems. It has been reported of Dugesia tigrina, the dominant turbellarian in the open water systems, that they act together when foraging and that they cover substrates with mucus to impede or trap their prey (Pickavance 1971). In contrast to Elodea-dominated systems, the macrophyte-free systems were characterized by small amounts of substrates in the form of solid surfaces that also provide shelter. For this reason both the prey animals and their predators were more concentrated on the surfaces supplied by the artificial substrates. This might have been of greater advantage for the cooperating individuals of Dugesia tigrina than for Hirudinea, which forage individually and which might also have suffered from the mucus secreted by
Dugesia. The response of Turbellaria and Hirudinea, the most important micro-invertebrate predators in our systems, to Dursban 4E
The observation in the high-dose systems dominated by macrophytes that on the artificial substrates the numbers of Hiru-
406
T.C.M. Brock et al.
100
-4-
® 80
tO
60
ZJ
'5
i
40
20
0 -4
-2
1
2
4.
6
8
10
250
13
16
4-
19
22
26
4-
°
4-
19
i
26
® 200
f-'-q CONTROL LOW TREATMENT HIGH TREATMENT
150 +
Y_
p
Archives
of
E nvironmental
c o nmlm tam ination and Iloxicology
© 1992Springer-VerlagNewYorkInc.
Fate and Effects of the Insecticide Dursban ® 4E in Indoor Elodea-Dominated and Macrophyte-Free Freshwater Model Ecosystems: II. Secondary Effects on Community Structure T. C. M. Brock*, M. van den Bogaert +, A. R. Bos +, S. W. F. van Breukelen +, R. Reiche +, J. Terwoert +, R. E. M. Suykerbuyk +, and R. M. M. Roijackers + *DLO the Winand Staring Centre for Integrated Land, Soil and Water Research, P.O. Box 125, 6700 AC Wageningen (correspondence address) and +Department of Nature Conservation, Wageningen Agricultural University, Ritzema Bosweg 32a, 6703 AZ Wageningen, The Netherlands
Abstract. Secondary effects of a single dose of the insecticide Dursban ® 4E (active ingredient chlorpyrifos) were studied in indoor experimental freshwater ecosystems intended to mimic drainage ditches. Two experiments were performed, one in which all model ecosystems were dominated by the macrophyte EIodea nuttallii, and one using systems devoid of macrophytes. In the Elodea-dominated and macrophyte-free model ecosystems, populations of primary producers, herbivores, carnivores and detritivores were indirectly affected via the loss of populations of Arthropoda as a direct result of insecticide application. However, the taxa in which secondary effects were observed differed considerably between these two types of model ecosystem. In macrophyte-dominated systems secondary effects were observed on populations of periphytic algae, the macrophyte Elodea nuttallii, the gastropod Bithynia tentaculata, Turbellaria, and sediment dwelling Oligochaeta. In open water systems it were populations of phytoplankton, the rotators Polyarthra and Asplanchna, bivalves (Sphaeriidae), Hirudinea, sediment dwelling Oligochaeta, and that of the isopod Proasellus coxalis in which secondary effects were observed. In aquatic ecosystems the presence or absence of a well-developed macrophyte vegetation may be a very important characteristic that determines the nature and route of secondary effects induced by pesticides. The differences in secondary effects observed between Elodea-dominated and macrophyte-free model ecosystems indicate that the system's structure and trophic dynamics should be taken into account when predicting ecological effects.
An undesirable side-effect of the use of pesticides in agriculture and horticulture is that these chemicals may enter freshwater ecosystems. Consequently, aquatic organisms may be affected in their survival, growth, or reproduction, due to direct toxicological effects of pesticide residues (the primary effects). Populations of organisms may also be affected in an indirect way when a reduction or elimination of pesticide-susceptible species results in a disturbance of biological interactions and processes.
Ecosystem changes that follow and result from primary effects are termed secondary effects (Hurlbert 1972). The present paper deals with secondary effects of the insecticide Dursban ® 4E (active ingredient chlorpyrifos; O,O-diethyl O-(3,5,6-trichloro-2-pyridyl phosphorothioate), on populations of animals and plants in two types of indoor experimental freshwater ecosystems. Both types of model ecosystem comprised biotic communities obtained from drainage ditches, but differed in the presence or absence of aquatic macrophytes. Macrophyte-dominated systems were treated with two doses of Dursban 4E, with the intention to obtain nominal chlorpyrifos concentrations of 5 (low dose) or 35 (high dose) Ixg/L. Macrophyte-free systems were only treated with the high dose. In view of single species toxicity tests performed in our laboratory (Van Wijngaarden et al., in prep.) and the large body of toxicity data on chlorpyrifos in the literature (Marshall and Roberts 1978; Odenkirchen and Eisler 1988) we postulated that, at the exposure levels used in our study, it would be the group of Arthropoda that predominantly suffered primary effects of the active ingredient chlorpyrifos. The fate of chlorpyrifos and its effect on aquatic arthropods in the indoor experimental freshwater ecosystems have been described in part I of the present series of papers (Brock et al. 1992). Some important results on responses of Arthropoda are summarized in Table 1. In both types of model ecosystem the application of a high dose of chlorpyrifos resulted, at least temporarily, in a substantial reduction in population densities of Cladocera, Copepoda, Amphipoda, Isopoda and Insecta. In the macrophyte-dominated systems that received a low dose the effects were less severe since no significant reductions in Copepoda could be demonstrated and the Cladocera and Isopoda showed a relatively rapid recovery. Most negative effects on the abundancy of Arthropoda could be explained from direct toxicity of chlorpyrifos. The first aim of the present paper is to elucidate shifts in the population densities of algae, macrophytes and invertebrates other than Arthropoda in the experimental freshwater ecosystems, resulting from the application of Dursban ® 4E. A second aim is to elucidate the impact of the presence or absence of macrophytes on the nature and route of secondary effects of
392
T . C . M . Brock et al.
Table 1. Response of various groups of Arthropoda to the application of the low (nominal 5 txg/L) or high (35 tzg/L) dose of chlorpyrifos in the model ecosystems (adapted from Brock et al. 1992). The data represent the mean relative numbers of arthropods in the treated systems, expressed as percentages of the control systems on each sampling date. An asterisk (*) indicates that treated systems differ significantly (p ~< 0.05; Mann-Whitney U test) from controls and a dot (°) is indicative of a trend of difference (0.05 < p 0.05). If these differences occurred on successive sampling dates this is indicated by underlining
"0 t-
10
,
,
-1
1
2
,
4
,
6
,
8
10
12
Week post insecticide application
3. Amounts of loosely attached periphytic algae on shoots of Elodea nuttallii, expressed as p.g periphyton chlorophyll-a per g dry weight macrophyte (mean ± s.d.; n = 4), in control (C), low-dose (L) and high-dose (H) model ecosystems of experiment I. An asterisk (*) indicates a significant difference (p ~< 0.05) from controls; a dot (,) indicates a trend of difference (0.05 < p ~< 0.10) Table
t-tg periphyton chlorophyll-a/g dry weight macrophyte Week after application
4
8
12
16
20
C
357 (127) 809" (372) 1265' (105)
382 (156) 670 (274) 769* (109)
1076 (408) 904 (367) 1592 (409)
644 (134) 611 (251) 1813 (1687)
1272 (654) 731 (666) 2624 (1774)
L H
mean (± s.d.) mean (± s.d.) mean (± s.d.)
the Turbellaria. In macrophyte-dominated systems the most abundant species of this group was Dugesia lugubris (Schmidt), while D. tigrina (Girard) and Bothromesostoma esseni M. Braun were codominant on several sampling dates. On several sampling dates macrophyte-dominated systems treated with Dursban 4E were found to contain significantly lower numbers of the total turbellarian population than control systems (Figures 7A and 7B). This effect was more pronounced in high-dose systems (weeks 8, 10, and 13) than in low-dose systems (week 13). In open water model ecosystems, nearly all individuals of the Turbellaria group were Dugesia tigrina, and a negative effect of insecticide application on densities of Tur-
bellaria could not be shown. Although not significant, mean numbers of turbellarians even were higher in treated than in control open water systems during the post-application period (Figure 7C). Other important carnivorous macro-invertebrates in our model ecosystems were Hirudinea, with Erpobdella octoculata (L.), Helobdella stagnalis (L.), and Glossiphonia heteroclita (L.) as the most important species from a quantitative point of view. In macrophyte-dominated systems densities of Hirudinea on artificial substrates were positively affected by the application of Dursban in weeks t and 2 post-application (Figures 8A and 8B). This effect contrasted with that observed in open
Secondary Effects of Dursban® on Community Structure
397
,;,;,; 1.2 ram) are able to suppress rotifers both by competition for algae as a shared food resource and by mechanical interference, as rotifers are swept into the branchial chambers of feeding cladocerans. In addition, it has been documented that carnivorous copepods (Lampert 1987) and Chaoborus larvae (Christoffersen 1990) can prey on rotifers. Several experimental ecosystem studies have documented that application of Dursban (Hurlbert etal. 1972; Lucassen and Leeuwangh, accepted) or other insecticides (Papst and Boyer 1980; Kaushik et al. 1985; Stephenson et al. 1986; Day et al. 1987; Helgen et al. 1988; Hanazato and Yasuno 1990) disrupted the zooplankton communities dominated by cladocerans and/or copepods and induced the abundant occurrence of rotifers.
Secondary Effects on Carnivores
application was remarkably different between the two types of model ecosystem. In the treated macrophyte-dominated systems the Turbellaria temporarily declined in numbers when compared with controls (Figure 7), while no negative effects on Hirudinea could be demonstrated. In contrast, Hirudinea in treated open water systems significantly declined in numbers (Figure 8), while no negative effects on the abundance of Turbellaria were observed. Several species of Turbellaria (Dugesia lugubris, D. tigrina) and Hirudinea (Erpobdella octoculata, Helobdella stagnalis) which were numerically important in our systems are reported to have greatly overlapping diets, consuming insects, crustaceans, oligochaetes and molluscs (Pickavance 1971; Schtirch and Walter 1978; Young 1981; Wrona et al. 1981; Dall 1983). Therefore, elimination or reduction of arthropods by the insecticide application most probably caused shifts in the diets of these predators, resulting in a more intense resource competition between Turbellaria and Hirudinea. It is possible that differences in foraging behaviour between these two groups of predators can explain the observed differences in their success between macrophyte-dominated and open water systems. It has been reported of Dugesia tigrina, the dominant turbellarian in the open water systems, that they act together when foraging and that they cover substrates with mucus to impede or trap their prey (Pickavance 1971). In contrast to Elodea-dominated systems, the macrophyte-free systems were characterized by small amounts of substrates in the form of solid surfaces that also provide shelter. For this reason both the prey animals and their predators were more concentrated on the surfaces supplied by the artificial substrates. This might have been of greater advantage for the cooperating individuals of Dugesia tigrina than for Hirudinea, which forage individually and which might also have suffered from the mucus secreted by
Dugesia. The response of Turbellaria and Hirudinea, the most important micro-invertebrate predators in our systems, to Dursban 4E
The observation in the high-dose systems dominated by macrophytes that on the artificial substrates the numbers of Hiru-
406
T.C.M. Brock et al.
100
-4-
® 80
tO
60
ZJ
'5
i
40
20
0 -4
-2
1
2
4.
6
8
10
250
13
16
4-
19
22
26
4-
°
4-
19
i
26
® 200
f-'-q CONTROL LOW TREATMENT HIGH TREATMENT
150 +
Y_
p
