Integrated Environmental Assessment and Management — Volume 7, Number 4—pp. 577–586 ß 2011 SETAC
577
Comparison of Environmental Risks of Pesticides Between Tropical and Nontropical Regions Francisco Sanchez-Bayo,*y and Ross V Hynez y Centre for Ecotoxicology, University of Technology Sydney, PO Box 29, Lidcombe, NSW 1825, Australia z Centre for Ecotoxicology, Office of Environment and Heritage (NSW), Lidcombe, Australia
(Submitted 5 December 2010; Returned for Revision 1 February 2011; Accepted 28 February 2011)
ABSTRACT
Keywords: Pesticides
Half-life
Leaching
Volatilization
INTRODUCTION In recent years, there has been a shift in focus in regard to risk assessment of agrochemicals from temperate to tropical regions (Ro¨mbke et al. 2008; Sarkar et al. 2008). Two main reasons may explain this trend: first, developed countries, situated in temperate regions, are shifting toward a chemicalfree agriculture (so-called ‘‘organic’’ agriculture) (Dimitri and Greene 2002) or decreasing pesticide use as a result of improvements in agronomic practices (Pimentel et al. 2005), whereas developing countries, most of which are in tropical regions, are increasing their use of pesticides and fertilizers as they become wealthier (Tan et al. 1998). Second, as developed countries restrict the use of many substances that pose risks to the environment, developing countries are manufacturing the same old chemicals, which usually have the greatest risks (Ecobichon 2001) and are using them without restraint in their own agriculture (Heong et al. 1994; Zhou and Jin 2009). This is understandable, because many of the older pesticides have run out of patent and are now being mass-produced by small companies in developing countries, which sell them at cheaper prices than the new chemicals produced by foreign enterprises. Thus, not only has agrochemical usage increased in tropical regions (Wilson and Tisdell 2001), where they are applied indiscriminately by farmers who are unaware of health and environmental hazards (Calumpang 1996), but also the overall human and ecological risks may have increased as a result of applying older and potentially more hazardous pesticides (Catan˜o et al. 2008). Health issues aside, it has become necessary to evaluate whether the environmental
* To whom correspondence may be addressed:
[email protected] Published online 24 March 2011 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ieam.189
Fugacity modeling
risks of such chemicals are greater or lesser under tropical conditions than in the temperate regions (Daam and Van den Brink 2010). Research on the fate of pesticides in tropical regions is sparse, spanning less than 3 decades. By contrast, the environmental fate and risk assessment of pesticides have been extensively studied in North American and European countries since the early 1960s (Brock et al. 2006). In order to evaluate the risks of chemicals between climatic and/or geographical regions, 2 aspects need to be addressed: 1) a comparison of susceptibility to chemicals among species from either region, and 2) a comparison of the behavior of chemicals between those environments. Although tropical aquatic organisms appear to be more sensitive to some organic chemicals and less sensitive to metal contaminants than their temperate counterparts (Kwok et al. 2007), the extrapolation of ecological risks between climatic regions also requires an understanding of the physicochemical factors involved in the dissipation and movement of chemicals, as well as the ecological characteristics of the areas in which they are applied. The latter characteristics are not considered here, because they are often dependent on biogeographical features of the areas concerned. Tropical regions are defined as those lying between the tropics of Cancer and Capricorn (23.48N and 23.48S), where environmental conditions are characterized by annual average temperatures in the range of 20 to 30 8C, and annual precipitation usually above 700 mm (even > 3000 mm in the wet tropics) except for the arid tropics. Intense rainfall events often result in loss of the fertile top layers of the soil. In addition, soils are rather deficient in organic carbon (OC) because of the intense microbial degradation that takes place uninterrupted throughout the entire year, whereas soils in temperate regions are rich in OC because their microbial activity is restricted only to the warm seasons (Lugo et al. 1986). Temperate regions lie between the tropics and the polar circles (23.4–66.58N and 23.4–66.58S) and are more variable in climatic conditions, with temperature ranges
Health & Ecological Risk Assessment
A comparison of environmental risks of pesticides between tropical and nontropical regions has been performed, using data from the literature and modeling outputs based on the physicochemical properties of the compounds. With a few exceptions, the level of risk of exposure for most pesticides in tropical agriculture is similar to that in other climatic regions of the world. Generally, dissipation of pesticides increases under the warm and wet conditions of the tropics, with most of the dissipation occurring through hydrolysis in water and biological degradation in water and soil. High temperatures in the tropics also foster volatilization rates, whereas high precipitation and poor soils tend to increase losses into runoff and, for certain chemicals, affects their leaching behavior. The environmental risk is determined by a balance of soil types, soil organic carbon, pH, and the rates of degradation in the various environmental compartments. Integr Environ Assess Manag 2011;7:577–586. ß 2011 SETAC
578
generally lower than in the tropical regions, even during the summer cropping season. This article compares the environmental behavior of pesticides under the conditions of both temperate and tropical regions. By understanding their different dissipation patterns and their level of persistence and mobility in these environments, the extent of their risk in the tropical regions can be properly assessed. In this context, it is important to realize that apart from the physicochemical properties, which are characteristic of each chemical, external factors such as temperature, precipitation, soil pH, and OC content influence the risk of exposure that an agrochemical may have in a particular area.
MATERIALS AND METHODS Data collection A literature search from the last 25 years was carried out in order to find information on the fate of pesticides in tropical countries. A total of 137 articles were found, but only 66% of those provided useful data (i.e., values for specific variables). The articles came from 26 countries, with India and Brazil composing 53% of all the studies, followed by Malaysia, Caribbean islands, Thailand, Taiwan, and 20 others. A total of 73 pesticides (36 herbicides, 33 insecticides, 2 fungicides, and 2 others) were covered, with the most commonly studied being DDT (24% of studies), atrazine, endosulfan and chlorpyrifos (> 10%), malathion, metolachlor, lindane, and imidacloprid (> 5%). Except for the last insecticide, all the other chemicals have already run out of patent. Data sought included dissipation half-lives (DT50) in water and soil, which were worked out in the majority of cases from field trials, or when no field data were available, from controlled laboratory experiments: adsorption and desorption isotherms and their corresponding OC partitioning coefficient (KOC) values for a range of representative soils of the countries concerned, and volatility and leaching data for a number of chemicals. Data on half-lives and sorption from temperate regions were obtained from the FootPrint online database (Anonymous 2009), the Pesticide Manual (Tomlin 2009), and Pesticide Profiles (Kamrin 1997). In addition, these 3 sources provided information on the characteristic physicochemical properties of the pesticides, namely solubility in water, vapor pressure, and octanol–water partitioning coefficient (KOW). The conditions of laboratory testing in tropical countries usually conformed with the standards applied in any other country. However, field conditions were very different, because they depended on the characteristics of the particular countries in which the research was carried out. In any case, only the data obtained under reliable conditions either in the field or the laboratory were considered. In general, the pH of the waters was above 7.5 (58% of data), with the remaining studies being conducted either in acidic (22%) or neutral waters (20%). The soil pH, on the contrary, was acidic in 59% of the cases studied, with the remainder being alkaline (23%) or neutral (18%). Acidic soils are common in Brazil and Southeast Asian countries; soils from some African countries and Caribbean islands tend to be neutral, whereas those from the Indian subcontinent are alkaline in most cases. A large proportion (58%) of the soils tested, whether in the field or laboratory, had normal OC contents, i.e., between 1.5 and
Integr Environ Assess Manag 7, 2011—F Sanchez-Bayo and RV Hyne
3%, but one-third could be considered rather poor ( 0.05 as determined by Student t test) of dissipation or sorption between temperate or tropical environments, the model output would be very similar under both conditions. Therefore, only those chemicals that showed significant differences in regard to their dissipation or sorption values between the 2 environments were selected for modeling the exposure. Thus, only a subset of 17 pesticides were considered here: 8 herbicides, 5 insecticides, 1 fungicide,
579
and 3 organochlorine insecticides (endosulfan, lindane, and DDT) that are still used in tropical countries and have the characteristic of being very volatile. This latter feature implies that their behavior would likely be affected by the temperature difference between the 2 scenarios, even if their dissipation and sorption rates did not change significantly. For the purpose of our comparison, concentrations were estimated for a theoretical application rate of 1 kg/ha1 of agricultural land (Table 1). Values of the dissipation variables (e.g., DT50 in water and soil) and sorption to soil (e.g., KOC as a function of the percent OC in soil and the KOW) for tropical environments were obtained from the literature search, whereas values for temperate regions were taken from the databases already mentioned.
RESULTS Comparison of data A comparison of DT50s in water for 12 pesticides for which suitable data was obtained showed that 91% of the
Table 1. Set parameters and variables used for steady-state calculations with fugacity modeling level 3 Parameters Constants
Description
Value in model 106 m2
Area of system Area fraction of soil Area fraction of vegetation
0.9 a
Area fraction of water
0.3 m
Pore water fraction of soil
0.15
Soil bulk density
1.4
b
0.04
Depth of water
0.5 m
Concentration of suspended solids
1.25 g L1
Biomass fraction in water (plankton)
1.8 104
Air mass transfer
90 m3 d1
Rate of air diffusion
0.1 m d1
Water leaching rate
1 m s1 2 mm d1
DT50 in soil
As per literature
DT50 in water
As per literature
Soil-partitioning (KOC)
See textc
Organic carbon fraction of soil (OC)
0.015 (T), 0.03 (NT)
Temperature
30 8C (T), 20 8C (NT)
Overlapping soil area. Fraction of carbon in biota (i.e., vegetation, plankton). c Determined in the model as a function of percent OC and KOW. b
0.3 m
Biomass factor
Wind speed
a
0.1
Soil depth
Height of vegetation
Variables
0.6
580
Integr Environ Assess Manag 7, 2011—F Sanchez-Bayo and RV Hyne
follow a similar pattern of dissipation as found for the other pesticides. Adsorption to soil, measured as KOC, depends on 2 factors: 1) the KOW, which is a characteristic value of the chemical, and 2) the OC content and soil type. The data collected from tropical countries indicate that some 45% of the herbicides and fungicides examined showed higher KOC values than they normally have in soils from temperate regions (Figure 1A). The ratio T/NT was significantly higher ( p < 0.05) in the case of acetochlor (30 times) and the fungicide carbendazim (8 times) in Brazilian soils (Ferri et al. 2005; Carbo et al. 2007). On the other end of the spectrum, paraquat, oxifluorfen, and terbutryn showed significantly lower KOC values under tropical conditions: between 6 and 50 times less sorption for a range of clayey, loamy, and sandy soils from Taiwan (Yen et al. 2003), Thailand (Amondham et al. 2006), and Brazil (Barriuso et al. 1992). In the case of insecticides, however, there seemed to be a trend toward more adsorption onto tropical soils, with 77% of the compounds showing slightly higher KOC values, but none of them to a significant extent (Figure 1B). The exception is permethrin, which appears to adsorb much less (74 times) on tropical soils of Malaysia, but this could be because the only data available are from predominantly sandy soils (Ismail and Kalithasan 2004), which have low sorption capacities.
chemicals (all but malathion, which practically shows no difference) have shorter half-lives in tropical regions than in temperate ones, and such differences appear to be significant for at least 3 compounds (carbendazim, parathion-methyl, and chlorpyrifos) (Table 2). Among the chemicals tested, 9 pesticides (75%) degraded twice as fast in the warm tropical waters. Dissipation rates from tropical soils are also twice as fast for half of the insecticides and for 55% of the herbicides compared. At least 5 herbicides (trifluralin, paraquat, 2,4-D, triclopyr, and simazine) and 2 insecticides (diazinon and deltamethrin) showed significantly faster dissipation ( p < 0.05) from tropical soils compared to their dissipation under temperate conditions. Volatile compounds make a large proportion (38%) among the pesticides with faster dissipation in tropical soils. For instance, lindane dissipation from Indian soils during the summer season (DT50 ¼ 24–45 d) is 7 times faster than in temperate regions, and this is mainly because of volatilization losses (Samuel and Pillai 1991), which could be as high as 86% of the applied rate (Samuel et al. 1988; Tanabe et al. 1991). In the case of DDT, volatilization may account for 8 to 30% of the total dissipation in tropical countries (Nair et al. 1992) but only 1% in temperate regions, whereas losses of endosulfan by volatilization from tropical soils are estimated at 5 to 20% (Singh et al. 1991). On the other extreme, longer half-lives in soils from tropical environments were found in the case of 3 herbicides (chlorthal-dimethyl, anilofos, and oxyfluorfen) and 3 insecticides (carbofuran, fipronil, and parathion), which amount to some 13% of the 46 compounds for which appropriate data was obtained (Table 3). Data on soil halflives for the fungicides carbendazim and chlorothalonil could not be found in the published literature from tropical countries, but it can be assumed these types of compounds
Modeling exposure Concentrations in water and soil of 17 selected pesticides (8 herbicides, 8 insecticides, and 1 fungicide) were obtained for both tropical and temperate conditions using the aforementioned model system. All parameters were fixed except temperature, soil OC, and dissipation rates of the chemicals in the respective compartments, which varied according to the regional conditions (Table 1).
Table 2. Comparison of half-lives (days) in water Tropical (T) Compound
Type
a
Chemical group
range
typical
range
typical
T/NT
Malathion
I
OPb
2.0–5.1
3.4
1–107
3
1.13
Carbaryl
I
Carbamate
4.9–9.0
5.8
1–12
7
0.89
Endosulfan
I
Cyclodiene
7.0–7.3
7.2
1–20
9
0.79
Glyphosate
H
–
7.6–12.6
10.1
3–70
28
0.36
Atrazine
H
Triazine
17.3–18.6
18.0
10–105
67
0.27
Chlorpyrifos
I
OP
3.3–7.8
5.6
21–78
25
0.22
Dimethoate
I
OP
3.2–4.8
4.1
5–175
19
0.22
Parathion-Me
I
OP
2.6–6.9
4.3
33–68
29
0.15
Carbofuran
I
Carbamate
3.9–35.0
5.2
28–56
43
0.12
Carbosulfan
I
Carbamate
1.1–3.2
2.2
1–72
27
0.08
Carbendazim
F
–
4.6–6.4
5.2
60–350
136
0.04
Fenitrothion
I
OPb
0.5
0.5
1–75
27
0.02
F ¼ fungicide; H ¼ herbicide; I ¼ insecticide. Organophosphorus insecticide. p < 0.05 (t test) a
b
Temperate (NT)
Risk Assessment in Tropical Regions—Integr Environ Assess Manag 7, 2011
581
Table 3. Comparison of half-lives (days) in soila Tropical (T) Type Herbicides
Compound
Range
Typical
Range
Typical
T/NT
Chlorthal-dimethyl
17–289
153
14–98
47
3.25
Anilofos
32–86
59
30–45
38
1.57
Oxyfluorfen
15–242
99
35–180
73
1.35
Butachlor
0.3–14
10
4–18
11
0.92
Dithiopyr
17–38
28
17–61
39
0.72
Ametryn
17–33
25
37–60
37
0.67
Terbuthylazine
9–210
49
30–167
75
0.65
Metsulfuron-methyl
7–13
10
4–29
17
0.63
Glyphosate
9–50
31
4–174
49
0.63
Metolachlor
14–47
33
8–100
54
0.61
Prometryn
2–43
23
30–90
41
0.55
11
11
14–79
21
0.51
Alachlor
1–15
7
7–35
14
0.48
Imazethapyr
19–46
33
14–290
90
0.36
Pendimethalin
13–1396
32
27–186
107
0.30
Epoxiconazole
20–29
25
52–226
120
0.20
Simazine
9–17
13
27–102
65
0.20
Hexazinone
20–21
21
30–182
106
0.19
Atrazine
5–41
13
28–108
75
0.17
Diuron
15–31
16
20–231
90
0.17
1–4
3
7–54
31
0.09
2,4-D
0.5–1.3
1
10–14
12
0.08
Paraquat
36–46
41
480–4745
2613
0.02
Trifluralin
1–4
2
35–375
170
0.01
Carbofuran
23–46
31
5–27
16
1.91
Fipronil
9–200
97
6–135
65
1.50
Parathion
5–84
36
2–58
30
1.20
Dimethoate
3–22
8
4–10
7
1.09
Lambda-cyhalothrin
5–43
22
6–40
23
0.96
Cadusafos
22–43
36
11–55
38
0.93
Permethrin
11–24
17
13–42
28
0.63
Carbosulfan
10–23
12
10–72
21
0.57
Acephate
1–3
2
2–7
5
0.52
Mevinphos
2–4
3
1–12
7
0.51
Bifenthrin
36–41
39
65–267
85
0.45
Deltamethrin
6–15
8
18–35
21
0.39
Triazophos
8–12
10
7–46
27
0.38
Malathion
2–17
5
1–25
13
Thiobencarb
Triclopyr
Insecticides
Temperate (NT)
0.37 (Continued )
582
Integr Environ Assess Manag 7, 2011—F Sanchez-Bayo and RV Hyne
TABLE 3. (Continued ) Tropical (T) Type Insecticides
Compound
Temperate (NT)
Range
Typical
Range
Typical
T/NT
Endosulfan
3–96
31
28–150
89
0.34
Phorate
15–17
16
14–90
63
0.25
Monocrotophos
1–12
5
1–35
18
0.25
Diazinon
3–6
4
8–30
19
0.22
24–2000
44
121–450
286
0.15
Chlorpyrifos
1–10
7
11–141
50
0.14
Cypermethrin
6–8
8
14–199
69
0.11
72–10 000
226
1460–10 950
6205
0.04
Lindane (g-HCH)b
DDT NT ¼ nontropical; T ¼ tropical. g-HCH ¼ gamma-hexachlorocyclohexane. p < 0.05 (t test).
a
b
The actual concentrations calculated by the model are of little interest, because they refer to equal theoretical applications for all compounds (1 kg ha1) and do not match the amounts applied for each of them in real agricultural
Figure 1. Comparative sorption of 21 herbicides and 1 fungicide (a) and 13 insecticides (b) between tropical (T) and temperate (NT) environments.
settings. Our interest here is in the ratio of the estimated concentration for the 2 scenarios, tropical versus temperate (T/NT). Figure 2 shows that the ratio of concentrations in soil does not appear to differ much across the whole gamut of chemicals (from 1.00 for oxyfluorfen and acetochlor to 0.70 for 2,4-D) except in the case of the herbicide triclopyr (T/NT ¼ 0.11), which in tropical soils would have its concentration reduced 9 times. Water concentrations, on the contrary, vary among chemicals: 1) those of trifluralin, oxyfluorfen, acetochlor, and terbutryn would increase in tropical waters by a factor of 2 (Figure 2), so their exposure risk to aquatic organisms would be greater in tropical environments; 2) most pesticides (60%), including all insecticides, are considered to have a nonsignificant increase in concentrations, and therefore risk,
Figure 2. Concentration ratios for 17 selected pesticides in water and soil as determined by fugacity modeling level 3, using the parameters and variables described in Table 1.
Risk Assessment in Tropical Regions—Integr Environ Assess Manag 7, 2011
in tropical waters; and 3) concentrations in water for the herbicides triclopyr, 2,4-D, and paraquat would not change, so their risk of exposure is expected to be the same in the tropics as in temperate regions. Looking in more detail, it is apparent that chemicals with longer half-lives in tropical environments (DT50 T/NT > 1.0) also show higher concentrations and consequently higher risk of exposure, but such risk is unrelated to the adsorption to soil (r ¼ 0.21, exposure T/NT versus KOC T/NT). For the modeling conditions used, the exposure risk in water, which was measured as the T/NT for water concentrations, increases twice as much whenever the DT50 ratios in water are > 1.2, whereas the risk decreases for ratios < 0.5. Risk of exposure to soil, measured as the T/NT for soil concentrations, appears to be always lower under tropical conditions, because the DT50 ratios from soil were also < 1.0 in all cases except for oxyfluorfen (T/NT ¼ 1.35), which turned out to show no difference in soil concentrations anyway. As indicated above, the risk of exposure to triclopyr in soil is significantly less in tropical environments because its DT50 ratio in soil is < 0.1 and its ratio for KOC is < 1.0.
Modeling losses by volatilization and leaching Fugacity level 3 allows estimation of losses by degradation as well as transport processes such as volatilization, leaching, and runoff. Volatilization and leaching are intrinsically linked
583
to the physicochemical properties of the compounds, more specifically to vapor pressure and water solubility, even though external factors such as temperature, wind speed, and soil type affect the rates at which chemicals move into other phases. Among these factors, only the average ambient temperature and soil OC were considered in the model, because the other factors are not specific to either the tropical or temperate regions. Runoff losses were excluded from this modeling, because they are highly dependent on topography, soil type, rainfall events, vegetation cover, and other variables that are not characteristic of any particular region. Eight (47%) among the 17 selected compounds have high volatilization rates (> 0.1 g d1) in both environments (Table 4). Although it is essential to know how much pesticide can be lost by volatilization, for the purposes of this discussion, it is more important to know which chemicals change their volatility behavior, and the extent of that change, between the 2 environments under consideration. Looking at the volatilization T/NT ratios, it can be seen that 4 herbicides (trifluralin, oxyfluorfen, acetochlor, and terbutryn) increase their volatility under our modeled tropical conditions by almost a factor of 2, whereas 2 other herbicides (2,4-D and paraquat) have no appreciable reduction because they are polar or ionic, not volatile compounds. With the exception of trifluralin, which is very volatile in either environment, the change in behavior for the other 3 compounds is only a matter of degree: under tropical
Table 4. Volatilization and leaching ratesa as determined by fugacity modeling level 3 for 17 selected pesticides Volatilization (g d1)
a
Leaching (g d1)
Chemical
Tropical (T)
Temperate (NT)
T/NT
Tropical (T)
Temperate (NT)
T/NT
Trifluralin
0.93
0.48
1.94
1
0
2.00
Oxyfluorfen
0.03
0.01
1.93
2
1
2.00
Acetochlor
0.01
0.00
1.93
9
5
1.99
Terbutryn
0.01
0.00
1.92
29
14
1.99
Simazine
0.00
0.00
1.74
821
455
1.80
DDT
0.00
0.00
1.63
0
0
1.68
Endosulfan
0.70
0.43
1.63
3
2
1.68
Permethrin
0.01
0.00
1.63
0
0
1.68
Chlorpyrifos
0.32
0.20
1.63
4
2
1.68
Deltamethrin
0.04
0.02
1.63
5
3
1.68
Diazinon
0.36
0.22
1.62
37
22
1.68
Lindane
0.85
0.52
1.62
37
22
1.68
Parathion-methyl
0.22
0.14
1.61
183
110
1.67
Carbendazim
0.74
0.51
1.46
2042
1349
1.51
Triclopyr
1.52
1.46
1.04
3873
3589
1.08
2,4-D
0.00
0.00
0.97
4190
4170
1.00
Paraquat
0.00
0.00
0.97
2
2
1.00
Estimated for an input of 100 kg in the model (1 kg ha1 100 ha).
584
conditions, their losses by volatilization are still at very low levels (0.01–0.03 g d1). It is worth noting that the most volatile compound, triclopyr, practically does not change its volatilization rate between the 2 environments (1.5 g d1). The remaining 6 volatile pesticides tend to increase their volatility under the warmer conditions of the tropics, although this change appears to be of little or no significance. Five compounds (30%) among the selected pesticides could leach at rates higher than 100 g d1, which is 0.1% of the applied rate in our model (Table 4). Once again, because these rates are rather theoretical, what matters here is to compare the change in leaching behavior between the 2 environmental conditions. Interestingly, the ratio in leaching rates between the 2 environments for all selected chemicals is similar to that of volatilization. As it can be seen in Figure 3, the largest changes in leaching go in parallel with those of volatilization, in such a way that those pesticides which volatilize more in tropical regions would tend to also leach more in that environment, and vice versa. This is particularly noticeable among the insecticides modeled here. However, it should be emphasized that only the increase of simazine rates (T/NT ¼ 1.8) appears to be of some concern, and perhaps parathion-methyl and carbendazim (T/NT > 1.5), because the other compounds that show high leaching ratios (i.e., trifluralin, oxyfluorfen, and acetochlor) practically do not move in the soil profile. Losses by degradation are more relevant in tropical regions, as can be seen in Figure 3. The herbicides trifluralin and paraquat in particular show rates of dissipation higher than 50 times compared to nontropical environments, whereas another 6 compounds disappear at more than double the normal rate in temperate countries. The exceptions, as already pointed out, are the herbicides 2,4-D and oxyfluorfen: for the latter compound, losses by volatilization and leaching take priority under tropical conditions.
DISCUSSION The data gathered here clearly show that dissipation of pesticides in tropical regions is greater than in temperate countries, particularly in freshwater environments (Table 2).
Figure 3. Estimated losses by degradation and transport by volatilisation or leaching for 17 selected pesticides in tropical and temperate environments.
Integr Environ Assess Manag 7, 2011—F Sanchez-Bayo and RV Hyne
It is certain that higher ambient temperatures are responsible for the increase in hydrolysis rates in tropical waters, which particularly favors the dissipation of organophosphorus insecticides (Sharmila et al. 1988; Chai et al. 2009). Apart from this chemical process, biological degradation carried out by algae, fungi, and bacteria is likely enhanced under the warm and wet conditions of the tropics (Racke et al. 1997; Arbeli and Fuentes 2007). Repeated applications of carbofuran to the soil led to enrichment of carbofuran-degrading microorganisms at soil temperatures of 28 and 35 8C and not at 6 8C, and the enrichment was more pronounced at 35 than at 28 8C (Ramanand et al. 1988). This study indicates that intensive use of the same pesticide may lead to more rapid buildup of an enriched population of very active pesticidedegrading microorganisms under hot and humid conditions of the tropical environment than would occur in a temperate environment. In a review of microbial pesticide degradation in tropical soils, Sethunathan and colleagues concluded that acceleration of microbial activities due to elevated temperatures was the major factor responsible for observations of increased degradation of pesticides under tropical rice paddy soil conditions (Sethunathan et al. 1982). As a result of this faster dissipation, the exposure of aquatic organisms to toxic waterborne residues of pesticides is likely to be reduced in tropical waters, because most chemicals will last a shorter time than they usually last in rivers, lakes, and ponds of temperate countries (Anyanwu and Odeyemi 2002). Dissipation from soil, however, is only enhanced for half of the chemicals used. Although chemical and biological degradation by soil microorganisms could be responsible for most of this dissipation (Figure 3), the effect of higher temperatures on volatilization from soil surfaces combined with the greater leaching under conditions of high precipitation play an important role in the overall dissipation process in tropical agricultural regions (Table 4). Further losses can also be due to runoff as a consequence of intense rainfall events such as monsoonal cyclones. The extent of these losses is deemed to be quite variable and cannot be discussed here, but they inevitably affect the residue concentrations found in waterways (Table 5). For instance, our modeling indicates that in the absence of runoff and despite the fast dissipation rates in the tropics, residue concentrations in water would increase for most pesticides ( 82% of compounds), solely as a result of their higher desorption from poor OC soils (Figure 2); this means that additional residues by runoff inputs would only exacerbate the ecological risk in aquatic environments. The critical factor affecting the movement from soil into the water phase appears to be soil OC content, although pH can also play a significant role in the case of polar herbicides and acidic compounds; other factors such as soil aeration and porewater fraction, which may be influenced by the activity of earthworms and soil larvae, were considered constant (Table 1) because no estimates of such biological influences are available. For soils with equal OC content in the 2 environments, the concentrations of residues in tropical waters would only vary between 0.85 to 1.0 times the concentrations found in temperate regions; by contrast, a 50% reduction of soil OC would result in increased water concentrations in the tropics by a factor in the 1.0 to 3.0 range. Even if the effect of soil OC is not directly proportional to the estimated concentrations in water (r ¼ 0.22), the adsorption and desorption of pesticides from soil is likely to be governed by this environmental factor.
Risk Assessment in Tropical Regions—Integr Environ Assess Manag 7, 2011
585
Table 5. Summary of factors affecting the environmental risk of pesticides under tropical conditions Factor Temperature
Tropics High
Comparative effect
Comparative risk of exposure
Faster degradation
Shorter exposure to organisms
Shorter recovery time
More long-range transport
Increased volatilization Rainfall
Soil OC
Soil pH
High
Increased runoff
More residues in surface waters
Increased leaching
More mobility to groundwater
Low
Less adsorption
More mobility into water and air
High
Typical adsorption
No change
Low
Desorption of acidic herbicides
More residues available
Persistence of organophosphorus and acidic herbicides
Longer exposure to organisms
Normal degradation
No change
High
Although the final amount of residues moving into the waterways is compound-specific, this kind of modeling predicts an increase in water concentrations in the tropics whenever the soils are poorer than those of temperate regions (Table 5). This does not contradict the statement above regarding a shorter exposure in tropical waters: no matter how high the residue levels may be, the chemicals will certainly disappear faster than they normally do in colder environments. After all, toxic effects on aquatic organisms depend not only on the concentration of the toxicant but also on the duration of exposure. Consequently, any adverse effect will be of short duration in the tropics and, therefore, the recovery of individuals and populations of small aquatic organisms will be enhanced as well (Van den Brink et al. 2002; Traas et al. 2004). The warm conditions of the tropics also affect the dissipation by volatilization, particularly in the case of insecticides and fungicides that have high vapor pressures (Table 4). It is important to realize that volatilization is already factored in when carrying out dissipation experiments under field conditions: some of the pesticide disappears into the air, whereas the remainder experiences chemical and biological degradation. For this reason, recalcitrant organochlorines such as DDT and g-hexachlorocyclohexane (lindane) appear to dissipate much faster in the tropics (Table 3). In the case of endosulfan, its dissipation from tropical soils is 3 times as fast as in temperate regions (Raupach et al. 2001), and this is largely due to its higher volatilization under tropical conditions (Singh et al. 1991). The implications of pesticide volatilization for the local environment may be regarded as positive, because the chemical residues of these compounds will not last too long in the agricultural areas applied. However, it is known that their airborne residues will eventually deposit in regions with colder temperatures such as the Arctic (Weber et al. 2009) and mountain tops (Blais et al. 1998; Daly et al. 2007), through the so-called global distillation process (Simonich and Hites 1995). Thus, volatilization of these and other pesticides must be considered an undesirable feature from the point of view of their longrange transport around the globe (Shen et al. 2005). Finally, losses by leaching through the soil profile are almost the same in either environment except for the
herbicide simazine and the fungicide carbendazim (Table 4), so the risk of groundwater contamination for most pesticides applied in the tropics is not greater than that in other countries. As in the case of runoff, losses by leaching are highly dependent on the amount of precipitation, but in this case, the soil type (e.g., sandy versus clayey) could be decisive to allow the movement of residues downward into the water table.
CONCLUSIONS The fate and transport of pesticides applied to agricultural areas in tropical regions is in many ways similar to their fate and transport in other climatic regions of the world, and their ecotoxicological risk is likely to be similar (Daam and Van den Brink 2010). Differences regarding the faster dissipation rates in warm and wet environments may translate into shorter exposures and higher rates of recovery of the organisms affected in tropical regions. However, the high precipitation in the tropics will undoubtedly increase the risk of runoff for most compounds and foster the leaching potential of certain chemicals. At the same time, other features specific to particular tropical regions may accentuate the ecological risks: poor soils may lead to a higher mobility of residues into the water phase, whether surface water or groundwater, and acidic soils contribute to this mobility, thus exacerbating the exposure of the soil microfauna and microbial communities to herbicides and other polar compounds. In summary, the balance between positive and negative risks appears to depend on the characteristics of the specific soils in each region, rather than on climatic characteristics of the tropics.
REFERENCES Amondham W, Parkpian P, Polprasert C, DeLaune RD, Jugsujinda A. 2006. Paraquat adsorption, degradation, and remobilization in tropical soils of Thailand. J Environ Sci Health B 41:485–507. Anonymous. 2009. FootPrint Database. In: Dubus I, editor: European Commission, Sixth Framework Programme. Anyanwu CU, Odeyemi O. 2002. Malathion degradation in tropical soil and aquatic ecosystems. Global J Pure Appl Sci 8:1–6.
586
Integr Environ Assess Manag 7, 2011—F Sanchez-Bayo and RV Hyne
Arbeli Z, Fuentes CL. 2007. Accelerated biodegradation of pesticides: An overview
Racke KD, Skidmore MW, Hamilton DJ, Unsworth JB, Miyamoto J, Cohen SZ. 1997.
of the phenomenon, its basis and possible solutions; and a discussion on the tropical dimension. Crop Prot 26:1733–1746.
Pesticide fate in tropical soils. Pure Appl Chem 69:1349–1371. Ramanand K, Panda S, Sharmila M, Adhya TK, Sethunathan N. 1988. Development
Barriuso E, Feller C, Calvet R, Cerri C. 1992. Sorption of atrazine, terbutryn and 2,4-D
and acclimatization of carbofuran-degrading soil enrichment cultures at
herbicides in two Brazilian oxisols. Geoderma 53:155–167. Baskaran S, Sa´nchez-Bayo F, Kennedy IR, Mortimer MR. 2002. Environmental
different temperatures. J Agric Food Chem 36:200–205. Raupach MR, Briggs PR, Ahmad N, Edge VE. 2001. Endosulfan transport II.
concentration of pesticides in cotton production systems: risk assessment by fugacity modelling. In: Kookana RS, Sadler R, Sethunatham N, Naidu R, editors.
Modeling airborne dispersal and deposition by spray and vapor. J Environ Qual 30:729–740.
Environmental Protection and Risk Assessment of Organic Contaminants. Plymouth: Science Publishers, Inc. p 79–94.
Ro ¨mbke J, Waichman AV, Garcia MVB. 2008. Risk assessment of pesticides for soils of the Central Amazon, Brazil: comparing outcomes with temperate and
Blais JM, Schindler DW, Muir DCG, Kimpe LE, Donald DB, Rosenberg B. 1998. Accumulation of persistent organochlorine compounds in mountains of
tropical data. Integr Environ Assess Manag 4:94–104. Samuel T, Agarwal HC, Pillai MKK. 1988. Persistence and binding of DDT and
western Canada. Nature 395:585–588. Brock TCM, Arts GHP, Maltby L, Van den Brink PJ. 2006. Aquatic risks of pesticides,
gamma-HCH in a sandy loam soil under field conditions in Delhi, India. Pestic Sci 22:1–15.
ecological protection goals, and common aims in European Union legislation. Integr Environ Assess Manag 2:e20–e46.
Samuel T, Pillai MKK. 1991. Impact of repeated application on the binding and persistence of [14C]-DDTand [14]-HCH in a tropical soil. Environ Pollut 74:205–
Calumpang SMF. 1996. Exposure of four Filipino farmers to parathion-methyl while spraying string beans. Pestic Sci 46:93–102.
216. Sarkar SK, Bhattacharya BD, Bhattacharya A, Chatterjee M, Alam A, Satpathy KK,
Carbo L, Martins EL, Dores EFGC, Spadotto CA, Weber OLS, De-Lamonica-Freire EM. 2007. Acetamiprid, carbendazim, diuron and thiamethoxam sorption in two
Jonathan MP. 2008. Occurrence, distribution and possible sources of organochlorine pesticide residues in tropical coastal environment of India:
Brazilian tropical soils. J Environ Sci Health B 42:499–507. Catan ˜o HC, Carranza E, Huamanı´ C, Herna´ndez AF. 2008. Plasma cholinesterase
an overview. Environ Int 34:1062–1071. Sethunathan N, Adhya TK, Raghu K. 1982. Microbial degradation of pesticides in
levels and health symptoms in Peruvian farm workers exposed to organophosphate pesticides. Arch Environ Contam Toxicol 55:153–159.
tropical soils. In: Matsumura F, Krishna-Murti CR, editors. Biodegradation of pesticides. New York (NY): Plenum. p 91–115.
Chai L-K, Mohd-Tahir N, Hansen HCB. 2009. Dissipation of acephate, chlorpyrifos, cypermethrin and their metabolites in a humid-tropical vegetable production
Sharmila M, Ramanand K, Adhya TK, Sethunathan N. 1988. Temperature and the persistence of methyl parathion in a flooded soil. Soil Biol Biochem 20:399–
system. Pest Manag Sci 65:189–196. Daam MA, Van den Brink PJ. 2010. Implications of differences between temperate
401. Shen L, Wania F, Lei YD, Teixeira C, Muir DCG, Bidleman TF. 2005. Atmospheric
and tropical freshwater ecosystems for the ecological risk assessment of pesticides. Ecotoxicology 19:24–37.
distribution and long-range transport behavior of organochlorine pesticides in North America. Environ Sci Technol 39:409–420.
Daly GL, Lei YD, Teixeira C, Muir DCG, Castilo LE, Wania F. 2007. Accumulation of current-use pesticides in neotropical montane forests. Environ Sci Technol
Simonich SL, Hites RA. 1995. Global distribution of persistent organochlorine compounds. Science 269:1851–1854.
41:1118–1123. Dimitri C, Greene C. 2002. Organic food industry taps growing American market.
Singh NC, Dasgupta TP, Roberts EV, Mansingh A. 1991. Dynamics of pesticides in tropical conditions. 1. Kinetic studies of volatilization, hydrolysis of dieldrin and
Agric Outlook October 2002: 4–7. Ecobichon DJ. 2001. Pesticide use in developing countries. Toxicology 160:27–33.
a- and b- endosulfan. J Agric Food Chem 39:575–579. Tan SH, Nursiah Aros MT, David S. 1998. Pesticide use and control in Malaysia. In:
Ferri MVW, Gomes J, Dick DP, Souza RF, de Vidal RA. 2005. Sorption of acetochlor
Kennedy IR, Skerritt JH, Johnson GI, Highley E, editors. Seeking agricultural
herbicide by soil samples, humic acids and humin from an Argisol under no-till and conventional tillage systems. Revista Brasileira de Ciencia do Solo 29:705–
produce free of pesticide residues. Canberra (AU): Australian Centre for International Agricultural Research (ACIAR). p 59–63.
714. Heong KL, Escalada MM, Mai V. 1994. An analysis of insecticide use in rice: case
Tanabe S, Ramesh A, Sakashita D, Iwata H, Tatsukawa R, Mohan D, Subramanian AN. 1991. Fate of HCH (BHC) in a tropical paddy field: application test in South
studies in the Philippines and Vietnam. Int J Pest Manag 40:173–178. Ismail BS, Kalithasan K. 2004. Adsorption, desorption, and mobility of permethrin
India. Int J Environ Anal Chem 45:45–53. Tao S, Yang Y, Cao HY, Liu WX, Coveney RM, Xu FL, Cao J, Li BG, Wang XJ, Hu JY., et
in Malaysian soils. J Environ Sci Health B 39:419–429. Kamrin MA. 1997. Pesticide profiles: toxicity, environmental impact and fate. Boca
al. 2006. Modeling the dynamic changes in concentrations of gammahexachlorocyclohexane (gamma-HCH) in Tianjin region from 1953 to 2020.
Raton (FL): Lewis. Kwok KWH, Leung KMY, Lui GS, Chu VKH, Lam PKS, Morritt D, Maltby L, Brock TCM,
Environ Pollut 139:183–193. Tomlin CDS. 2009. The e-Pesticide Manual. In: Tomlin CDS, editor. V4.2 ed. Surrey
Van den Brink PJ, Warne MSJ, et al. 2007. Comparison of tropical and temperate freshwater animal species’ acute sensitivities to chemicals:
(UK): British Crop Protection Council. Traas TP, Janse JH, Van den Brink PJ, Brock TCM, Aldenberg T. 2004. A freshwater
implications for deriving safe extrapolation factors. Integr Environ Assess Manag 3:49–67.
food web model for the combined effects of nutrients and insecticide stress and subsequent recovery. Environ Toxicol Chem 23:521–529.
Lugo AE, Sanchez MJ, Brown S. 1986. Land use and organic carbon content of some subtropical soils. Plant Soil 96:185–196.
Van den Brink PJ, Roelsma J, van Nes EH, Scheffer M, Brock TCM. 2002. PERPEST model, a case-based reasoning approach to predict ecological risks of
Mackay D. 2001. Multimedia environmental models: The fugacity approach. Boca Raton (FL): Lewis. 261 p.
pesticides. Environ Toxicol Chem 21:2500–2506. Weber J, Halsall CJ, Muir D, Teixeira C, Small J, Solomon K, Hermanson M, Hung H,
Mackay D, Webster E. 2007. Simple fugacity models of off-site exposure to agrochemicals. In: Kennedy IR, Solomon KR, Gee SJ, Crossan AN, Wang S, Sa´nchez-Bayo F, editors. Rational environmental management of
Bidleman T. 2009. Endosulfan, a global pesticide: A review of its fate in the environment and occurrence in the Arctic. Sci Total Environ 408:2966–2984.
agrochemicals. Washington (DC): American Chemical Society. p 14–36.
Wilson C, Tisdell C. 2001. Why farmers continue to use pesticides despite environmental, health and sustainability costs. Ecol Econ 39:449–462.
Nair A, Samuel T, Pillai MKK. 1992. Behaviour of DDT in 3 soils exposed to solar radiations under different conditions. Pestic Sci 34:333–340.
Yen J-H, Sheu W-S, Wang Y-S. 2003. Dissipation of the herbicide oxyfluorfen in subtropical soils and its potential to contaminate groundwater. Ecotoxicol
Pimentel D, Hepperly P, Hanson J, Douds D, Seidel R. 2005. Environmental, energetic, and economic comparisons of organic and conventional farming
Environ Saf 54:151–156. Zhou J, Jin S. 2009. Safety of vegetables and the use of pesticides by farmers in
systems. BioScience 55:573–582.
China: Evidence from Zhejiang province. Food Control 20:1043–1048.
