Monitoring and risk assessment of pesticides in irrigation systems in Debra Zeit, Ethiopia.

Chemosphere 161 (2016) 280e291

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Monitoring and risk assessment of pesticides in irrigation systems in Debra Zeit, Ethiopia. Berhan M. Teklu a, b, Paulien I. Adriaanse c, Paul J. Van den Brink a, c, * a

Department of Aquatic Ecology and Water Quality Management, Wageningen University, Wageningen University and Research centre, P.O. Box 47, 6700 AA Wageningen, The Netherlands b The College of Natural Sciences, University of Addis Ababa, 4 Kilo campus, Addis Ababa, Ethiopia c Alterra, Wageningen University and Research centre, P.O. Box 47, 6700 AA Wageningen, The Netherlands

h i g h l i g h t s
Some of the 18 target organochlorine pesticides (OCPs) may pose a risk for aquatic organisms.
One OCPs pose high chronic risks to humans when surface water is used as drinking water.
Four pesticides presently used by small-scale farmers pose high risks to aquatic organisms.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 March 2016 Received in revised form 23 June 2016 Accepted 8 July 2016 Available online 18 July 2016

Since Ethiopia is going through a rapid transformation of its agricultural sector, we assessed the human health and environmental risks due to the past use of organochlorine pesticides (OCPs) as well as the risks of the current pesticide use by farmers. A monitoring programme and risk assessment was carried out for the Wedecha-Belbela irrigation system in the Debra Zeit area. The Wedecha and Belbela rivers and adjacent temporary ponds were sampled and examined for the presence of OCPs between August and October 2014, while data on the current pesticide use by small- and large-scale farmers was collected by interviews. The usage patterns were evaluated for risks of using the river or temporary ponds as source of drinking water and for risks for the aquatic ecosystems in the river and ponds with the aid of the PRIMET_Registration_Ethiopa_1.1 model. The samples were collected in five sampling periods, and results indicate that most of the 18 target OCPs were not detected above the detection limit, while g-chlordane may pose chronic risks when surface water is used as drinking water. Endosulfan and heptachlor pose risks to aquatic organisms at second-tier level, while for heptachlor-epoxide B, gchlordane and b-BHC only risks could be determined at the first tier due to a lack of data. For all nine pesticides used by small-scale farmers the calculated acute risks to humans were low. Second tier risk assessment for the aquatic ecosystem indicated that lambda-cyhalothrin, endosulfan, profenofos, and diazinon may pose high risks. © 2016 Elsevier Ltd. All rights reserved.

Handling Editor: Prof. A. Gies Keywords: OCPs Chemical monitoring Risk assessment PRIMET_Registration_Ethiopa_1.1

1. Introduction Due to the ongoing agricultural transformation in Ethiopia, the impact of pesticides on human health and the environment has recently become a major concern (Teklu et al., 2015). In such a dynamic era, it is important to evaluate the risks caused by the legacy use of hazardous substances like organochlorine pesticides (OCPs), as well as assessing the risks of current pesticide use to * Corresponding author. Department of Aquatic Ecology and Water Quality Management, Wageningen University, Wageningen University and Research Centre, P.O. Box 47, 6700 AA Wageningen, The Netherlands. E-mail address: [email protected] (P.J. Van den Brink). http://dx.doi.org/10.1016/j.chemosphere.2016.07.031 0045-6535/© 2016 Elsevier Ltd. All rights reserved.

human health and the environment in locations with past and current use of pesticides, like the Wedecha-Belbela irrigation system in Ethiopia. The use of OCPs has been banned or partially restricted in developed nations, but they are still being used for agricultural and public health purposes in developing countries, including for the control of agricultural pests as well as mosquitoes (Westbrom et al., 2008; Safford and Jones, 1997; Wei et al., 2007). Although most OCPs are banned from Ethiopian agriculture following the enforcement of the Stockholm convention in 2004 (Fiedler et al., 2013), DDT has continued to be used for indoor spraying purposes in malaria control. Illegal use to control agricultural pests has, however, been reported on various occasions, and endosulfan is widely used for insect pest control in vegetables,

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

even though it is registered only for controlling pests in cotton (Mengistie et al., 2016). It is therefore important to evaluate the risks of current environmental concentrations of OCPs, as they are among the serious pollutants of global concern (Tenabe et al., 1994). They are known for their environmental persistence, their ability to bioaccumulate and biomagnify in the food chain, and their chronic toxicity to wildlife and humans. Hence, OCPs are categorized among the POPs (persistent organic pollutants), an example of which is DDT, which stays in the environment for a long time (4e30 years). Other chlorinated pesticides share this persistence due to their resistance to biochemical degradation once released into the environment (Elvira et al., 2011; Jones and de Voogt, 1999). The impacts of OCPs on human and animal health include failure of the reproductive system and increased cancer risk (Perry et al., 2016; Bouman, 2004), immune system malfunction (Repetto and Baliga, 1996), endocrine disruption (Mnif et al., 2011), pancreatic cancers (Andreotti et al., 2009) and breast cancers (Olaya-Contrras et al., 1998). Little work has been done on monitoring the residues of OCPs in surface waters in Ethiopia, and most of the studies have concentrated on evaluating the OCP and heavy metals residue levels in big Rift Valley lakes like Ziway and Hawasa, focusing on quantifying the residue levels in fish (Abayneh et al., 2003; Yohannes et al., 2013) and nutrients and pesticides in the water (Teklu et al., 2016a). Some other studies in Africa examined the status of OCPs in freshwater systems, and detected a number of OCPs in water samples (Adeyemi et al., 2011). The current use of pesticides in Ethiopia has been surveyed at local and national level (Mengistie et al., 2016). National level (2000e2010) data showed that the average import of pesticides has now grown to over 2400 tonnes per annum, with an overall increasing trend and with herbicides being used most (www.prrpethiopia.org). Although pesticide use surveys among small-scale farmers have been performed in the past in Ethiopia (Mengistie et al., 2016; PAN-UK, 2006), so far no work has been done to quantify the risks posed to surface water organisms and humans using surface water as a source of drinking water from the actual pesticide use patterns of farmers cultivating their treated crops near small rivers, irrigation cannels and/or small temporary ponds. Objectives of the present study were to (i) assess the current OCP residue levels in the Wedecha-Belbela irrigation system, (ii) quantify the risks to humans and aquatic organisms posed by the detected OCP concentrations, (iii) quantify the risks to humans and aquatic organisms posed by the actual pesticide use by small-scale farmers in the Wedecha-Belbela irrigation system, and (iv) compare actually measured concentrations of endosulfan with predicted environmental concentration (PEC) values obtained with the PRIMET_Registration_Ethiopia_1.1 model (Wipfler et al., 2014), being the only pesticide for which both measured as predicted concentration data exists. 2. Materials and methods 2.1. The study area The study area is located in the Oromia region, Debra Zeit, some 55 km south of the capital Addis Ababa of Ethiopia (Fig. 1). Because of Ethiopia’s location near the equator, elevation has a very strong influence on temperature and, to a lesser extent, on rainfall. The Wedecha-Belbela irrigation system is situated in the highlands with altitudes ranging from 1895 m at the downstream end of the irrigation system to 2437 m in the upstream parts. It has annual average minimum and maximum temperatures of 10.5  C and 25.4  C, respectively, and an average annual rainfall of 815 mm. The major growing season in the area is associated with the long rains,

281

occurring from June to December (called ‘Kiremt’), while the short rains occur from March to May (called ‘Belg’). The irrigation system is dominated by small-scale farmers in the upstream and middle sections, while small-scale and a few medium-scale vegetable farms and commercial flower greenhouses concentrate in the downstream part of the system (Michael and Seleshi, 2007). Farms in the area are mostly 1e3 ha in size. During the main production period of June to December, the crops receive both rainfall and irrigation water. Major crops in the area include vegetables (cabbage, tomato and onion), while only a few farmers grow cereals and flowers. 2.2. Sample and data collection and laboratory analysis of OCPs Water samples were collected from three river sampling sites in the Wedecha and Belbela rivers, and three temporary ponds, representing the upstream, midstream and downstream parts of the irrigation system. The river sampling sites were at elevations of 2190, 1935 and 1879 m, while the pond sampling sites were situated at 2228, 1943 and 1882 m (Fig. 1). Locations were chosen based on the presence of a small- or large-scale farming activity adjacent to the river/canal or pond sampling sites. Samples were collected during five sampling periods, encompassing the crop development stages as well as seasonal variations in the study area. A total of 15 samples were taken by sampling three river sites four times and one sample from each of the three additional temporary ponds, which were only formed at the end of rainy season. A hand grab method was used to collect the water samples in 4 L amber glass bottles. Collected water was thoroughly mixed in a bucket and transferred to the 4 L amber bottles, supplemented with some drops of sodium thiosulfate as a preservative and filled up to the seal, leaving no space for air bubbles to be included. Properly sealed samples were taken to the quality monitoring and pesticide testing laboratory of the Plant Health Regulatory Directorate (PHRD) of the Ethiopian Ministry of Agriculture and stored at 4  C (Forrest, 2000). Together with the water sample collection, the physiological parameters of the water, including pH, temperature, dissolved oxygen (DO) and electrical conductivity (EC) were measured using a Handy Polaris (OxyGuard, USA) and a WTW multi 340i (WTW, USA) meter. The total dissolved solids (TDS) and total suspended solids (TSS) of the water were measured at a laboratory (JIJE laboglass plc; http://jijelaboglassplc.com/), which is ISO/IEC 17025:2005 accredited by the Ethiopian National Accreditation Office. TSS was measured by filtering (mesh size 1.12 mm) and drying in an oven at 103e105  C (Clesceri et al., 1999), while TDS was measured using a TDS meter (TDS_EZ water quality tester HMDigital USA). Samples were analysed at the accredited laboratory of the PHRD for pesticide residues. For all 15 samples, quality control was implemented, to keep the laboratory assessment free from unnecessary interference and cross-contamination (Adeyemi et al., 2011; Kashyap et al., 2005). A simple liquid/liquid extraction method followed by a solid phase extraction florisil clean-up was _ used (Zwir-Ferenc and Biziuk, 2006; De Koning et al., 2009). Analysis was performed by screening for the 17 OCPs using a gas chromatograph (GC) with an electron capture detector (ECD) (Agilent Technologies Inc., Palo Alto, CA, USA). OCP standards (99.95%) for 17 target chemicals were purchased from Restek Ltd, New Haven, CT, USA. A column (Stx-clpesticides 30 m  250 mm x 0.25 mm) was used to separate and analyse extracted samples, with 1 mL volume being injected automatically. GC temperature was programmed at 120, 225 and 280  C, at an injector and detector temperature of 230  C. Helium was used as a carrier gas at a flow rate of 1 mL/min, while nitrogen was used as a makeup gas at a

282

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

Figure 1. Map of the study area within Ethiopia.

flow rate of 60 mL/min (Chauhan et al., 2014; Adeyemi et al., 2011). 2.3. Data collection and calculation of PECs for nine pesticides used by small-scale farmers Data on cultivated crops and current pesticide use was collected by interviewing 15 representative small-scale farmers from the upstream, midstream and downstream parts of the river. In addition, seven large-scale farms located along the downstream part of the river were also surveyed. Farmers were considered to be representative when they cultivated treated crops adjacent to the river or a large irrigation canal. Data for the small-scale farmers was verified by the district agricultural experts in the area. In addition, some basic information about resource ownership and awareness of pesticide application was collected using a simple questionnaire (Annex 1). In total 16 pesticides, reported with their common names, were used by small-scale farmers. Pesticides with similar active ingredients but different common names are considered as one pesticide in the remainder of this paper. The original number of 16 was further reduced to nine by selecting only those pesticides for which physicochemical data were available (Annex 2). Observations in the field and discussions with the farmers revealed that cabbage and tomato, grown during the long rainy season ‘Kiremt’, are the major crops so these were selected for the scenario development for the risk assessment. When different application schemes were recorded for one pesticide the application scheme with the highest application rate was selected. So, the highest actual application rate and associated frequency of use were used in the PEC calculations (Annex 3).

For the nine pesticides, used on cabbage or tomato, the PECs were calculated with the aid of the PRIMET_Registration_Ethiopia_1.1 software model (Wipfler et al., 2014; Teklu et al., 2015). This model is recently developed to support pesticide registration in Ethiopia and is based on risk assessment procedures similar to procedures applied at EU level and in the USA. Calculation of the exposure concentrations is performed with the PRZM (for runoff) and TOXSWA (for fate in surface water) models, also used in the EU and USA for registration, using scenarios specifically tailored to Ethiopia. So, these scenarios contain Ethiopian data on soils, crops and cropping calendars, elevation, meteorology (e.g. rain), land use management practices and surface water bodies (small streams and ponds). They are used to assess risks from drinking water obtained from surface water, representing ‘realistic worst case’ situations and intend to protect man not in the average situation, but in more vulnerable situations. For risks from drinking water the scenarios were selected in such a way that they are protective for 99% of all possible agro-environmental conditions in time and space across Ethiopia. The same scenarios are used to assess the risks for the aquatic ecosystem, but then the 90th percentile PEC value is selected by selecting a lower PEC value from the 33 year series of yearly maximum concentration in the small streams or pond i.e. lowering the probability of the PEC value in time (Adriaanse et al., 2015). In this paper we selected the following two scenarios: (i) small streams above 1500 m elevation, represented by grid 191 with a long term annual average precipitation of 2581 mm and (ii) temporary ponds between 1500 and 2000 m, represented by grid 217 with a long term annual precipitation of 2779 mm (see Adriaanse et al., 2015 for more details). For these two scenarios

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

PEC values were calculated for cabbage or tomato, cultivated during the long rainy season of ‘Kiremt’ and the actual use patterns of the nine pesticides, used by the small-scale farmers. Since the entry routes of pesticides into surface water from greenhouse production systems are very different from those from crops grown in fields outside, a greenhouse scenario is not included in the PRIMET_Registration_Ethiopia_1.1 model. So it was not possible to calculate the PEC values for the surrounding surface waters regarding the usage data of the large-scale greenhouse farmers (Annex 4). The physicochemical data for the nine pesticides was obtained from the Footprint pesticides properties database (Lewis et al., 2016). The data include molar mass, saturated vapour pressure at 20  C, water solubility, half-life of transformation in soil (DT50,soil), half-life of transformation in water (DT50,water), dissociation constant (pKa), and coefficient of sorption to soil based on organic carbon content (Koc) (Annex 2).

283

concentrations determined using data from the US-EPA database (www.epa.gov/ecotox). The ETX 2.0. software (Van Vlaardingen et al., 2003) was used to calculate the HC5, data was selected as discussed by Maltby et al. (2005). In the second tier the chronic HC5 values were used as NEC values (Teklu et al., 2015) When no chronic HC5 value was available, the acute HC5 value was divided by an assessment factor of 10 to calculate the NEC. Risk category intervals were as follows; 0e1 equals low risk for algae, Daphnia and fish; 1e100 medium risk for algae and Daphnia; 1e10, medium risk for fish; >100 high risk for Daphnia and algae and >10 high risk for fish (Teklu et al., 2015). All risk calculations for the detected OCPs were performed by hand, using the highest concentration detected.

ETRwater org ¼

OCPconcentration NECorgðacute or chronicÞ

(3)

with: 2.4. Risk assessment 2.4.1. Acute and chronic risks to humans and aquatic organisms from detected OCPs For the OCPs we detected, the risks to humans were determined using the highest detected concentrations. Acute and chronic health risks were calculated using acute reference dose (ARfD) and acceptable daily intake (ADI) values, respectively. Estimated shortterm intake ratio (ESTI) and the internationally estimated daily intake ratio (IEDI) values were determined to indicate acute and chronic risks to humans from using surface water as a source of drinking water. Background calculation formulas are given in Eqs. (1)-(2) and in other publications (Wipfler et al., 2014; Adriaanse et al., 2015; Deneer et al., 2014).

ESTI ¼

LP dw  OCPconcentration 100% ARfD  BW

DI  OCPconcentration 100% ADI  Fdw  BW

NECfish

acute

NECDaphnia NECalgae NECfish

¼ 0.01 * LC50 of fish (mg/L)

acute

acute

chronic

NECDaphnia

¼ 0.01 * EC50 of Daphnia (mg/L)

¼ NECalgae

chronic

¼ 0.1 * EC50 of algae (mg/L)

¼ 0.1 * NOEC of fish (mg/L)

chronic

¼ 0.1 * NOEC of Daphnia (mg/L)

(4) (5) (6) (7) (8)

(1)

with: ESTI ¼ estimated short-term intake ratio (%); LP_dw ¼ Large portion of drinking water (L/d); OCPconcentration ¼ measured OCP concentrations (mg/L); ARfD ¼ acute reference dose (mg/kg BW*d) and BW ¼ body weight (kg). The ESTI was determined using a body weight of 60 kg and assuming a portion of drinking water (LP_dw) of 2 L per day (WHO, 2010).

IEDI ¼

NECorg(acute or chronic) ¼ acute or chronic no-effect concentration for the respective aquatic organism (fish, Daphnia or algae) (mg/ L).where:

(2)

with: IEDI ¼ internationally estimated daily intake ratio, expressed as % of the total acceptable intake of the pesticide during a lifetime (%); DI ¼ daily intake (L/d); ADI ¼ acceptable daily intake, expressed in mg pesticide per kg BW per day (mg/kgBW*d) and Fdw ¼ fraction of ADI allocated to drinking water (). BW was set at 60 kg, DI at 2 L/d and the fraction of ADI allocated to drinking water at 0.1 for Ethiopia. Similarly, both acute and chronic risks to aquatic organisms were determined using the measured OCP concentrations. A firsttier risk assessment was performed for aquatic organisms by calculating the exposure-to-toxicity ratio (ETR) by estimating acute and chronic no-effect concentrations (NECs) from the corresponding EC50, LC50 or NOEC values for three representative aquatic organisms (fish, Daphnia, and algae) (Eqs. (3)e(8)) (Lewis et al., 2016) (Annex 5). A second-tier risk assessment was performed for ETR values > 1 using HC5 (hazardous concentration protective of 95% of the population) values from acute and chronic species sensitivity distributions (SSDs), based on acute EC50’s and chronic NOECs, respectively. SSDs were calculated and HC5

2.4.2. Acute risks to humans and aquatic organisms for nine pesticides used by small-scale farmers Nine pesticides in current use by the small-scale farmers were evaluated for the risks they may pose to humans and aquatic organisms. The PRIMET_Registration_Ethiopia_1.1 model (Wipfler et al., 2014) was used to perform a lower tier risk assessment with exposure concentrations based on the two selected exposure scenarios and threshold levels of effects for the aquatic ecosystem and human health using the equations as provided in 2.4.1. (Teklu et al., 2015; Annex 6). Second-tier risks were calculated using the same PRIMET_Registration_Ethiopa_1.1 model calculated PEC values, but using higher-tier NEC values, derived from SSDs based on acute EC50 values for arthropods from which the HC5 values were derived. Data from the US-EPA database (www.epa.gov/ecotox) were included and the ETX 2.0. software (Van Vlaardingen et al., 2003) was used to calculate the HC5s. Data was selected as discussed by Maltby et al. (2005). 3. Results and discussion 3.1. Measured water properties and OCP concentrations Values of physicochemical parameters measured in this study are given in Annex 7. No correlation was found between the physicochemical parameters and the OCP residues detected in this study. EC and pH were found to be within the maximum permissible limits set by the WHO (WHO, 2010; Teklu et al., 2016a) while the value for the total suspended solids in the samples was at a maximum during the period of land preparation and sowing,

284

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

indicating high erosion at the onset of the rainy season in the area. Average pH and temperature of the river and pond sites sampled in this study were 7.62 and 20.4  C, respectively. This justified the use of the DT50,water for aqueous hydrolysis at pH 7 for the PRIMET_Registration_Ethiopia_1.1 model calculations (Annex 2). In this study OCPs were not detected or below LOQ (Level of Quantification) in most of the 15 samples (Table 1). The results on OCP residues in the river samples indicated that the highest concentrations detected were those of heptachlor epoxide B (0.115 mg/ L) and a-chlordane (0.192 mg/L). High concentrations of g-chlordane (10.1 mg/L) and b-BHC (2.72 mg/L) were detected in the downstream temporary pond samples. These results are in line with the model predictions (Teklu et al., 2015), which indicated higher PEC values for temporary ponds than for streams. A similar study done in Tanzania detected no heptachlor epoxide, this pesticide is banned from use both in Ethiopia and Tanzania (Hellar, 2011; PHRD, 2015). In our study, we detected no DDT or breakdown products of DDT, whereas DDT and its breakdown products were detected in more than 90% of the samples in water from the four rivers investigated in Tanzania (Hellar, 2011). In a study from Nigeria, most of the target OCPs analysed were reported not to be detected above the detection limits in the majority of the river water samples, as was also found in the present study, but with higher reported concentrations of 1.51, 0.11, 0.13 and 0.13 mg/L for dieldrin, p,p0 -DDE, cischlordane and a-endosulfan, respectively (Okoya et al., 2013). 3.2. Pesticide use and handling by small-scale farmers and PEC values for the nine pesticides The results of the survey on pesticide use and handling by smallscale farmers show that 69% of the respondents did not know that spray drift and run-off can be a possible source of pesticide pollution to the surrounding surface waters. The majority (91%) did not maintain a safe distance from the river (or canal) while spraying pesticides, and 71% of the farmers mixed pesticides and washed pesticide containers near the river/canal. Eighty-two percent of all farmers had increased the frequency and dosage of pesticide applications in the past five years, and 67% of them mentioned pest resistance as the major reason for the increment, while 67% indicated they had recently been trained in pesticide application methods (n ¼ 45) (Annex 1). Almost all pesticide types were used by the small-scale farmers with increasing amounts and frequencies of application, except for the herbicides pyroxsulam and glyphosate, for which no change was found. Flower farms data showed that they all followed the prescribed dosage and frequency of application (Annex 4). These results are in line with similar other studies conducted in Ethiopia. At local level, a survey of pesticide use conducted by FAO in the Dugda and Ada districts found that small-scale farmers were using higher application frequencies than commercial farmers in the same area. In addition, mismanagement in handling, transporting, storing and applying pesticides has been reported (FAO, 2012 unpublished). A similar study indicated that

farmers applied pesticides in violation of standard recommendations, used unsafe storage facilities, ignored risks and safety instructions, did not use protective devices when applying pesticides and disposed of containers unsafely. The same study found that 74% of the farmers mixed their pesticides close to a river and that 96% did not know that pesticides can cause damage to water bodies, while 88% indicated an increase in their pesticide use in the past 5 years (Mengistie et al., 2016). According to a study undertaken by the Pesticide Action Network-United Kingdom (PAN-UK) on pesticide use and management by small-scale farmers around the Central Rift Valley, Ethiopia, 97% of the respondents reported using pesticides once or twice a year, and about 91% of them prepared their pesticides close to water sources used by local people for drinking, cooking and other household purposes, while 61% washed their pesticides sprayers and other equipment on the farm field (PAN-UK, 2006). While we found that 82% of the farmers increased their pesticide use in the last 5 years, Macharia (2015) found an increase of 47% in pesticide application rate per hectare and per season for small-scale vegetable producers in Kenya. Although pest resistance has been reported as the main reason for the increase in pesticide use by small-scale farmers, the lack of awareness regarding pesticide handling issues could also explain the difference in pesticide misuse between small-scale and large-scale farmers (Mengistie et al., 2016). Usage data by commercial flower farms in the area and a general description of the current status of liquid waste management by these farms are provided in Annex 4 and 8. PEC values for the nine pesticides used on cabbage and tomato by the small-scale farmers are needed for the risk assessment for drinking water by humans and the aquatic ecosystem (Tables 2 and 3). They range from 0.00014 mg/L for deltamethrin (stream), an extremely low value caused by its extremely high sorption coefficient resulting all mass to be transferred from the water into the sediment, to 30.8 mg/L (Table 2) or 28.0 mg/L (Table 3) for diazinon (pond), being the highest value caused by the combination of a relatively high application rate of 1.5 kg/ha and relatively low sorption coefficient Koc of 609 L/kg (Annex 2 and 3). The PEC values for drinking water (Table 2) are somewhat higher than the corresponding PEC values for the aquatic ecosystem (Table 3), although the scenarios, including the crop, and all input (pesticide and properties, application pattern) are identical. The reason is that the former PEC values represent a 99th percentile worst-case situation, while the latter PEC values represent a less strict, 90th percentile worst-case situation. The 90th percentile exposure concentrations correspond to the current risk assessment procedure for the aquatic ecosystem in the EU. The 99th percentile exposure concentrations for the risk assessment for drinking water are clearly higher than the 0.1 mg/L cut-off value used in the EU (EC, 1998), thus demonstrating that the toxicological approach for assessing risks for drinking water, followed in this paper, generally is less strict than the use of the 0.1 mg/L cut-off value. Exceptions are the pyrethroids that have high sorption coefficients, Koc values.

Table 1 Concentration of OCPs measured (mg/L) in rivers and temporary ponds in 2014. Upstream river OCP Dieldrin Heptachlor Epoxide B Endosulfan a-Chlordane Heptachlor b-BHC Aldrin g-chlordane

LOQ 0.06 0.07 0.03 0.09 0.05 0.03 0.05 0.06

11-07 ND ND ND ND ND ND ND ND

11-08 ND 0.115 ND 0.192 ND ND ND ND

Midstream river 11-09 ND ND ND ND 0.009 ND ND ND

11-10 0.010 ND 0.023 ND ND 0.029 ND ND

11-07 0.027 ND ND ND ND ND ND ND

11-08 ND 0.100 ND ND 0.007 ND ND ND

Downstream river 11-09 ND 0.094 ND ND 0.007 ND ND ND

11-10 0.012 ND ND ND 0.024 ND ND ND

11-07 ND 0.015 0.031 ND ND ND ND ND

11-08 ND 0.100 ND ND ND ND ND ND

11-09 ND ND ND ND ND ND ND ND

11-10 ND ND ND ND ND ND ND ND

USP

MSP

DSP

20-10 ND ND ND ND ND 2.27 ND 5.02

20-10 0.012 ND ND 0.004 ND ND 0.001 ND

20-10 ND ND ND ND ND 2.72 ND 10.1

Note: USP ¼ Upstream pond; MSP ¼ Midstream pond; DSP ¼ Downstream pond; ND ¼ Not detected, LOQ ¼ Limit Of Quantification: values under the LOQ are indicative values, above the LOD (¼ Limit Of Detection)

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291 Table 2 PEC values and ESTI values (%) indicating acute toxic effects to humans for the nine pesticides used by the small-scale farmers (Calculated by PRIMET_Registration_Ethiopia_1.1)

285

available for a-chlordane, heptachlor epoxide B, g-chlordane and bBHC (Table 5). 3.4. Acute risks to humans and aquatic organisms from nine pesticides used by small-scale farmers

Active ingredient

ARfD (mg/kgBW*d)

PEC (mg/L)

ESTI

Stream

Pond

Stream

Pond

Lambda-cyhalothrin Profenofos Malathion Endosulfan Deltamethrin Pyroxsulam Propiconazole Glyphosate Diazinon

0.005 1 0.3 0.02 0.01 0.9* 0.3 0.3* 0.025

0.0251 16.5 0.32 3.54 0.00014 3.37 13.5 17.7 25.1

0.053 14.9 0.33 6.49 0.00091 1.32 20.6 21.6 30.8

<0.1 <0.1 <0.1 0.006 <0.1 <0.1 <0.1 <0.1 0.03

<0.1 <0.1 <0.1 0.011 <0.1 <0.1 <0.1 <0.1 0.0

* ADI (no ARfD available)

3.3. Acute and chronic risks to humans and aquatic organisms from detected OCPs Our risk assessment results for humans indicate low acute risks for humans for the two pesticides dieldrin and endosulfan with their IESTI values lower than 1% (Table 4). Similarly, all the OCPs we detected pose a low chronic risk to humans, except g-chlordane, whose IEDI values of 334 and 673% indicates a high risk when surface water is used as drinking water. These results are in line with those of Teklu et al. (2015) who found a low acute risk to humans for all evaluated pesticides when surface water is used as drinking water. Although beta-hexachlorocyclohexane (b-BHC) detected at 2.27 and 2.72 mg/L and g-chlordane detected at 5.02 and 10.1 mg/L (Table 1) exceed by a factor 22 to 100 the acceptable concentration of 0.1 mg/L in the EU (EC, 1998), this is not in contradiction with the conclusion of low human risks, as the EU standards have no toxicological basis. Also, the four heptachlor epoxide B concentrations detected above the LOQ of between 0.094 and 0.115 mg/L are up to a factor of nearly 4 higher than the acceptable concentration of 0.03 mg/L in the EU and for a-chlordane the detected concentration of 0.192 mg/L is a factor 2 higher than the acceptable concentration of 0.1 mg/L in the EU (EC 1998). The first-tier risk assessment for the aquatic ecosystem results indicate many medium and two high risks that pesticides may pose to fish (Table 5). For fish, a high chronic ETR of 3086 was calculated for endosulfan for one river sampling site, as well as a high acute ETR for g-chlordane (11) and b-BCH (9.1) for a temporary pond. Two medium acute risks were calculated for b-BHC for algae (2.3 and 2.7) in ponds, as well as two medium risks for g-chlordane for Daphnia (chronic, ETR ¼ 1.4 and acute, ETR ¼ 1.7), also in ponds (Table 5). The second-tier assessment, which is based on SSDs using acute or chronic toxicity data for arthropods analogous to Maltby et al. (2005), revealed chronic risks to aquatic organisms for endosulfan and heptachlor (Table 5). No threshold levels were

All nine pesticides used by the small-scale farmers pose a low risk when the surface water is used as drinking water (Table 2). This is in line with earlier model-based risk assessments by Teklu et al. (2015). The highest ESTI values were found for diazinon, 0.03% for the small stream scenario and 0.04% for the pond scenario, which are both far below the 100% representing the total acceptable shortterm intake. Results from first-tier risk assessments for the aquatic ecosystem indicate that lambda-cyhalothrin, endosulfan, and profenofos are associated with high risk for fish in both the pond as the stream scenario, while malathion poses medium risks (Table 3). Diazinon poses a high risk for Daphnia, in both the river and pond sites in the area, while lambda-cyhalothrin, profenofos and malathion poses medium risks. Only propiconazole posed a medium risk to algae (Table 3). Similar work done using model-based first-tier risk assessment in South Africa found that analysis of the application patterns of aldicarb, methomyl, linuron, bromoxynil, carbaryl, dichlorvos, parathion, and two pyrethroids, cypermethrin and deltamethrin, indicated a possible risk at their respective predicted environmental concentrations (Ansara-Ross et al., 2008). These results are not in line with those of the present study, which identified four pesticides with high risks at the first-tier level. Second-tier risk assessment for aquatic organisms indicated that lambda-cyhalothrin, profenofos, endosulfan, and diazinon posed medium to high risks for both the river and pond sites, while propiconazole and malathion were found to pose a low risk (Table 6). Slightly higher second-tier risks were observed in this study for lambda-cyhalothrin and profenofos, compared to their first-tier estimates, while risks of endosulfan and diazinon were considerably lower in the second-tier assessment (Table 6). The latter is in line with the logic of the tiered approach in the risk assessment for registration purposes, where higher tiers should be more realistic and less conservative than lower tiers (Brock et al., 2006; Wheeler et al., 2002). The data used in this study for assessing SSDs mostly relies on non-African species, yet it is believed that they represent reliable results as some works done to show difference in sensitivity between tropical and temperate species resulted indicate similarity in sensitivity with recommendations the possibility of using temperate data until much tropical (African) data is compiled (Teklu et al., 2016b, Rico et al., 2011). Our study was unable to quantify the risks of pesticide use for aquatic ecosystems near greenhouses. The scenario for such systems is different from those included in the PRIMET_Registration_Ethiopia_1.1 model, for instance in that all spray activity

Table 3 PEC values and first-tier ETRs () indicating acute toxic effects to aquatic ecosystems for the nine pesticides used by the small-scale farmers (calculated by PRIMET_Registration_Ethiopia_1.1). Active ingredient

Lambda-cyhalothrin Profenofos Malathion Endosulfan Deltamethrin Pyroxsulam Propiconazole Glyphosate Diazinon

PEC(mg/L)

NEC(mg/L)

ETR fish

ETR Daphnia

ETR algae

Stream

Pond

Fish

Daphnia

Algae

Stream

Pond

Stream

Pond

Stream

Pond

0.024 16.0 0.19 3.41 0.00014 2.78 11.7 16.5 24.0

0.050 11.5 0.26 4.93 0.00086 0.71 15.5 12.4 28.0

0.0021 0.8 0.18 0.02 0.0026 870 26 380 31

0.0036 5 0.007 44 0.0056 1000 102 400 0.01

30 10000 1300 215 910 92.4 9.3 440 640

11.6 19.9 1.04 171 0.054 <0.01 0.449 0.043 0.774

24.0 14.3 1.45 247 0.33 <0.01 0.60 0.033 0.904

6.77 3.19 26.7 0.078 0.025 <0.01 0.11 0.041 2398

14.0 2.29 37.2 0.11 0.15 <0.01 0.15 0.031 2802

<0.01 e <0.01 0.016 <0.01 0.03 1.26 0.038 0.037

<0.01 e <0.01 0.023 <0.01 <0.01 1.66 0.028 0.044

e No output for algae since the toxicity database provides no EC50 value of profenofos for algae.

286

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

Table 4 IEDI/ESTI values (%) indicating chronic/acute toxic effects to humans for the detected OCPs (Calculated according to Eqs. (1) and (2)). When only one value is present, it represents the IEDI. B-BCH is not included since no ARfD or ADI was available. OCPs detected

Human risk (IEDI/ESTI%) River

a-Chlordane g-Chlordane Dieldrin Endosulfan Heptachlor- Epoxide B Heptachlor

Pond

Upstream

Middle part

Downstream

Upstream

Middle part

Downstream

13 e e e 38 e

e e 8.9/0.03 e 33 31

e e e 0.17/0.007 33 e

e 334 e e e e

e e e e e e

e 673 e e e e

Note: e ¼ no risk quotient since the concentration is below the detection limit. Table 5 First- and second-tier ETRs indicating acute/chronic toxic effects to aquatic ecosystems for the detected OCPs (calculated according to Eqs. (3)e(8)). OCPs detected

First tier risk (ETR)

Second tier risks (SSD)

upstream river or pond

a-Chlordane Dieldrin Endosulfan Heptachlor- Epoxide B Heptachlor g-Chlordane b-BHC

midstream river or pond

downstream river or pond

algae

Daphnia

fish

algae

Daphnia

fish

algae

Daphnia

fish

e e e 0.043 e

0.027a e e 0.27 e 1.4a* 0.45*

0.21 e e 1.6 e 5.6* 7.6*

e < 0.01 e 0.037 0.035 e e

e 0.01 e 0.24 0.22 e e

e 2.2 e 1.4 1.3 e e

e e <0.01 0.037 e e 2.7*

e e <0.01 0.24 e 1.7* 0.54*

e e 3086a 1.4 e 11* 9.1*

2.3*

Highest measured concentration

NECssd

ETRssd

e 0.027 0.031 0.12 0.095 10 2.3

NA 0.046b,c 0.0075b,c /0.0066d NA 0.057b,c NA NA

e 0.59 4.1c /4.7d e 1.7 e e

Note: e¼ no risk quotient since the concentration is below the detection limit; * ¼ risk values for temporary pond; a Chronic risks; NA ¼ not enough data available (see Maltby et al. (2005) for criteria) b NEC based on HC5 value calculated using acute EC50 values divided by 10. c HC5 value based on data for arthropods. d HC5 values based on data for fish and amphibians.

takes place indoors. Most large-scale farmers reported possessing a mechanism for recycling liquid waste from their greenhouses, besides the use of soakaway pits and wetlands for further detoxification of liquid wastes to protect the surrounding soil and surface waters (Annex 8). Future studies need to concentrate on the applicability of these systems and the generation of plausible scenarios for surface water risk assessment of pesticides for Ethiopian greenhouse systems. Moreover, the two grids (191 and 217) from the PRIMET_Registration_Ethiopia_1.1 represent vulnerable exposure situations in Ethiopia (and thus not average situations), i.e. the 99th percentile worst case for drinking water and the 90th percentile worst case for aquatic organisms, of all possible situations in time and space in Ethiopia. Comparison of the elevation, rainfall pattern (i.e. total annual average, period, and if possible number of events >20 mm/d per year) and agricultural crops between the small stream scenario (grid 191) and the pond scenario (grid 217) with those of the Wedecha-Belbela irrigation system showed that the grid scenarios provided a worst case scenario for the irrigation system. 3.5. Comparison of measured endosulfan concentrations and model-predicted PEC values In this study we found endosulfan concentrations of 0.023

(upstream river) 0.031 mg/L (downstream river) (Table 1). This is lower than the levels predicted in a recent model-based study (Teklu et al., 2015), with model predicted PEC-values of 1.3 mg/L and 0.6 mg/L for endosulfan used on maize. These values are also lower than those obtained in the present study, also by the PRIMET_Registration_Ethiopia_1.1, with PECs of 3.14 and 4.93 mg/L for the cabbage and tomato crops in the same stream and pond scenarios (Table 3). There are many explanations for the difference between measured and model predicted concentrations, the most important one, probably being the fundamental difference between river water sampled at more or less arbitrarily times and locations and concentrations purposely modelled to be realistic worst case. The concentrations of PRIMET_Registration_Ethiopia_1.1 represent 99th and 90th percentile concentrations, selected out of 33 yearly maximum concentrations at realistic worst case locations: grids 191 and 217 with their 2581 and 2779 mm of rain, while the grabbed river and pond samples are taken in Debra Zeit area with only 815 mm rain at times not necessarily leading to yearly maximum, e.g. immediately after pesticide applications followed by runoff events, as does happen for the scenario calculations in grids 191 and 217. Also peak concentrations by spray drift deposition are probably not captured by the hand grabbed river samples, while the model selects these peak concentrations. Finally, hydrophobic substances such as endosulfan may create a film on the

Table 6 Second-tier ETRs indicating acute toxic effects to aquatic ecosystems for six pesticides used by the small-scale farmers (calculated by PRIMET_Registration_Ethiopia_1.1). NECs are calculated using SSD including acute toxicity data for arthropods (Maltby et al., 2005). Active ingredient

PEC (mg/L) Stream

PEC (mg/L) Pond

NEC (mg/L)

ETRssd

Lambda-cyhalothrin Profenofos Malathion Endosulfan Propiconazole Diazinon

0.024 15.95 0.19 3.414 11.68 23.98

0.05 11.45 0.26 4.93 15.47 28.02

1.75E-03 2.09E-01 5.50E-01 4.83E-02 4.26Eþ02 5.06E-01

13.7 76.3 0.35 70.7 0.027 47.4

river

ETRssd 28.6 54.8 0.472 102.2 0.036 55.4

pond

283707 2016 1800 11500 10240000 33.22 1086 1424 609 2.00E-04 2.53 3.1 0.83 1.24E-05 1.00E-04 0.056 0.0131 11.97 0.005 28 148 0.32 0.0002 3200 150 10500 60 449.85 373.63 330.36 406.93 505.2 434.35 342.22 169.1 304.35

Vapour pressure at 25  C (mPa)

1000 1000 6.2 20 1000 1000 53.5 1000 138 175 7 0.17 39 26 3.3 90 15.3 9.1 IN IN IN IN IN HB FU HB IN Lambda-cyhalothrin Profenofos Malathion Endosulfan Deltamethrin Pyroxsulam Propiconazole Glyphosate Diazinon

Pyrethroid Organophosphate Organophosphate Organochlorine Pyrethroid Triazolopyrimidine Triazole Phosphonoglycine Organophosphate

Type

Substance group

DT50_Soil (lab at 20  C (d)

DT50 water Aqueous hydrolysis at pH 7(d) at 20  C

Response (%) (n ¼ 45)

Trained on pesticide application methods yes 66.6 no 33.4 Understanding pesticide labelling yes 64.4 no 35.6 Fate of used pesticide containers: thrown into the river, 0 used as drinking water container 8.9 burned 60 sold 11.1 used as kerosene containers 20 Awareness that drift and run-offs can cause surface water pollution yes 31.1 no 68.9 Keeping a safe distance from canal/river while spraying Yes 8.9 No 91.1 Location of pesticide mixing and container washing near river/canal 71.1 at home 8.9 in the field (farm) 20 Trends in amount and frequency of pesticides used in last five years increase 82.2 decreased 0 no change 17.8 Reason for increment everyone increases 22.2 pest resistance 66.7 pesticide sellers said so 0 as a trial 11.1

a.i.

Question

15.1 e 0.4 e 65 e 636 74.5 10.4

Annex 1. Training and knowledge regarding pesticide handling and application trends

Water solubility at 20  C (mg/L)

The present study was funded by the Pesticide Risk Reduction Programme e Ethiopia (PRRP-Ethiopia), a collaborative project on pesticide registration and post-registration jointly set up by the Ministry of Agriculture of the Federal Republic of Ethiopia, and the State of the Netherlands represented by the Ministry of Foreign Affairs/Foreign Trade and Development Cooperation, and the Technical Cooperation Programme (TCP) of the Food and Agricultural Organisation of the United Nations. We thank Jan Klerkx for improving the language.

Molar mass pesticide (g)

Acknowledgements

DT50_water sediment(d)

water surface, and sampling done below the water surface may therefore miss part of it, while the model assumes ideal mixing across the transversal cross section of flow. Generally, higher PEC values were calculated for the stream and pond scenarios by PRIMET_Registration_Ethiopia_1.1 when actual, higher use patterns were used as input (Table 2) than when using use patterns based on good agricultural practices (GAP) (Teklu et al., 2015). Endosulfan PEC values found by Teklu et al. (2015) were based on GAP, while the present study used the actual application rate and frequency, higher than the GAP ones to predict the PEC values, which explains the higher concentrations.

287

KOC (mL/Kg)

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

Annex 2. Physico-chemical properties for nine pesticides used by small scale farmers (Source: Lewis et al. 2016)

288

a.i

Type

Substance group

Conc. a.i. in product (g/Kg or g/L)

Method of application

Dose of formulated product (According to GAP (k/ha or L/ha))

Actual dose of formulated product (Kg/ha or L/ha )

Application rate according to GAP (g a.i./ha)

Actual Application rate (g a.i./ha) actual

prescribed # of application/ cropping season

actual number of application

application intervals (According to GAP as well as actual (d))

Crop type in PRIMET_ Registration _Ethiopa_1.1

Application start

Lambdacyhalothrin Profenofos Malathion Endosulfan Deltamethrin Pyroxsulam Propiconazole Glyphosate Diazinon

IN

Pyrethroid

80

spray

0.4

1.2

32

96

3

6

7

C1st

July 15

IN IN IN IN HB FU HB IN

Organophosphate Organophosphate Organochlorine Pyrethroid Triazolopyrimidine Triazole Phosphonoglycine Organophosphate

720 500 350 60 450 250 480 600

spray spray spray spray spray spray spray spray

1.5 1.5 2.5 1 0.4 1 4 2

3.5 3.5 3.5 1.5 0.4 2.5 4 2.5

1080 750 875 60 180 250 1920 1200

2520 1750 1225 90 180 625 1920 1500

2 2 2 2 1 2 1 2

4 4 4 4 1 4 1 3

20 25 20 15

T1st T1st T1st C1st T1st T1st C1st C1st

July 15 July 15 July 15 July 15 June 15 July 15 June 15 July 15

Note: C1st ¼ Cabbage first cycle; T1st ¼ Tomato first cycle.

7 15

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

Annex 3. Pesticide use by small-scale farmers, according to GAP and applied according to farmers themselves

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

289

Annex 4. Pesticide use by large scale flower farms

#

common name

Trade name

Mode/Method Used dose (Kg/ha, Prescribed dose Formulation Use (Kg/ha, L/ha or (1 ¼ liquid category of application L/ha or g/m2) g/m2) 2 ¼ powder 3 ¼ granule 4 ¼ other)

Application intervals

No of application

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Abamectin Dodemorfacetate Bupirimate Spnosad Imidacloprid Spiromesifen Spiroxamine Lufenuron Organosilicone Mancozeb þ Metalaxyl Chlorothalonil Pyrimethanil Cadusafos Bifenazate Methomyl Propamocarb Clofentezine Dodemorfacetate Sulfur Iprodione Oxymatrine Pyriproxyfen Teflubezuron Tetraconazole Fenhexamid Boscalld þ Keroxymeth Dodemorfacetate Methiocarb Cyprodinil Fostail-Alumunum 80 WP Carbandazim 50 SC Acetamiprid 20 SP Boscalld Deltamethrin

Biomectin Metatox Nimrod Tracer 480 sc confidor 200 SL Oberon 240 SC Impulse 500 EC Match aminogold Mistress 72% WP Adona 72 SC Scala Rugby 100 ME Floromite Lannate Pervicure N 12 Apollo 50% SC Metatox Thiovet Rovral 8þ64% WP Oxymetrine 1% SL Prempt Nomolt Domark Teldor 50 WG Collis Metatox Mesurol 500 EC Switch Fostail WP

1 2 2 1 2 1 1 2 1 2 1 1 1 2 1 1 1 2 2 1 2 1 2 2 2 1 1 2 2 1

IN FU FU IN IN FU FU IN IN/FU FU FU FU NE FU IN FU IN/AC FU FU FU IN IN IN FU FU FU FU IN FU FU

spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar

1 lit/ha 1.5 lit/ha 1.5 lit/ha 1 lit/ha 1 lit/ha 1.5 lit/ha 1.5 lit/ha 1 lit/ha 0.6 lit/ha 2.5 kg/ha 1.5 lit/ha 1.5 lit/ha

same same same same same same same same same same same same

as as as as as as as as as as as as

used used used used used used used used used used used used

dose dose dose dose dose dose dose dose dose dose dose dose

5e7 days 7 days 7 days 5e7 days 5e7 days 7 days 7 days 5e7 days 10 days 6 days 7 days 7 days

2rounds 2rounds 2rounds 2rounds 2rounds 2rounds 2rounds 2rounds everyday 3rounds 2rounds 2rounds

spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar spray/foliar

1.5 lit/ha 1 lit/ha 1.5 lit/ha 1 lit/ha 1.5 lit/ha 0.7 ke/ha 1.5 lit/ha 1 lit/ha 1 lit/ha 1 lit/ha 1.5 lit/ha 1.5 lit/ha 1.5 lit/ha 1.5 lit/ha 1 lit/ha 1.5 lit/ha 1.5 lit/ha

same same same same same same same same same same same same same same same same same

as as as as as as as as as as as as as as as as as

used used used used used used used used used used used used used used used used used

dose dose dose dose dose dose dose dose dose dose dose dose dose dose dose dose dose

7 days 5e7 days 7 days 5e7 days 7 days 10 days 7 days 5e7 days 5e7 days 5e7 days 7 days 7 days 7 days 7 days 5e7 days 7 days 7 days

2rounds 2rounds 2rounds 2rounds 2rounds everyday 2rounds 2rounds 2rounds 2rounds 2rounds 2rounds 2rounds 2rounds 2rounds 2rounds 2rounds

Bavistin Golan Bellis Decis 2.5 EC

1 1 2 2

FU IN FU IN

spray/foliar spray/foliar spray/foliar spray/foliar

1.5 lit/ha 1 lit/ha 1.5 lit/ha 1 lit/ha

same same same same

as as as as

used used used used

dose dose dose dose

7 days 5e7 days 7 days 5e7 days

2rounds 2rounds 2rounds 2rounds

31 32 33 34

Annex 5. Toxicity data for detected OCPs (Source: Lewis et al. 2016)

OCP detected

a-Chlordane g-Chlordane b-BHC Dieldrin Endosulfan Heptachlor Epoxide B Heptachlor

ARfD (mg/kg_bw*d)

0.0005* 0.0005* NA 0.003 0.02 0.0001* 0.0001*

EC/LC50 values (mg/L) Algae

Daphnia

Fish

NA NA 1 10 215 2.7 2.7

5.9 5.9 5 2.5 4.4 0.42 0.42

0.9 0.9 0.3 0.012 0.02 0.07 0.07

Note: NA ¼ Not Available; ADI ¼ Acceptable Daily Intake; ARfD ¼ Acute reference dose; EC/LC50 ¼ Effect/Lethal concentrations; * ADI (ARfD: none allocated)

290

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

Annex 6. Toxicity data for nine pesticides used by small scale farmers (Source: Lewis et al. 2016)

a.i.

Aquatic ecosystem

Human Health

Acute

Lambda-cyhalothrin Profenofos Malathion Endosulfan Deltamethrin Pyroxsulam Propiconazole Glyphosate Diazinon

Acute

Acute

Fish LC50 (mg/L)

Daphnia EC50 (mg/L)

Algae EC50 (mg/L)

Macrophyts EC50 (mg/L)

ARfD (mg/kg_bw*d)

0.00021 0.08 0.018 0.002 0.00026 >87 2.6 38 3.1

0.00036 0.5 0.0007 0.44 0.00056 100 10.2 40 0.001

>0.3 NA 13 2.15 9.1 0.924 0.093 4.4 6.4

NA NA NA NA NA 0.0026 4.9 12 NA

0.005 1.0 0.3 0.02 0.01 0.9* 0.3 0.3* 0.025

* ADI (no ARfD allocated)

Annex 7. Measured water physical parameters as a function of sampled locations and times in 2014

Sampling location and (elevation (m))

Sampling time in 2014

EC mS/cm

pH

DO(mg/L)

T (oC)

TDS (mg/L)

TSS (mg/L)

TS(mg/L)

Crop development stage

USR (2190 m)

11-07 11-08 11-09 11-10 11-07 11-08 11-09 11-10 11-07 11-08 11-09 11-10 20-10 20-10 20-10

670 420 560 545 490 570 520 598 320 616 440 470 360 510 540

7.82 6.96 7.65 7.45 8.17 7.45 8.12 6.97 7.52 7.98 7.34 8.21 7.34 7.98 7.34

7.45 5.46 6.45 7.45 6.34 6.34 4.34 7.45 6.22 7.22 6.34 8.98 7.56 5.45 7.34

18.9 19.3 20.1 22.1 22.2 20.1 20.4 22.2 19.5 19.8 21.4 21.4 22.5 21.3 22.1

420 260 370 320 310 360 330 380 200 390 280 300 220 320 340

1180 1060 700 240 750 1190 440 260 1460 810 670 500 320 180 1260

1600 1320 1070 560 1060 1550 770 640 1660 1200 950 800 540 500 1600

land preparation seedling flowering maturity land preparation seedling flowering maturity land preparation seedling flowering maturity maturity maturity maturity

MSR (1935 m)

DSR (1879)

USP (2228 m) MSP (1943 m) DSP (1882 m)

Note: USR ¼ Up Stream River; MSR ¼ Mid Stream River; DSR ¼ Down Stream River; USP ¼ Up Stream Pond; MSP ¼ Mid Stream Pond: DSP ¼ Sown Stream; EC ¼ Electrical Conductivity; DO ¼ Dissolved Oxygen; T ¼ Temperature; TDS ¼ Total Dissolved Solids; TSS ¼ Total Suspended Solids; TS ¼ Total Solids (Dissolved þ Suspended).

Annex 8. The status of liquid waste management by the seven commercial farms downs stream (Source: Survey 2014)

#Farm

Type

Production system

1 2 3 4 5 6 7

Commercial Commercial Commercial Commercial Commercial Commercial Commercial

Green Green Green Green Green Green Green

house house house house house house house

under under under under under under under

irrigation irrigation irrigation irrigation irrigation irrigation irrigation

Major source of liquid waste( 1 ¼ greenhouses 2 ¼ pack houses and cleaning rooms 3 ¼ pesticide mixing rooms 4 ¼ any other)

Availability of liquid waste accumulation and detoxification (recycling) (1 ¼ yes 2 ¼ No)

Type of liquid waste management( 1 ¼ wet land plates 2 ¼ soak away pits 3 ¼ recycling through silo 4 ¼ any other)

1,2,3 2,3,4(fertigation) 2 1 1,2,3 1,2,3 1,2,3

1 1 2 1(under construction) 1 1 1

2,3 2,3 none 1(under construction)2,3 2,3(under construction) 1,2,3 2

B.M. Teklu et al. / Chemosphere 161 (2016) 280e291

References Abayneh, A., Taddese, W., Chandravanshi, B.S., 2003. Trace metals in selected fish species from lakes Awassa and Ziway, Ethiopia. Ethiop. J. Sci. 26, 103e114, 2003. Adeyemi, D., Anyakora, C., Ukpo, G., Adedayo, A., Darko, G., 2011. Evaluation of the levels of organochlorine pesticide residues in water samples of Lagos Lagoon using solid phase extraction method. J. Environ. Chem. Ecotoxicol. 3, 160e166. Adriaanse, P.I., Ter Horst, M.M.S., Teklu, B.M., Deneer, J.W., Woldeamanuel, A., Boesten, J.J.T.I., 2015. Development of Scenarios for Drinking Water Produced from Groundwater and Surface Water for Use in the Pesticide Registration Procedure of Ethiopia. Alterra report 2674. Wageningen, The Netherlands. Andreotti, G., Freeman, L., Hou, L., Coble, J., Rusiecki, J., Hoppin, J., Silverman, D., Alavanja, M., 2009. Agricultural pesticide use and pancreatic cancer risk in the agricultural health study cohort. Int. J. Cancer 124, 2495e2500. Ansara-Ross, T.M., Wepener, V., Van den Brink, P.J., Ross, M.J., 2008. Probabilistic risk assessment of the environmental impacts of pesticides in the Crocodile (west) Marico catchment, North-West Province. Water SA 34, 637e644. Bouman, H., 2004. South Africa and the Stockholm convention on persistent organic pollutants. Afr. J. Sci. 100, 323e328. Brock, T.C.M., Arts, G.H.P., Maltby, L., Van den Brink, P.J., 2006. Aquatic risks of pesticides, ecological protection goals, and common aims in European union legislation. Integr. Environ. Assess. Manag. 2, 20e46. Chauhan, A., Goyal, M.K., Chauhan, P., 2014. GC-MS technique and its analytical applications in science and technology. J. Anal. Bioanal. Tech. 5, 222. Clesceri, L.S., Arnold, E.G., Andrew, D.E. (Eds.), 1999. Standard Method for the Examination of Water and Waste Water, 20 ed. American Public Health Association and American Water Works Association, Washington DC, USA. De Koning, S., Janssen, H.G., Brinkman, U.A.T., 2009. Modern methods of sample preparation for GC analysis. Chromatogr. Suppl. 69, S33eS78. Deneer, J.W., Adriaanse, P.I., de Boer, P., Busschers, M., Lahr, J., Van der Schoor, C., Van Vliet, P., Woldeamanual, A., 2014. A Scientific Evaluation System for the Registration of Pesticides in Ethiopia. Wageningen, The Netherlands. Altera report 2547. EC, European Commission (1998). Council Directive 98/83/EC of 3 November 1998 on the quality of water intended for human consumption. OJ L 330/32, 5.12.98, p. 1e23. Elvira, V.B., Philipp, M., Lars, G., Katrin, V., Ole, R.N., 2011. Occurrence of organochlorine pesticides in indoor dust. J. Environ. Monit. 13, 522e526. Fiedler, H., Abad, E., van Bavel, B., de Boer, J., Bogdal, C., Malisch, R., 2013. The need for capacity building and first results for the Stockholm Convention Global Monitoring Plan. Trends Anal. Chem. 46, 1e206. Forrest, J., 2000. Collecting water quality samples for dissolved metals in water. In: USEPA, Region 6. Water Quality Handbook, 2nd edition. EPA, USA. Hellar, K.H., 2011. Pesticide residues in four rivers running through an intensive agricultural area, Kilimanjaro, Tanzania. J. Appl. Sci. Environ. Manage 15, 307e316. Jones, K.C., de Voogt, P., 1999. Persistent organic pollutants (POPs): state of the science. Environ. Pollut. 100, 209e221. Kashyap, S.M., Pandya, G.H., Wachasunder, S.D., Kondawar, V.K., 2005. QC/AC aspects of GC-MS analytical instrument for environmental analysis. Indian J. Chem. Technol. 12, 477e487. Lewis, K.A., Tzilivakis, J., Warner, D., Green, A., 2016. An international database for pesticide risk assessments and management. Hum. Ecol. Risk Assess. An Int. J. 22, 1050e1064. Macharia, I., 2015. Pesticides and health in vegetable production in Kenya. BioMed Research International. http://dx.doi.org/10.1155/2015/241516. Maltby, L., Blake, N., Brock, T.C.M., Van den Brink, P.J., 2005. Insecticide species sensitivity distributions: the importance of test species selection and relevance to aquatic ecosystems. Env. Toxicol. Chem. 24, 379e388. Mengistie, B.T., Mol, A.P.J., Oosterveer, P., 2016. Pesticide use practices in the central rift valley of Ethiopia. Environ. Dev. Sustain. http://dx.doi.org/10.1007/s10668015-9728-9. Michael, M.G., Seleshi, B.A., 2007. Irrigation Practices in Ethiopia: Characteristics of Selected Irrigation Schemes. International Water Management Institute, Addis Ababa, Ethiopia. Working Paper 124. Mnif, W., Hassine, A.I., Bouaziz, A., Bartegi, A., Thomas, O., Roig, B., 2011. Effect of endocrine disruptor pesticides: a Review. Int. J. Environ. Res. Public Health 8,

291

2265e2303. Okoya, A.A., Ogunfowokan, A.O., Asubiojo, O.I., Torto, N., 2013. Organochlorine Pesticide Residues in Sediments and Waters from Cocoa Producing Areas of Ondo State, Southwestern Nigeria. ISRN Soil Science Article ID 131647. Olaya-Contrras, P., Rodríguez - Villamil, J., Posso-Valencia, H.J., Cortez, J.E., 1998. Organochlorine Exposure and Breast cancer Risk in Colombian Women Cad. Saúde Pública, Rio de Janeiro 14, pp. 125e132. PAN-UK, 2006. Pesticide Use, Accumulations and Impacts: a Case Study in the Rift Valley. Ethiopia Environment and Social Assessment International. http://www. panuk.org/archive/Projects/Obsolete/Pesticide%20Use%20Accumulation%20and %20Impacts.pdf. Perry, M.J., Young, H.A., Grandjean, P., Halling, J., Petersen, M.S., Martenies, S.E., Karimi, P., Weihe, P., 2016. Sperm aneuploidy in faroese men with lifetime exposure to dichlorodiphenyldichloroethylene (DDE) and polychlorinated biphenyl (PCB) pollutants. Environ. Health Perspect. 124, 951e956. PHRD, 2015. List of Registered Pesticides as of June 2015. Ministry of Agriculture Plant Health Regulatory Directorate, Ethiopia. http://www.moa.gov.et/ documents/93665/4536019/-ListþofþRegisteredþPesticidesþasþofþJuneþ%2C %202015-.pdf/34309f8f-f790-4119-ac55-5d0b52a52ff2. Repetto, R., Baliga, S.S., 1996. Pesticides and the Immune System: the Public Health Risks. World Resource Institute, Washington DC, USA. ^a, R., Van den Brink, P.J., 2011. Effects of Rico, A., Waichman, A.V., Geber-Corre malathion and carbendazim on Amazonian freshwater organisms: comparison of tropical and temperate species sensitivity distributions and water quality criteria. Ecotoxicology 20, 625e634. Safford, R.J., Jones, C.G., 1997. Did organochlorine pesticide use cause declines in Mauritian forest birds? Biodivers. Conserv. 6, 1445e1451. Teklu, B.M., Adriaanse, P.I., Ter Horst, M.M.S., Deneer, J.W., Van den Brink, P.J., 2015. Surface water risk assessment of pesticides in Ethiopia. Sci. Total Environ. 508, 566e574. Teklu, B.M., Hailu, A., Wiegant, D.A., Scholten, B.S., Van den Brink, P.J., 2016a. Impacts of nutrients and pesticides from small and large scale horticulture on the water quality and ecology of the Lake Ziway catchment, Ethiopia. Env. Sci. Pollut. Res.,. http://dx.doi.org/10.1007/s11356-016-6714-1. Teklu, B.M., Retta, N., Van den Brink, P.J., 2016b. Sensitivity of Ethiopian aquatic macroinvertebrates to the pesticides endosulfan and diazinon, compared to literature data. Ecotoxicology 25, 1226e1233. Tenabe, S., Iwata, H., Tatsukawa, R., 1994. Global contamination by persistent organochlorines and their ecotoxicological impact on marine mammals. Sci. Total Environ. 154, 163e177. Van Vlaardingen, P., Traas, T.P., Aldenberg, T., 2003. Normal distribution Based Hazardous Concentration and Potentially Affected Fraction. ETX-2000. Rijksinstituut voor Volksgezondheid en Milieu, Bilthoven, The Netherlands. Wei, D., Kameya, T., Urano, K., 2007. Environmental management of pesticidal POPs in China: past, present and future. Environ. Int. 33, 894e902. Westbrom, R., Ahmed, H., Negussie, M., Negussie, R., Lennart, M., Erland, B., 2008. Assessment of organochlorine pesticide pollution in Upper Awash Ethiopian state farm soils using selective pressurized liquid extraction. Chemosphere 8, 1181e1187. Wheeler, J.R., Grist, E.P.M., Leung, K.M.Y., Morritt, D., Crane, M., 2002. Species sensitivity distributions: data and model choice. Mar. Pollut. Bull. 45, 192e202. WHO, 2010. Rapid Assessment of Drinking Water Quality in the Federal Democratic Republic of Ethiopia. Country Report. World Health Organization. http://www. wssinfo.org/fileadmin/user_upload/resources/RADWQ_Ethiopia.pdf. Wipfler, E.L., Adriaanse, P.I., ter Horst, M.M.S., Vlaming, J., Van den Brink, P.J., Peeters, F.M., Deneer, J.W., Boesten, J.J.T.I., Groenwold, J.G., 2014. PRIMET_ Registration_Ethiopia_1.1, technical description and manual. In: A Decision Support System for Assessing Pesticide Risk in the Tropics to Man, Environment and Trade, for Pesticide Registration in Ethiopia. Wageningen, The Netherlands. Altera report 2573. Yohannes, Y.B., Ikenaka, Y., Nakayama, S., Saengtienchai, A., Watanabe, K., Ishizuka, M., 2013. Organochlorine pesticides and heavy metals in fish from Lake Awassa, Ethiopia: insights from stable isotope analysis. Chemosphere 91, 857e863. _Zwir-Ferenc, A., Biziuk, M., 2006. Solid phase extraction technique e trends, opportunities and applications. Pol. J. Environ. Stud 15, 677e690.