A preliminary study of the content and distribution of pesticide residues in soil samples from the Kathmandu valley, Nepal

The increasing use of pesticides for agricultural production is causing soil pollution problems in different parts of Nepal. Uncontaminated agricultural soils are of great importance as they have a direct impact on food security and human health. The objective of this study was to investigate the quality and quantity of pesticides in soil samples from districts near the capital city of Kathmandu, from where fruit and vegetables are brought to the city for consumption. A questionnaire survey was carried out in four districts around Kathmandu city to investigate the types of pesticides that are most commonly used in these districts. A total of 15 soil samples were taken at a depth of 10 cm and four complete soil profiles were sampled at three different depths (10 cm, 30 cm and 50 cm) on the farms of those who were interviewed. A total of four replicates of each soil sample were extracted and analyzed. The pH, soil texture and organic carbon content of the soil samples were analyzed to understand the general soil characteristics. The QuEChERS method used for the analysis of


Introduction
The increasing production of food to fulfill the demands of a growing population throughout the world has resulted in the widespread use of pesticides. Some pesticides are highly persistent; they can last for many years before breaking down. These persistent substances can be highly mobile and are capable of bioaccumulation. They circulate globally: although they are released in one region, they can be easily transported through the atmosphere and to regions far away from the original source by repeated processes of evaporation and deposition (Williams 2000).
Organophosphorous insecticides like chlorpyrifos and chlorpyrifos-methyl are the pesticides that are most commonly used throughout the world !" #$%&'#()#*()#$#+*"(,*-./.0,*"&'0" -),.$"1,tabolites are still detected in the environment although several of them have been banned for several years (Kreuger et al. 2006). About 900 chemical pesticides are still used worldwide, both legally and illegally, with various food products and for the treatment of crops and soils (Thurman et al. 2008). The level of pesticide application for agricultural production is even more severe in developing countries due to efforts to eradicate insect borne diseases, to protect farms and forests and to produce adequate food (Schumann 2005). For several years, organophosphorous pesticides and pyrethroids have been widely used in agriculture, especially in developing countries. Although the use of organophosphorous pesticides and pyrethroids increases crop production, their usage has a severe negative impact on the environment (Wang et al. 2008).
Like any other country in the world, Nepal is confronted with the problems of extensive pesticide use and food security (Baker and Gyawali 1994;Palikhe 2002;Upadhyaya 2002). According to Dahal (1995), chemical pesticides were introduced into this country as early as 1955 when Paris Green, Gamaxone, and nicotine sulfates were imported from the United States of America (USA) for malaria control. Dichlordiphenyltrichlorethane (DDT) made its first impact in Nepal in 1956. This was soon followed by a variety of other organochlorine pesticides in the 1950s, organophosphorous pesticides in the 1960s, carbamates in the 1970s, and synthetic pyrethroids in the 1980s. The most commonly used pesticides in Nepal are malathion, chlorpyrifos-methyl, cypermethrin, deltametrin, mancozeb, parathion-methyl, fenvelarate, dichlorvos, endosulfan sulphate, dimethoate and carbendazim (Palikhe 2001). Many misuses have been reported generally from farmers who do not realize the extent to which pesticides are poisonous and hazardous to humans and the environment. Farmers and retailers of pesticides do not have adequate knowledge regarding pesticide use and health safety (Giri 1998;Baker and Gyawali 1994;Dahal 1995). Furthermore there are no strict government control mechanisms to control the purchase, trading, import and export of pesticides. A significant proportion -ranging from 20 to 70%-of an applied pesticide or its associated degradation products may remain in the soil as a persistent residue bound to soil colloids (Miglioranza et al. 2003). The presence of contaminants in agricultural soils above a certain level entails several negative consequences for food production and for the agricultural ecosystem. Through food chain and other pathways, like inhalation, pollutants can be accumulated within the human body and have an adverse impact on human health (Tang et al. 2010). There is a potential risk of pesticide contamination of ground waters and river systems due to their haphazard use. Chemicals seeping into shallow wells may quickly appear, but could possibly be cleared away within couple of years of treatment. Even so, in deep wells, they may still appear after a very long time because of the percolation time in soils. Such contamination cannot be easily treated and it can take many years for the chemicals to degrade if they remain distant from the active microbial zone (Khanal 2012). Areas with sandy soils face a higher risk of contamination than clayey soils due to easier water percolation (Engle et al. 1993). Carelessness during tank filling, mixing, spraying, re-filling, hand washing, tank emptying and cleaning, the disposal of pesticides near ground water sources like dug wells and tube wells, and/or deep boring, etc., may pose a serious threat.
Due to the lack of technical, financial and trained human resources required for the monitoring and analysis of pesticide residues in agricultural soils and food products, there is no up-to-date database on the pesticide concentrations in different parts of Nepal. Occasional tests for pesticides carried out by the Central Food Research Laboratory affiliated to the Nepalese Government showed contaminated fruit and vegetables brought to Kathmandu markets (Poudel 2011). This means that city dwellers depending on food products grown in peri-urban and rural parts all over the country unknowingly consume pesticides. A systematic study of agricultural soil pollution in different parts of the country is essential in order to elucidate the extent of contamination due to the use of haphazard pesticides and to assess potential risks for the health of local residents and the security of the agricultural products. In this regard, it was of utmost importance to conduct a study in these areas to investigate the types of pesticides used for agricultural production and their residue levels in soil samples.
The main objective of this study was therefore to investigate the quality and quantity of pesticides found in soil samples taken from districts neighboring the capital city of Kathmandu, from where fruit and vegetables are brought to market and thereby to quantify the risk with both a questionnaire survey and sample analyses. This study could be considered a first survey conducted on a pilot scale in order to detect the magnitude of pesticide pollution in the soils of the Kathmandu valley.  Figure  1). From these districts, fruits and vegetables are brought to Kathmandu markets every morning. A multiple choice questionnaire survey was carried out with thirty farmers from the selected sites at the end of September 2008. The main objective of the questionnaire survey was to gather information about the types of pesticides used by farmers to be used as baseline data for the purposive soil sample collection and laboratory analysis.

Sample collection
The soil samples collected on the farms of interviewed farmers were randomized across the field. A total of fifteen soil samples were collected at a depth of 10 cm. In order to determine the distribution of pesticide residues at different soil depths, a total of four soil profiles from Godavari, Lalitpur district were collected from the depths of 10 cm, 30 cm and 50 cm. About 2 kg of each soil sample was collected and then packed in a clean plastic bag and stored at -18 0 C until analysis. The extraction and analyses of soil samples were carried out at the Institute of Soil Research, University of Natural Resources and Life Sciences (BOKU), Vienna, Austria.  and parathion-methyl. Stock solutions (1000 mg l -1 ) of each pesticide were prepared in acetonitrile. From each of the single stock standards, a mix standard of 10 mg l -1 was prepared and used for the preparation of calibration standards with subsequent dilution with acetonitrile. The single and mix standards were stored at -4 0 C while the calibration standards were prepared fresh on the day of analysis.
The HPLC-MS/MS analyses were performed using an Agilent 1200 liquid chromatographic system interfaced to an Agilent 6440 triple quadrupole mass spectrometer (Agilent Technologies, Waldbronn). Mass Hunter software was used for data acquisition and data processing.

Soil sample preparation and extraction
The soil samples were freeze dried for 24 h, gently crushed in a ceramic mortar and sieved (< 2 mm). A total of four replicates of each soil sample were made for extraction followed by the laboratory analysis. The QuEChERS method reported by Anastassiades et al. (2003) and developed for the extraction of pesticides from fruits and vegetables was modified and applied to the extraction of soil samples as follows: 10 g of homogenized soil sample was weighed in a 50 ml teflon tube and 10 ml of ultrapure water was added. The mixture was vigorously shaken for 1 min using a vortex, and then 10 ml of ACN was added and the suspension sonificated with an ultrasound system for 1 min. Four g MgSO 4 and 1 g NaCl were added and the sample shaken again for 1 min. Fifty µl of 10 mg l -1 TPP as internal standard was added and the tube shaken for 30 s followed by centrifugation at 3452 g for 30 min. Six ml of an aliquot of the supernatant was transferred to a teflon centrifuge tube (15 ml), and 0.90 g MgSO 4 and 0.15 g PSA were added. The tube was shaken for 30 s and centrifuged at 3452 g for 15 min. A 1.5 ml aliquot of the supernatant was filtered through a Whatmann syringe filtering vial (0.45 µm) and transferred to a LC vial. The extract was evaporated to dryness in a nitrogen stream and the residue was redissolved with ACN to make up 1.5 ml, and was used for further qualification and quantification in the LC-MS/MS system.

Recovery experiments
The recovery was determined using 3 replicates at 8 spiking concentrations from 0.01 mg kg -1 to 2 mg kg-1, with eight calibration solutions prepared in acetonitrile. The linearity was tested with eight concentrations between 0.01 mg l -1 to 2 mg l -1 and correlation factors (R 2 ) were calculated.

Chromatography and mass spectrometry
The chromatographic separation was achieved using a Zorbax SB-C-18 column 2.1×100 mm, 1.7 µm particle size from Agilent Technologies at a flow rate of 0.6 ml min -1 . The column was thermostated at 40 0 C and the injection volume was 5 µl. The mobile phases consisted of A: H 2 O-MeOH, 90% -9.95% (v/v) with 0.05% HCOOH and B: H 2 O-MeOH, 9.95%-90% (v/v) with 0.05% HCOOH. The solvent gradient used is given in Table 1.
The MS/MS was equipped with an electrospray ionization (ESI) interface operated in positive mode. The nebulizer gas (nitrogen) pressure was 25 psi, the drying gas flow rate was 8 ml min -1 and the drying gas temperature was 300 0 C. Collision cell energy and fragmentor voltage were optimized in the dynamic Multiple Reaction Monitoring mode (MRM) for each pesticide and are listed in Table 2.
All the above mentioned compounds were analyzed in positive polarity mode with a dwell time of 30 ms.

Questionnaire survey
The questionnaire survey revealed that the farmers were using various pesticides for agricultural production: carbendazim (fungicide), chlorpyrifos-methyl (organophosphorous insecticide), cypermethrin (insecticide-synthetic pyrethroid), parathion-methyl (organophosphorous insecticide), imidacloprid (systemic insecticide), metalaxyl (systemic fungicide), dimethoate (organophosphorous insecticide), omethoate (organophosphorous insecticide) and dichlorvos (organophosphorous insecticide). According to a survey carried out by Dahal (1995) in some parts of eastern, western and central Nepal organochlorine insecticides like aldrin, endosulfan and BHC (benzene hexachloride) dust were found to be used in larger amounts by 95% of interviewed farmers. The study also concluded that farmers used chemical pesticides not only to control pests on crops but also to store the food grains, lentils, vegetables and fruits (Dahal 1995). However, in the present study, out of a total of thirty farmers interviewed, none of them were using organochlorine pesticides. The reason could be that the number of interviewed farmers was not big enough and was restricted only to a few districts around Kathmandu city. It might also be that organochlorine pesticides are not purchased anymore and have been substituted by persistent organophosphorous pesticides.

Recovery experiments
The recovery of the analyzed compounds was found to be 77.5-112% (standard deviation between 1.78% and 4.01%). All the compounds under study were found to be linear over a concentration range between 0.01 mg kg -1 -2 mg kg -1 with correlation factors R 2 between 0.996-0.999. The LOQ (limit of quantitation) of the analyzed compounds were between 0.2 µg kg -1 and 6.25 µg kg -1 . The highest value of LOQ was found to be 6.25 µg kg -1 for parathion-methyl while the lowest value of 0.20 µg kg -1 was found for metalaxyl. The recoveries, relative standard deviation, correlation factors and LOQ are represented in Table 3.

Standard deviation (%)
Correlation factor (R 2 ) LOQ (μg kg -1 )  Table 3. The recoveries, relative standard deviation, correlation factor and LO Q (limit of quantitation) of the studied pesticides. The me asurements were made for eight calibration solutions with thre e replicates of e ach

Results of the soil analyses
The soil texture, pH and organic carbon content are presented in Table 4. The sites Kathmandu, Godavari, Thimi and Chattradeurali had loamy sand, Kirtipur and Bhaktapur had silty sand, and Lalitpur and Dhading Besi had clay sand and sandy loam respectively. Generally, sandy soils tend to have a low organic matter content. The lesser the organic matter content in soil, the lesser the microbial activity, which increases the probability of pesticide contamina-tion on the soil surface (Kerle et al. 2007). The soils from Kathmandu and Thimi were strongly acidic, from Kirtipur, from Godavari and Bhaktapur very strongly acidic and from Chattradeurali and Dhading Besi moderately acidic when the values were compared with Bruce and Rayment (1982). The organic carbon content of the soil from Thimi was higher than that of the other sites with a value of 3.7%. Soils that have an organic layer, such as crop residues or thatch in turf grass, may strongly sorb pesticides and reduce their mobility (Kerle et al. 2007).
The soil samples at 10 cm from all the selected sites except that from Kathmandu (Kathmandu and Kirtipur) were contaminated by various pesticides (Figure 2). The soil samples were contaminated with the fungicide carbendazim up to 0.038 mg kg -1 , the insecticide chlorpyrifosmethyl up to 0.038 mg kg -1 and the systemic insecticide imidacloprid up to 0.016 mg kg -1 . A study carried out by the Soil Entomology Division of Nepal Agricultural Research Council (NARC) revealed that the analytical results of soil samples from the vegetable growing area of Kavre, Dhading, Chitwan and Bhaktapur showed only a few isomers of ! and "!234"5!-BHC -0.001 mg kg -1 and "!234"67668"1%"9% -1 ) in some samples (Soil Entomology Division 1998).
The pesticide amounts in the soil samples of the present study were lower than the reported amount of applied pesticides according to the field survey. As far as we know, target values for pesticide residues in soil have not yet been developed, and therefore the present data could not be compared with any of the internationally authorized target values. However, according to environmental quality standards for soil (Wang et al. 2008), three soil contamination degrees are depicted: slightly polluted soils containing 0.05-0.5 mg kg -1 , moderately polluted soils with 0.5 to 1 mg kg -1 and heavily polluted soils with > 1 mg kg -1 . In the present study, the soil samples collected at the depth of 10 cm were found to be contaminated with the fungicide carbendazim and the insecticide chlorpyrifos-methyl up to 0.038 mg kg -1 and the systemic insecticide imidacloprid up to 0.016 mg kg -1 . When the pesticide residue levels in the soil samples of this study are compared with these target values, each soil sample of this study can be defined as having little pollution. The reason could be due to different soil properties and climatic conditions (Herrmann et al. 2002). An investigation into the degradation of pesticide in soil samples contaminated with malathion, dimethoate, fenvalarate and metalaxyl at different moisture contents and temperatures under defined laboratory conditions showed that the disappearance of pesticides was mainly dependent on moisture content and incubation temperature (Vinke et al. 2002). Other factors responsible for the concentration of pesticide residues in soil are the chemical properties of soil, soil use and type, persistence of the pesticide, the technique and rate of application, the frequency and timing of precipitation, soil organic carbon, the tillage system etc. (Redondo et al. 1994;Wang et al. 2006;Fabietti et al. 2009;Ling et al. 2010). The soil samples in the present study were collected during monsoon season at the end of September 2008, which could be one of the reasons for the lower than expected pesticide concentration when the values are compared with the standard values as described in Wang et al. (2008).
The distribution pattern of pesticide residues at different soil depths (10 cm, 30 cm and 50 cm) of the four collected soil profiles from Godavari (Lalitpur district) showed that they were contaminated with dimethoate, omethoate, dichlorvos, parathion-methyl and metalaxyl at different concentrations. The soil profile survey revealed a homogenous distribution of the pesticides in all the studied depths (Figures 3a to 3e). As all the samples of this study were agricultural soils, the tillage activity could have churned and homogenized the concentration of the pesticide residues in the studied soil depths. A similar study carried out to investigate the pesticide residue level at different soil depths (surface (0-30 cm) and subsurface (30-60 cm)) showed that the distribution in surface layer was higher than in the sub surface layer (Al-Wabel et al. 2011). The movement of pesticides through a soil profile de-  pends mainly on solubility, adsorption and desorption of the compounds. Most pesticides and their biologically active metabolites only penetrate to deeper soil layers in very special circum-stances, although penetration in the deeper soil layers has been reported by some researchers (Manandhar 2005).

Conclusions
The field study revealed that various organophosphorous pesticides, insecticides, systemic fungicides and synthetic pyrethroids were detectable in the soils. The soils were acidic and sandy with an organic carbon content in the range of 1.5% for all the samples except those from Thimi. The soil samples collected at 10 cm depth were found to be contaminated with pesticides at various concentrations. Maximum concentrations of 0.038 mg kg -1 for carbendazim and 0.038 mg kg -1 for chlorpyrifos-methyl were found in the Chattradeurali II of Dhading and Godavari of Lalitpur districts respectively. However, the concentration of pesticide residues in the analyzed samples was found to be less than that observed in the field application. Another test carried out at different soil depths (10 cm, 30 cm, and 50 cm) showed that the pesticides were distributed homogenously through the profile. An "old" insecticide parathion-methyl, which has been banned for use in many countries, was still applied and found in some samples. The findings of the present study are based on a small sample size, which decreases the probability of finding contaminated samples. However, this study is the first to quantify the magnitude and type of soil contamination due to pesticide use in central Nepal, and by doing so, reveal the pesticide use and safety among farmer communities in the region. It could act as a base for a comprehensive study covering all the ecological divisions of Nepal (Terai, Hills and Mountain), which would result in a better understanding and more meaningful conclusions to be drawn for the whole country. This study also shows that an information campaign regarding pesticide use and food safety for farmers and consumers is essential.