Temporal–spatial loss of diffuse pesticide and potential risks for water quality in China

Science of the Total Environment 541 (2016) 551–558

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Temporal–spatial loss of diffuse pesticide and potential risks for water quality in China Wei Ouyang ⁎, Guanqing Cai, Weijia Huang, Fanghua Hao State Key Joint Laboratory of Environmental Simulation and Pollution Control, School of Environment, Beijing Normal University, Beijing 100875, China

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Pesticide uses of six crops in China over two decades were summarized. • Temporal-spatial differences of three kinds of pesticides were highlighted. • Spatial pattern of diffuse pesticide loss and potential risk for water was figured. • Estimation uncertainties of pesticide use and loss were quantified.

a r t i c l e

i n f o

Article history: Received 21 August 2015 Received in revised form 23 September 2015 Accepted 23 September 2015 Available online 3 October 2015 Editor: D. Barcelo Keywords: Pesticide Temporal–spatial pattern Agricultural development Diffuse pollution Water risk Uncertainty

⁎ Corresponding author. E-mail address: [email protected] (W. Ouyang).

http://dx.doi.org/10.1016/j.scitotenv.2015.09.120 0048-9697/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t Increasing amount of pesticide has been used in Chinese agricultural system with effects on environmental quality and human health. The comprehensive inventory of pesticide use in six main crop categories over the period from 1990 to 2011 in China was conducted. The national average pesticide use intensity was estimated 1.74 kg·ha−1 for grain crops in paddy land, 1.31 kg·ha−1 for grain crops in dry land, 1.38 kg·ha−1 for economic crops, 3.82 kg·ha−1 for vegetables, 1.54 kg·ha−1 for tea plantations, and 3.49 kg·ha−1 for orchards. The pesticide use was estimated to be approximately 5.24 × 104 t for grain crops in paddy land, 1.05 × 105 t for grain crops in dry land, 3.08 × 104 t for economic crops, 7.51 × 104 t for vegetables, 3.26 × 103 t for tea plantations, and 4.13 × 104 t for orchards. Based on the pesticide use and loss coefficients for each category, the distribution of pesticide loss in China was calculated. Total pesticide loss in China was estimated about 4.39 × 103 t in 2011. The pesticide loss from six main crop categories was about 14.84% for grain crops in paddy land of total pesticide loss, 33.31% for grain crops in dry land, 10.47% for economic crops, 26.37% for vegetables, 1.08% for tea plantations and 13.93% for orchards. The results indicated that the highest pesticide use intensity and highest pesticide loss rate occurred in China's eastern and central provinces. The Monte Carlo simulation was used to quantify the uncertainties associated with estimation of pesticide use and loss rate for the six types of crops. The potential risk to national water quality was assessed and the water in the provinces of Henan, Shandong, Hebei, Beijing and Shanghai was at high risk for pesticide pollution. The implication for the future agricultural and environmental policies on reducing the risk to environmental quality was also summarized. © 2015 Elsevier B.V. All rights reserved.

552

W. Ouyang et al. / Science of the Total Environment 541 (2016) 551–558

1. Introduction

2. Methodology and key parameters

The pesticides play important role in safeguarding crop yields, which are used in agricultural tillage around the world and contribute to global food safety (Carvalho, 2006). However, the excessive application of pesticides over the past half century has posed serious risks to human health and water quality (Kolpin et al., 1998). Pesticide residues in water and soil are the significant environment threats and have been classified as carcinogens pollutants in many countries (Dich et al., 1997; Bressa et al., 1997). In China, the food security is a national priority due to the big population and lead to widespread application of pesticides on farmland. Consequently, the pesticides have been detected in some Chinese rivers (Feng et al., 2003; Zhou et al., 2006). Thus, it is necessary to determine the temporal–spatial patterns of pesticide use and loss and to assess risk for water quality. The total cultivation area in China increased to 1.63 × 107 ha in 2013, and the agricultural planting structure has shifted in response to recent economic development (Zhao et al., 2008). The cultivated area devoted to grain and economic crops remain stable, but the areas sown with vegetables and orchards increased (Liu and Savenije, 2008). Given the differences in pesticide use for various crop categories, the variation in the spatial pattern of agricultural planting directly impacts pesticide use. It is well known that grain crops in paddy land, tea bushes and orchards are commonly planted in southern China, and the grain crops in dry land are concentrated in northern China (Ministry of Agriculture, 2011). However, there is limited understanding of the detailed temporal–spatial characteristics of pesticide use with crop categories. Some parts of the pesticides sprayed on crops will remain in farmland, but some of them will enter the surrounding air, soil and water (Malone et al., 2004; Lefrancq et al., 2013). As the artificial organic compounds, the pesticides can remain in the environment for many years and may be transported over long distance (Scholtz et al., 2002). The potential environmental risks posed by pesticide loss during runoff and erosion are of considerable concern and have become a priority issue for the Chinese Environmental Protection Agency (Bao et al., 2012). The National Agricultural Diffuse Pollution Action Plan, which was issued in 2015, set detailed diffuse pollution control goal for the near future (Ministry of Agriculture, 2011). To effectively prevent water pollution from diffuse pesticide loss, the detailed information concerning the main types of pesticides used on crops at provincial scale is required for subsequent action. Some methods have been used to assess pesticide loss at the watershed level, and the field monitoring of pesticide discharge is the basic method for determining pesticide concentrations on-site. The measurements also provide sound data for the assessment of larger areas (Müller et al., 2006). On the watershed scale, diffuse pollution modelling has gained popularity in recent decades (Bannwarth et al., 2014). However, these distributed models cannot provide reliable results on the national scale because of complicated territorial dynamics and limited monitoring data. Therefore, a multiple-year comprehensive inventory of pesticide use is employed in this study, which has been applied in diverse case studies (Sarigiannis et al., 2013). To improve the understanding of diffuse pesticide pollution, we analyzed the temporal–spatial dynamics of Chinese pesticide use from 1990 to 2011. The potential pesticide loss due to runoff and erosion at a provincial scale was determined and used to evaluate the risk to water quality. The pesticide use and loss was investigated by crop categories, which provided information about the response to agricultural structure change and had implications for national water management. Finally, the uncertainty analysis was conducted to improve the accuracy and applicability of the estimation.

2.1. Pesticide use calculations Three main types of pesticides (insecticides, herbicides and fungicides) account for more than 95% of the total use, which were calculated separately. The pesticide use amounts were estimated for the following crop categories: grain crops in dry land (GCDL), grain crops in paddy land (GCPL), economic crops (EC), vegetables (VG), tea plantations (TP) and orchards (OC). The GCDL referred to cereals (wheat and corn), soybeans and tubers. The GCPL was rice. The EC covered the oilbearing crops, cotton, fiber crops, sugar crops and tobacco. The pesticide use amounts for these six categories were calculated using the bottomup approach with a refined pesticide type database and agricultural planting structure database (Tian et al., 2014). The basic equation for pesticide use inventory was followed: U Total ðnÞ ¼

X

U i ðnÞ ¼

i

XX i

P i;m ðnÞ  Ai ðnÞ  0:001

ð1Þ

m

where U is the annual pesticide use amount for a crop category in China (t/year); P is the pesticide use intensity of crops (kg/ha); A is the cultivated crop area (ha/year); and i, m, and n indicate the province (autonomous region or municipality), pesticide type, and crop category, respectively. The pesticides were counted based on their active ingredients. The pesticide use intensity was defined as the annual pesticide use amount per unit of cultivated area (kg/ha). There are large differences in pesticide use intensity among crop types because of different growing conditions and the crop's resistance to disease, pests or weeds (Lucas, 2011). The pesticide use intensity factor was defined as the ratio of the pesticide use amount for a specific crop to that of grain crops, which was used to evaluate the pesticide use intensity for different crops. Based on previous studies, the pesticide use intensity factor was set at 1.0 for grain crops, 1.0 for economic crops, 2.5 for vegetables, 1.0 for tea plantations and 2.5 for orchards (Kolpin et al., 1998; Fu and Qi, 1998; General Administration, 2007; Short and Colborn, 1999). Pesticide use intensity was calculated as Eq. (2): P i;m ðnÞ ¼ Gi;m  F ðnÞ

ð2Þ

where P is the pesticide use intensity of crop (kg/ha); G is the pesticide use intensity of grain crop (kg/ha); and F is the pesticide use intensity factor of crop. The pesticide use intensity for grain crop was calculated as Eq. (3): U i;m  1000 Gi;m ¼ X A  Fn n i;m

ð3Þ

where U is the pesticide use amount by province (t); F is the pesticide use intensity factor of crop; and A is the annual cultivated crop area (ha). The planting structure and cultivated area data of crops in each province from 1991 to 2011 was obtained from the China Statistical Yearbook. The pesticide use data at a provincial scale over the same period was referred to the Development Center of Science and Technology, Ministry of Agriculture, China. 2.2. Pesticide loss calculation from farmland The representative chemicals of insecticide, herbicide and fungicide categories were chosen to define the pesticide loss coefficient. The chlorpyrifos and carbendazim were chosen as the representatives for insecticides and fungicides, respectively. The acetochlor and butachlor were selected as the representatives of herbicides used on crops in dry land and paddy land, respectively. The use amounts of these four

W. Ouyang et al. / Science of the Total Environment 541 (2016) 551–558

representative pesticides accounted for a large proportion of the total pesticide use, as these four chemicals have been widely used in China for over 30 years (Shu, 2009). The physical–chemical properties of the four representative pesticides were listed in Table 1. The pesticide loss for crops is calculated as Eq. (4): LTotal ðnÞ ¼

X

Li ðnÞ ¼

i

XX

Ri;m ðnÞ  Ai ðnÞ  10−6

ð4Þ

m

i

where L is the annual pesticide loss amount for crop category in China (t/year) and R is the pesticide loss rate for crop (g/ha). The pesticide loss rate was estimated using the loss coefficient method (Riise et al., 2004) as Eq. (5): R ¼ P  η  1000

ð5Þ

where η is the pesticide loss coefficient in farmland (%) and it was assumed that the same of a crop in one province. The η was calculated as Eq. (6): η¼

Ldissolvable þ Lsorbed  100% T

ð6Þ

where Ldissolvable and Lsorbed are the pesticide loss amounts in soluble and sorbed forms in farmland (g/ha), respectively; T is the pesticide use amount in a typical agricultural region (g/ha). The algorithms used to determine the Ldissolvable and Lsorbed values, which was calculated with the pesticide transport module of the Soil and Water Assessment Tool (SWAT) model (Neitsch et al., 2011). The amount of soluble pesticides transported via surface runoff was determined by the concentration, amount of surface runoff, and pesticide percolation coefficient, as followed: Ldissolvable ¼ βpst  concpst;flow  Q surf

ð7Þ

where βpst is the pesticide percolation coefficient; concpst, flow is the concentration of pesticide in mobile water at top 10 mm of soil (g/mm H2O); and Qsurf is the amount of surface runoff (mm H2O). The pesticide attached to soil particles and transported by surface runoff. The amount of pesticides transported to streams via sorption to sediments was expressed by Eq. (8): Lsorbed ¼ 0:001  C soilphase 

sed ε area pst:sed

ð8Þ

where Csolidphase is the pesticide concentration in sediment (g/t); sed is the sediment yield (t); area is the area of farmland (ha); and εpst:sed is the pesticide enrichment ratio. The pesticide loss coefficients were calculated for six groups of crop in each province. The range of pesticide loss coefficients was 0.10%– 3.67%, which is similar to estimates from farm fields in America and Norway (Kellogg et al., 2002; Riise et al., 2004).

553

The pesticide property data were obtained from the pesticide database of the SWAT model and the National Pesticide Information Center of America. The soil properties data were part of National Soil Scientific Database. The precipitation data were obtained from the National Meteorological Information Center of the China Meteorological Administration. The irrigation data were referred to the China Water Resources Bulletin. The agricultural activity data were obtained from the China Statistical Yearbook. 2.3. Data analysis and uncertainty analysis To better understand the spatial pattern of pesticide use in China, the pesticide use intensity was estimated at provincial scale. Based on the national land use distribution in 2011 at 30 m resolution, the spatial variability of pesticide loss rates in farmland was analyzed by resampling to 1000 m. After that, pesticide concentrations in water were estimated based on the pesticide losses and surface water resource of each province. With this analysis, the water safety in China with respect to pesticide contamination was assessed. To evaluate the accuracy of the calculations, the Monte Carlo simulation was used to analyze the uncertainty of pesticide use and losses for six types of crops. It was assumed a normal distribution with a coefficient of variation of 5% for cultivated areas (Tian et al., 2014). The pesticide use intensity factors and pesticide loss coefficients observed the uniform distribution (Zhao et al., 2011). 3. Results 3.1. Pesticide use calculation and spatial distribution The pesticide use of six crop categories in each province of China in 2011 were calculated (Table 2). The total pesticide use in China was approximately 3.08 × 105 t (log value: 5.49). The insecticides accounted for the largest proportion (40.53%) of the total pesticide use. The herbicides and fungicides accounted for 31.64% and 23.62%, respectively. The proportions of the three main pesticide types varied remarkably by province (see Supporting Information (SI) Tables S1, S2 and S3). The five provinces (SD, GD, HLJ, HeN and AH; see province abbreviations in Table 2) with the highest pesticide use accounted for 35.3% of the national total value. Pesticide use in GCPL was estimated to be 5.24 × 104 t (log value: 4.72), which was the largest among the six crop categories. Pesticide use was 1.05 × 105 t (log value: 5.02) in GCDL and 3.08 × 104 t (log value: 4.49) in EC in 2011, respectively. Because of the difference in the cultivated area of each crop and the planting structure among provinces, there were large differences in provincial pesticide use inventories. To better understand the spatial pattern of pesticide use in China, the pesticide use intensity and ratio of pesticide types at provincial scale was estimated (Fig. 1). The national averaged pesticide use intensity was 1.90 kg·ha−1 in 2011. The five provinces with the largest pesticide use intensities were HN, GD, FJ, ZJ and JX. It was noted that the eastern and central provinces had higher pesticide use intensities than the other

Table 1 The physical–chemical properties of the representative pesticidesa. Pesticides

MWb (g/mol)

Water solubility (mg/L)

Bioaccumulation (Log Kowc)

Sorption coefficient (Kocd)

Soil half-life (days)

Pesticide movement rating

Acetochlor Butachlor Chlorpyrifos Carbendazim

269.8 311.9 350.6 191.2

223 23 0.4 8

4.14 2.5 4.70 1.5

176 700 6070 400

6.3 12 30 120

Moderate Low Very low Moderate

a The properties of butachlor, carbendazim and chlorpyrifos refer to OSU Extension Pesticide Properties Database, National Pesticide Information Center, USA. The properties of acetochlor refer to a report named with Pesticide Fate in the Environment: A Guide for Field Inspectors (Gillespie et al., 2011). b Molecular weight. c Octanol–Water Partition Coefficient as the log vale. d Soil adsorption coefficient normalized for soil organic carbon content.

554

W. Ouyang et al. / Science of the Total Environment 541 (2016) 551–558

Table 2 Pesticide use amounts for various crops by province in China, 2011 (Log ton/year)a. Province

GCPLb

GCDL

EC

VG

TP

OC

Total

Beijing (BJ) Tianjin (TJ) Hebei (HeB) Shanxi (SX) Inner Mongolia (IM) Liaoning (LN) Jilin (JL) Heilongjiang (HLJ) Shanghai (SH) Jiangsu (JS) Zhejiang (ZJ) Anhui (AH) Fujian (FJ) Jiangxi (JX) Shandong (SD) Henan (HeN) Hubei (HuB) Hunan (HuN) Guangdong (GD) Guangxi (GX) Hainan (HN) Chongqing (CQ) Sichuan (SC) Guizhou (GZ) Yunnan (YN) Tibet (TB) Shaanxi (SaX) Gansu (GS) Qinghai (QH) Ningxia (NX) Xinjiang (XJ) National total

–c 1.15 1.96 0.00 1.72 3.06 3.13 3.69 2.29 3.58 3.38 3.62 3.26 3.92 2.38 2.86 3.56 3.82 3.66 3.43 3.01 2.81 3.34 2.69 3.11 0.00 1.88 – – 1.57 1.51 4.72

2.41 2.46 3.84 3.52 3.51 3.64 3.88 4.15 2.17 3.71 2.98 3.92 2.92 2.92 4.13 4.02 3.57 3.12 3.14 3.11 2.56 3.17 3.69 3.24 3.59 2.06 3.27 3.25 2.23 2.53 2.95 5.02

0.85 1.79 3.09 2.33 2.65 2.85 2.74 2.65 1.32 3.12 2.79 3.37 2.61 3.33 3.49 3.38 3.56 3.42 3.09 3.23 2.51 2.47 3.19 2.74 3.14 1.20 2.38 2.41 2.01 1.59 2.95 4.49

2.32 2.33 3.51 2.76 2.60 3.30 3.06 2.96 2.80 3.72 3.62 3.57 3.56 3.53 3.94 3.69 3.68 3.69 3.85 3.53 3.26 3.16 3.52 3.11 3.34 1.58 2.85 2.81 1.81 2.07 2.56 4.88

– – – – – – – – – 1.73 2.69 2.41 2.66 2.17 1.56 1.95 2.64 2.22 1.98 1.84 0.60 1.51 2.42 2.15 2.66 – 1.75 0.78 – – – 3.51

2.30 1.90 3.46 2.91 1.99 3.19 2.43 2.16 1.98 2.93 3.33 2.71 3.46 3.38 3.46 3.12 3.25 3.34 3.81 3.50 3.17 2.79 3.20 2.49 3.02 0.60 3.24 2.83 0.90 2.14 3.05 4.62

2.83 2.82 4.16 3.69 3.63 3.99 4.04 4.31 3.04 4.22 4.03 4.28 4.00 4.23 4.45 4.30 4.26 4.25 4.32 4.09 3.70 3.65 4.14 3.65 4.01 2.24 3.67 3.53 2.54 2.82 3.52 5.49

a

The unit for the data is a log value of ton per year. GCPL is for grain crops in paddy land, GCDL for grain crops in dry land, EC for economic crops, VG for vegetables, TP for tea plantations, OC for orchards. c “–” means no value. As the pesticide use for specific crops in a province is 0 t, a log value of 0 works no value. b

provinces. The proportions of the three types of pesticide also varied by province. The herbicide was the dormant pesticide in northeastern China, especially in IM and QH. However, the insecticides were dominant in other provinces. This difference was probably based on the different occurrences of crops, pests and weeds in each province. It was also found the remarkable difference in pesticide use intensity among crop types. The average pesticide use intensity at national level was estimated 1.74 kg·ha− 1 for GCPL, 1.31 kg·ha− 1 for GCDL, 1.38 kg·ha−1 for EC, and 1.54 kg·ha−1 for TP. The estimated average pesticide use intensity at national level for VG and OC was much higher than for the other crops, at 3.82 kg·ha−1 and 3.49 kg·ha−1, respectively. The larger proportions of vegetables and orchards in agricultural planting structure resulted in higher pesticide use intensity in southern China, which was also the area with the plenty water resource. 3.2. Temporal patterns of pesticide use The temporal trends of pesticide use for different crops were shown in Fig. 2. The total cultivated area in China increased from 1.48 × 108 ha in 1990 to 1.62 × 108 ha in 2011, which increased 9.38%. However, the total amount of pesticide use increased by approximately 50% over the same period. Temporal correlation analysis indicated that cultivated area was the key factor affecting pesticide use, and the correlation coefficient (r) between them was 0.729 (p b 0.01). The annual pesticide use for VG, OC and TP increased rapidly by 5.5% from 1990 to 2011, but the pesticide use for GCDL, GCPL and EC only increased at annual rate of 0.8% at same time. The differences in pesticide use growth rates among crops mainly because the differences in the cultivated area for each crop (see SI Table S4). The cultivated areas of GCDL, GCPL and EC grew by 0.14%, −9.09% and 10.69% from 1990 to 2011, respectively. In contrast, the cultivated areas of VG, TP and OC grew by 209.86%, 99.15%, and 128.44%, respectively. The increased proportion of cultivated areas devoted to VG, OC and TP contributed more to the increase in pesticide use than GCDL, GCPL and EC (r = 0.858, p b 0.01). The grain output per hectare increased from 4206 kg/ha to 5824 kg/ha from 1990 to 2011, which showed

Fig. 1. Spatial distributions of pesticide use intensity and ratio of pesticide types by province in 2011.

W. Ouyang et al. / Science of the Total Environment 541 (2016) 551–558

555

Table 3 Pesticide loss amounts by province in China (Log kg/year)a.

Fig. 2. Temporal trend of pesticide use by various crops in China from 1990 to 2011.

positive correlation with pesticide use (r = 0.896, p b 0.01). In addition, the rapid growth of gross domestic product also showed significant correlation with pesticide use (r = 0.918, p b 0.01). As the economic development and the decreased proportion of grain crops in the future, the pesticide use in China will increase. 3.3. Potential loss of diffuse pesticides The intensive pesticide use in China posed huge environmental risk due to the potential diffuse pesticide loss. The total pesticide loss in China was estimated about 4.39 × 103 t (log value: 3.64) in 2011 (Table 3), which varied among crop categories and provinces. Grain crops were the largest contributor, and their pesticide loss amount was 2.11 × 103 t (log value: 3.32), accounting for 48.15% of the total loss. GCDL contributed 1.46 × 103 t (log value: 3.16) of the total pesticide loss, a greater amount than GCPL. Because of the higher associated pesticide use intensity, the pesticide loss from VG accounted for 26.37% of the total loss. In addition, the OC and EC were responsible for 13.93% and 10.47% of the loss, respectively. The large amount of pesticide loss occurred in the eastern and central provinces of China. The top six provincial contributors were SD, GD, AH, JX, HuB, and HeN, which accounted for nearly half of the total loss due to higher pesticide loss rates and larger grain output. Based on the national land use distribution at 30 m resolution, the spatial distribution of pesticide loss rate in farmland was analyzed (Fig. 3). Pesticide loss rates in China distributed unevenly because of differences in pesticide use intensity, crop categories, and meteorological and surface conditions. The high pesticide loss rates were concentrated in eastern and southern China. The national average loss rates of insecticides, herbicides and fungicides in 2011 were 3.13, 8.18, 7.81 g·ha−1, respectively. Owing to the variation in pesticide types among provinces, the loss rates of the three types of pesticides observed different spatial distributions. The highest insecticide loss rate was concentrated in the southern provinces and the highest herbicide loss rate occurred in the northeastern provinces. The eastern provinces had the highest fungicide loss rate (see SI Figure S1). 3.4. Risk assessment for Chinese water security Pesticide concentrations in water were estimated with the potential pesticide loss amount and surface water resources of each province. The national average concentration of pesticides in water was estimated 2000 ng/L. The average concentrations of dimethoate, dichlorvos, and malathion were 8 ng/L, 19 ng/L, and 6 ng/L, respectively. These values were similar to the results of other studies in China (Table 4). At the provincial level, the pesticide concentrations in water were higher than the national average in HeN, SD, HeB, BJ, and SH.

Province

GCPLb

GCDL

EC

VG

TP

OC

Total

BJ TJ HeB SX IM LN JL HLJ SH JS ZJ AH FJ JX SD HeN HuB HuN GD GX HN CQ SC GZ YN TB SaX GS QH NX XJ Total

–c 2.28 3.06 1.00 2.59 4.16 3.99 4.46 3.30 4.70 4.48 4.79 4.35 5.07 3.69 3.96 4.74 4.91 4.81 4.39 4.32 3.99 4.41 3.65 4.08 0.48 3.01 – – 2.18 2.04 5.81

3.54 3.41 4.91 4.55 4.51 4.80 4.98 5.16 3.29 4.90 4.18 5.13 3.94 4.20 5.45 5.14 4.74 4.27 4.38 4.23 3.96 4.37 4.86 4.29 4.77 2.48 4.27 3.91 2.68 3.07 3.33 6.16

1.95 2.72 4.16 3.36 3.66 4.00 3.84 3.66 2.43 4.32 3.99 4.58 3.64 4.64 4.80 4.50 4.72 4.57 4.33 4.35 3.91 3.67 4.37 3.79 4.31 1.60 3.38 3.06 2.46 2.15 3.33 5.66

3.45 3.27 4.58 3.79 3.60 4.46 4.17 3.98 3.91 4.92 4.82 4.78 4.59 4.84 5.25 4.81 4.85 4.83 5.09 4.65 4.66 4.36 4.69 4.16 4.52 2.00 3.85 3.47 2.26 2.61 2.94 6.06

– – – – – – – – – 2.92 3.89 3.62 3.68 3.48 2.88 3.07 3.81 3.37 3.23 2.96 1.95 2.71 3.59 3.20 3.83 – 2.75 1.48 – – – 4.68

3.42 2.85 4.54 3.94 3.00 4.35 3.53 3.17 3.10 4.12 4.53 3.92 4.48 4.69 4.77 4.24 4.42 4.49 5.05 4.62 4.57 3.99 4.38 3.54 4.19 1.00 4.24 3.49 1.30 2.68 3.43 5.79

3.96 3.77 5.23 4.72 4.63 5.14 5.11 5.28 4.14 5.39 5.21 5.49 5.04 5.47 5.77 5.42 5.42 5.38 5.54 5.18 5.08 4.85 5.30 4.69 5.17 2.66 4.67 4.18 2.98 3.37 3.90 6.64

a

The unit for the data is a log value of kg per year. GCPL is for grain crops in paddy land, GCDL for grain crops in dry land, EC for economic crops, VG for vegetables, TP for tea plantations, OC for orchards. c “–” means no value. As the pesticide loss for specific crops in a province is 0 kg a log value of 0 works no value. b

The calculated pesticide concentrations in water were compared to detected concentrations in other countries and water quality standards. The results showed that China had lower concentrations of dimethoate, dichlorvos and malathion in water than the Thailand, the U.S. and Iran (Table 4). The pesticide concentrations in water rarely exceeded the Environmental Quality Standards for Surface Water of China and the critical values in the Guidelines for Drinking-water Quality from the WHO (Table 4). Compared with the National Primary Drinking Water Standards from the U.S. EPA, though, China's water security was at risk due to high malathion concentrations.

3.5. Uncertainty analysis of diffuse pesticide assessment The uncertainties (95% confidence interval) in the calculation of pesticide use and loss for various crops were assessed (Table 5). The uncertainties for the six types of crops ranged from −27.5% to 37.3% for pesticide use and − 32.2% to 34.0% for pesticide loss. The coefficients of variation for the six types of crops were 7.20%–9.60% for pesticide use and 7.95%–10.48% for pesticide loss. The uncertainties mainly resulted from the pesticide use intensity factor and pesticide loss coefficients. These parameters were defined as constants throughout the study period to highlight the temporal–spatial variation, but this predefinition contributed to some uncertainties in the spatial differences of pesticide uses for the same crop category in a province. The comprehensive inventory on pesticide use and loss, as well as their temporal–spatial distribution, has provided essential input data for control of water pollution, soil protection and sustainable agricultural management in China.

556

W. Ouyang et al. / Science of the Total Environment 541 (2016) 551–558

Fig. 3. Pesticide loss rates in Chinese farmland in 2011.

4. Discussion 4.1. Implications for sustainable agricultural management The temporal pattern of pesticide use (Fig. 1) showed that the agricultural planting proportions of crops and cultivated area were the most important factors for pesticide use. The proportion of grain crops has declined in Chinese dietary structure, and the proportions of vegetables and fruits have increased in recent decades. The variations of agricultural planting structure also lead to changes in the spatial distribution of pesticide use. From 1990 to 2011, the cultivated areas of vegetables and fruits increased by 209.86% and 128.44%, respectively. During this period, the increase in pesticide use for vegetables and fruits accounted for 50.87% and 26.27% of the total increase in pesticide use, respectively. The cultivated areas of vegetables and orchards will continue to increase in the future due to the climbing market demand. Based on historical patterns, the pesticide application is expected to increase. If environmentally friendly farming practices are not adopted

to reduce the pesticide use intensity, the increase in pesticide use will pose greater risks to water quality. The temporal trend in pesticide use is close correlated to agricultural and economic development (Fig. 2). The Chinese economy has achieved huge advance over the last two decades, which also impacts pesticide use. The change in diet from grains to fruits and vegetables is a consequence of the improved economic level. More affluent farmers can also afford to spend more on pesticides, which is already occurred in the provinces near Beijing, Shanghai and Guangdong (Yu et al., 2013). Rapid urbanization in these areas has induced a shift in agricultural structure because of increased demand for fruit and vegetables, which directly affects the distribution of pesticide use. 4.2. Implications for control of agricultural diffuse pollution Agricultural diffuse pollution attracts more attentions in China nowadays after the control of industrial point source pollution has been achieved (Ministry of Agriculture, 2015). The reduction of pesticide

Table 4 Comparison between pesticide (dimethoate, dichlorvos and malathion) concentrations in Chinese water with other locations or water quality standards (ng/L).a, b, c Location

Year

Dimethoate

Dichlorvos

Malathion

References

Yangtze River, China Taihu Lake, China Wuchuan River, China Haihe River, China Pearl River Estuary, China Mae Sa River, Thailand San Joaquin River, USA Southern Caspian Sea basin, Iran

2003–2004 2003–2004 2000 2008 2000 2007–2008 2007–2008 2008

NDb-16 346 9.84–22.86 1.3–120 ND-8.81 ND-400 74–190 –

2–1552 51.6 8.04–15.10 10–50 0.17–5.80 ND-1100 – 400–1900

Gao et al. (2009) Na et al. (2006) Z. Zhang et al. (2002) and Z.L. Zhang et al. (2002) Gao et al. (2012) Z. Zhang et al. (2002) and Z.L. Zhang et al. (2002) Sangchan et al. (2014) Ensminger et al. (2011) Nasrabadi et al. (2011)

Entire nation

2011

ND-540 11.6 8.31–43.53 10–120 ND-22.8 – – 200–1500 0–123 (Mean = 6) 50,000 * 100

Standards 1a Standards 2 Standards 3

0–41 (mean = 8)

0–136 (mean = 19)

80,000 6000 *

50,000 *c *

This study China WHO U.S. EPA

a Standards 1 = Environmental quality standards for surface water, GB 3838-2002, China; Standards 2 = Guidelines for Drinking-water Quality, WHO; Standards 3 = National Primary Drinking Water Standards, U.S. EPA 816-F-02-013. b ND = not detected. c * = a guideline value has not been established.

W. Ouyang et al. / Science of the Total Environment 541 (2016) 551–558 Table 5 Uncertainties of pesticide use and loss estimation for six types of crops in China.

Crop

GCPL GCDL EC VG TP OC

Uncertainties in pesticide use

Crop

Boundsa

CVb

−19.9%, 28.5% −18.3%, 24.8% −22.0%, 31.6% −23.1%, 28.0% −26.1%, 37.3% −27.5%, 31.9%

7.45% 7.20% 8.49% 7.56% 9.60% 8.72%

GCPL GCDL EC VG TP OC

Uncertainties in pesticide loss Boundsa

CVb

−23.5%, 24.4% −22.4%, 28.1% −23.7%, 32.7% −22.9%, 30.3% −24.9%, 34.0% −32.2%, 29.5%

8.38% 7.95% 9.22% 8.49% 10.48% 9.43%

a Expressed as the lower and upper bounds of a 95% confidence interval around a central estimate. b Coefficient of variation.

use is crucial to control the agricultural diffuse pollution. For the eastern and central provinces with higher pesticide use intensity, the implementation of green crop pest control technologies and promotion of bio-pesticides can reduce pesticide pollution. Crops with low pesticide demand should be advocated in the provinces at risk of water pesticide pollution. As grain crops are in less pesticide demand than vegetable and orchards, it would bring about less pesticide use to enlarge the planting proportion of grain crops and reduce vegetable and orchards. In addition, a pesticide use fee is an appropriate solution for pesticide reduction, and could maintain environment risks below target levels for the lowest cost (Zilberman et al., 1991). Construction of demonstration sites for control of agricultural diffuse pollution should be promoted in major river watersheds and in priority risk areas. These sites could highlight new technologies such as vegetative buffers, constructed wetlands, biochar sorption for the control of agricultural diffuse pesticide pollution (Cole et al., 1997; Zhao et al., 2013). Compared to the national average pesticide use intensity (1.90 kg·ha−1) and loss rate (19.12 g·ha−1), the large areas in northeastern and southeastern China are sensitive to diffuse pesticide pollution. The four provinces in southeastern China (ZJ, JX, GD and HN) are at risk of pesticide loss rates of up to 150% more than the national average. It is mainly because of the higher pesticide use intensities and loss coefficients in the four provinces. The pesticide use intensities in the four provinces rank top five in China. Abundant rainfall in southeastern China results in a higher pesticide loss coefficient than other regions of China. Moreover, elevated pesticide concentrations in farmland have already been detected in these provinces (Zhang et al., 2012). With the help of pesticides and fertilizers, the grain output in China has risen over the last 11 years (National Bureau of Statistics, 2014). However, industrial agriculture also contributes to soil depletion and environmental issues (Tilman et al., 2002). To control diffuse pollution, a series of measures to strengthen agricultural ecological management were listed in a document distributed by the Chinese government in 2015 (Central People's Government, 2015). In this document, the government gives priority to agricultural environmental protection rather than one-sided pursuit of grain output growth. 4.3. Implications for national water risk management Based on the potential pesticide loss rate and the water resources of each province, the diffuse pesticide concentration in water of China was calculated. Compared with national standards and values from other countries, the present pesticide risk to Chinese waters is still under control. Based on higher standard and demands for environmental protection in future, the diffuse pollution simulation and control will shift from the conventional nutrient pollutants and heavy metals to the organic pollution (Ministry of Environmental Protection, 2002; Schulz and Peall, 2001). The diffuse pesticides from tillage practices add pressure to water pollution control efforts in developing countries. The northern provinces of China have fewer water resources than the southern provinces. However, the northern provinces are at risk

557

because of high pesticide concentrations of more than 8000 ng/L, which is four times the national average level. The spatial distribution of pesticide loss rates shows that the potential discharge to rivers in northern China is higher than in southern China (Gao et al., 2008). The water resources per capita in the northern provinces are only one twentieth the world average, which causes more conflicts between agricultural tillage and environmental quality (Guan and Hubacek, 2008). The high concentrations of pesticides in water add more pressures to water security in the provinces of northern China. Therefore, the control of pesticide use is essential for water resource protection and environmental management in northern China. 5. Conclusions The inventory of pesticide use and loss for six primary categories of crops showed that total pesticide use in China increased from 2.09 × 105 t (log value: 5.32) to 3.08 × 105 t (log value: 5.49) during 1990 to 2011. The pesticide use for vegetables, orchards and tea plantations increased rapidly during this period, while pesticide use for economic crops and grain crops in dry and paddy land increased more gradually over the same period. Grain crops in dry land were the major pesticide consumer, especially in northern China. The insecticides were the major type of pesticide used in most parts of China, while herbicides were the major type in the northeast. Pesticide use intensity in the eastern and central provinces was higher than in other areas. The yearly pesticide loss amount was estimated to be 4.39 × 103 t (log value: 3.64) in China, and the highest pesticide loss rate occurred in the eastern and southern provinces. Given the scarcity of water resources in northern China, it was suggested that the area was at risk from high pesticide concentrations. Agricultural and environmental policies have increasingly focused on pesticide loss and pesticide use reduction. The results of this inventory study had direct implications for the rational use and control of pesticides in China. Notes The authors declare no competing financial interest. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (Grant Nos. 41271463 and 41371018) and the Supporting Program of the “Twelfth Five-year Plan” for Science & Technology Research of China (2012BAD15B05). Appendix A. Supplementary data Use inventories of the three main pesticide types by province, cultivated areas of various crops by province, and risk identification of the three main pesticide types in Chinese farmland. Supplementary data associated with this article can be found in the online version, at http://dx.doi.org/10.1016/j.scitotenv.2015.09.120. References Bannwarth, M.A., Sangchan, W., Hugenschmidt, C., Lamers, M., Ingwersen, J., Ziegler, A.D., Streck, T., 2014. Pesticide transport simulation in a tropical catchment by SWAT. Environ. Pollut. 191, 70–79. Bao, L.J., Maruya, K.A., Snyder, S.A., Zeng, E.Y., 2012. China's water pollution by persistent organic pollutants. Environ. Pollut. 163, 100–108. Bressa, G., Sisti, E., Cima, F., 1997. PCBs and organochlorinated pesticides in eel (Anguilla anguilla L.) from the Po Delta. Mar. Chem. 58 (3), 261–266. Carvalho, F.P., 2006. Agriculture, pesticides, food security and food safety. Environ. Sci. Pol. 9 (7), 685–692. Central People's Government (China), 2015. Deepen reform and Innovation to Accelerate Agricultural Modernization. Central People's Government of the People's Republic of China, Beijing (in Chinese).

558

W. Ouyang et al. / Science of the Total Environment 541 (2016) 551–558

Cole, J.T., Baird, J.H., Basta, N.T., Huhnke, R.L., Storm, D.E., Johnson, G.V., 1997. Influence of buffers on pesticide and nutrient runoff from bermudagrass turf. J. Environ. Qual. 26, 1589–1598. Dich, J., Zahm, S.H., Hanberg, A., Adami, H.-O., 1997. Pesticides and cancer. Cancer Causes Control 8 (3), 420–443. Ensminger, M., Bergin, R., Spurlock, F., Goh, K.S., 2011. Pesticide concentrations in water and sediment and associated invertebrate toxicity in Del Puerto and Orestimba Creeks, California, 2007–2008. Environ. Monit. Assess. 175 (1–4), 573–587. Feng, K., Yu, B.Y., Ge, D.M., Wong, M.H., Wang, X.C., Cao, Z.H., 2003. Organo-chlorine pesticide (DDT and HCH) residues in the Taihu Lake Region and its movement in soil–water system I. Field survey of DDT and HCH residues in ecosystem of the region. Chemosphere 50 (6), 683–687. Fu, Z.T., Qi, L.J., 1998. Over-use of pesticide and approaches to reduce pesticide dosage. Trans. Chin. Soc. Agric. Eng. 14 (2), 13–18 (in Chinese with abstract in English). Gao, J., Liu, L., Liu, X., Lu, J., Zhou, H., Huang, S., 2008. Occurrence and distribution of organochlorine pesticides – lindane, p,p′-DDT, and heptachlor epoxide – in surface water of China. Environ. Int. 34 (8), 1097–1103. Gao, J., Liu, L., Liu, X., Zhou, H., Lu, J., Huang, S., 2009. The occurrence and spatial distribution of organophosphorous pesticides in Chinese surface water. Bull. Environ. Contam. Toxicol. 82 (2), 223–229. Gao, J., Liu, L., Liu, X., Lu, J., Hao, H., Yuan, H., Zhou, H., 2012. The organic contamination survey and health risk assessment of 16 source water reservoirs in Haihe River basin. Water Sci. Technol. 65 (6), 998–1006. General Administration (China), 2007. Guideline for Safety Application of Pesticides. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, Beijing (in Chinese). Gillespie, W.E., Czapar, G.F., Hager, A.G., 2011. Pesticide fate in the environment: guide for field inspectors. Contract Report. Guan, D., Hubacek, K., 2008. A new and integrated hydro-economic accounting and analytical framework for water resources: a case study for North China. J. Environ. Manag. 88 (4), 1300–1313. Kellogg, R.L., Nehring, R.F., Grube, A., Goss, D.W., Plotkin, S., 2002. Environmental Indicators of Pesticide Leaching and Runoff From Farm Fields. Springer. Kolpin, D.W., Barbash, J.E., Gilliom, R.J., 1998. Occurrence of pesticides in shallow groundwater of the United States: initial results from the National Water-Quality Assessment Program. Environ. Sci. Technol. 32 (5), 558–566. Lefrancq, M., Imfeld, G., Payraudeau, S., Millet, M., 2013. Kresoxim methyl deposition, drift and runoff in a vineyard catchment. Sci. Total Environ. 442, 503–508. Liu, J., Savenije, H.H.G., 2008. Food consumption patterns and their effect on water requirement in China. Hydrol. Earth Syst. Sci. 12 (3), 887–898. Lucas, J.A., 2011. Advances in plant disease and pest management. J. Agric. Sci. 149 (S1), 91–114. Malone, R.W., Ahuja, L.R., Ma, L., Don Wauchope, R., Ma, Q., Rojas, K.W., 2004. Application of the Root Zone Water Quality Model (RZWQM) to pesticide fate and transport: an overview. Pest Manag. Sci. 60 (3), 205–221. Ministry of Agriculture (China), 2011. National Agricultural and Rural Economic Development Plan During the 12th Five-year Period. Ministry of Agriculture of People's Republic of China, Beijing (in Chinese). Ministry of Agriculture (China), 2015. Implementation Options on Agricultural Nonpoint Source Pollution Control. Ministry of Agriculture of People's Republic of China, Beijing (in Chinese). Ministry of Environmental Protection (China), 2002. Environmental quality standard for surface water. Ministry of Environmental Protection of People’s Republic of China, Beijing (in Chinese). Müller, K., Stenger, R., Rahman, A., 2006. Herbicide loss in surface runoff from a pastoral hillslope in the Pukemanga catchment (New Zealand): role of pre-event soil water content. Agric. Ecosyst. Environ. 112 (4), 381–390. Na, T., Fang, Z., Zhanqi, G., Ming, Z., Cheng, S., 2006. The status of pesticide residues in the drinking water sources in Meiliangwan Bay, Taihu Lake of China. Environ. Monit. Assess. 123 (1–3), 351–370.

Nasrabadi, T., Bidhendi, G.N., Karbassi, A.R., Grathwohl, P., Mehrdadi, N., 2011. Impact of major organophosphate pesticides used in agriculture to surface water and sediment quality (Southern Caspian Sea basin, Haraz River). Environ. Earth Sci. 63 (4), 873–883. National Bureau of Statistics (China), 2014. China Statistical Yearbook. China Statistics Press, Beijing, p. 2014 (in Chinese). Neitsch, S.L., Arnold, J.G., Kiniry, J.R., Williams, J.R., 2011. Soil and Water Assessment Tool Theoretical Documentation Version 2009. Texas Water Resources Institute. Riise, G., Lundekvam, H., Wu, Q.L., Haugen, L.E., Mulder, J., 2004. Loss of pesticides from agricultural fields in SE Norway — runoff through surface and drainage water. Environ. Geochem. Health 26 (2), 269–276. Sangchan, W., Bannwarth, M., Ingwersen, J., Hugenschmidt, C., Schwadorf, K., Thavornyutikarn, P., Pansombat, K., Streck, T., 2014. Monitoring and risk assessment of pesticides in a tropical river of an agricultural watershed in northern Thailand. Environ. Monit. Assess. 186 (2), 1083–1099. Sarigiannis, D.A., Kontoroupis, P., Solomou, E.S., Nikolaki, S., Karabelas, A.J., 2013. Inventory of pesticide emissions into the air in Europe. Atmos. Environ. 75, 6–14. Scholtz, M.T., Voldner, E., McMillan, A.C., Van Heyst, B.J., 2002. A pesticide emission model (PEM) Part I: model development. Atmos. Environ. 36, 5005–5013. Schulz, R., Peall, S.K.C., 2001. Effectiveness of a constructed wetland for retention of nonpoint-source pesticide pollution in the Lourens River catchment, South Africa. Environ. Sci. Technol. 35 (2), 422–426. Short, P., Colborn, T., 1999. Pesticide use in the US and policy implications: a focus on herbicides. Toxicol. Ind. Health 15 (1–2), 240–275. Shu, F., 2009. Prediction and analysis of pesticide demand in China in 2009. Chin. Plant Prot. 29 (3), 40–41 (in Chinese). Tian, H., Liu, K., Zhou, J., Lu, L., Hao, J., Qiu, P., 2014. Atmospheric emission inventory of hazardous trace elements from China's coal-fired power plants—temporal trends and spatial variation characteristics. Environ. Sci. Technol. 48 (6), 3575–3582. Tilman, D., Cassman, K.G., Matson, P.A., Naylor, R., Polasky, S., 2002. Agricultural sustainability and intensive production practices. Nature 418 (6898), 671–677. Yu, H.Y., Li, F.B., Yu, W.M., Li, Y.T., Yang, G.Y., Zhou, S.G., 2013. Assessment of organochlorine pesticide contamination in relation to soil properties in the Pearl River Delta, China. Sci. Total Environ. 447, 160–168. Zhang, Z., Dai, M., Hong, H., Zhou, J.L., Yu, G., 2002a. Dissolved insecticides and polychlorinated biphenyls in the Pearl River Estuary and South China Sea. J. Environ. Monit. 4 (6), 922–928. Zhang, Z.L., Hong, H.S., Zhou, J.L., Yu, G., 2002b. Occurrence and behaviour of organophosphorus insecticides in the River Wuchuan, southeast China. J. Environ. Monit. 4 (4), 498–504. Zhang, A.P., Fang, L., Wang, J.L., Liu, W.P., Yuan, H.J., Jantunen, L., 2012. Residues of currently and never used organochlorine pesticides in agricultural soils from Zhejiang Province, China. J. Agric. Food Chem. 60 (12), 2982–2988. Zhao, J.Z., Luo, Q.S., Deng, H.B., Yan, Y., 2008. Opportunities and challenges of sustainable agricultural development in China. Philos. Trans. R. Soc. B 363 (1492), 893–904. Zhao, Y., Nielsen, C.P., Lei, Y., McElroy, M.B., Hao, J., 2011. Quantifying the uncertainties of a bottom-up emission inventory of anthropogenic atmospheric pollutants in China. Atmos. Chem. Phys. 11 (5), 2295–2308. Zhao, X.C., Ouyang, W., Hao, F.H., Lin, C.Y., Wang, F.L., Han, S., 2013. Properties comparison of biochars from corn straw with different pretreatment and sorption behaviour of atrazine. Bioresour. Technol. 147, 338–344. Zhou, R.B., Zhu, L.Z., Yang, K., Chen, Y.Y., 2006. Distribution of organochlorine pesticides in surface water and sediments from Qiantang River, East China. J. Hazard. Mater. 137 (1), 68–75. Zilberman, D., Schmitz, A., Casterline, G., Lichtenberg, E., Siebert, J.B., 1991. The economice of pesticide use and regulation. Science 253 (5019), 518–522.