- Email: [email protected]
An Integrated Quantitative Method to Simultaneously Monitor Soil Erosion and Non-Point Source Pollution in an Intensive Agricultural Area
Pedosphere 24(5): 674–682, 2014 ISSN 1002-0160/CN 32-1315/P c 2014 Soil Science Society of China Published by Elsevier B.V. and Science Press
An Integrated Quantitative Method to Simultaneously Monitor Soil Erosion and Non-Point Source Pollution in an Intensive Agricultural Area∗1 MA Li1 , BU Zhao-Hong1,∗2 , WU Yong-Hong1 , P. G. KERR2 , S. GARRE3 , XIA Li-Zhong1 and YANG Lin-Zhang1,4 1 State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008 (China) 2 School of Biomedical Sciences, Charles Sturt University, Wagga Wagga, NSW 2678 (Australia) 3 Universit´ e de Li` ege, Gembloux Agro-Bio Tech, AgricultureIsLife, Passage des d´ eport´ es 2, Gembloux 5030 (Belgium) 4 Institute of Agricultural Resources and Environmental Sciences, Jiangsu Academy of Agricultural Sciences, Nanjing 210014 (China)
(Received August 22, 2013; revised May 21, 2014)
ABSTRACT In China, some areas with intensive agricultural use are facing serious environmental problems caused by non-point source pollution (NPSP) as a consequence of soil erosion (SE). Until now, simultaneous monitoring of NPSP and SE is difficult due to the intertwined effects of crop type, topography and management in these areas. In this study, we developed a new integrated method to simultaneously monitor SE and NPSP in an intensive agricultural area (about 6 000 km2 ) of Nanjing in eastern China, based on meteorological data, a geographic information system database and soil and water samples, and identified the main factors contributing to NPSP and SE by calculating the NPSP and SE loads in different sub-areas. The levels of soil total nitrogen (TN), total phosphorus (TP), available nitrogen (AN) and available phosphorus (AP) could be used to assess and predict the extent of NPSP and SE status in the study area. The most SE and NPSP loads occurred between April to August. The most seriously affected area in terms of SE and NPSP was the Jiangning District, implying that the effective management of SE and NPSP in this area should be considered as a priority. The sub-regions with higher vegetation coverage contributed to less SE and NPSP, confirming the conclusions of previous studies, namely that vegetation is an effective factor controlling SE and NPSP. Our quantitative method has both high precision and reliability for the simultaneous monitoring of SE and NPSP occurring in intensive agricultural areas. Key Words:
nutrient load, soil particles, spatial variation, vegetation coverage, water runoff
Citation: Ma, L., Bu, Z. H., Wu, Y. H., Kerr, P. G., Garre, S., Xia, L. Z. and Yang, L. Z. 2014. An integrated quantitative method to simultaneously monitor soil erosion and non-point source pollution in an intensive agricultural area. Pedosphere. 24(5): 674–682.
INTRODUCTION Soil erosion (SE), nitrogen (N) and phosphorus (P) losses occur simultaneously in a continuous, dynamic process and are the main contributors to non-point source pollution (NPSP) (Huang et al., 2001; Wu et al., 2008; Shen et al., 2012). When N and P in soil particles move into downstream waters, some will be dissolved and become soluble N and P (Leon et al., 2001; Braskerud, 2002; Wu et al., 2010a, b). In recent years, the NPSP caused by SE in some intensive agricultural areas in China has become a serious environmental concern (Bu et al., 2005; Jiang et al., 2005; Wu et al., 2010a, b). Apart from the contributions of N and P carried by soil particles during SE, the N and P ∗1 Supported
carried by water runoff is the other major contributor to NPSP (Yan and Tang, 2005; Wu et al., 2011). To date, several quantitative methods have been applied to monitor SE and NPSP. For example, the universal soil loss equation (USLE) and soil and water assessment tool (SWAT) are the most widely used for monitoring SE (Wischmeier and Smith, 1978; Renard et al., 1997) and NPSP (Yong et al., 1989; Ouyang et al., 2010). The USLE is commonly used in China to monitor SE (Xia et al., 2007). However, there have been few studies focusing on the monitoring NPSP with the USLE or modified USLE in China. This is due to the differences in crop land type, topography and management practices between China and USA (Edwin et al., 2010; Ongley et al., 2010). Therefore, a
by the State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences (No. 0812201210), the National Natural Science Foundation of China (No. 41301307), and the Knowledge Innovation Program of Chinese Academy of Sciences (No. ISSASIP1114). ∗2 Corresponding author. E-mail: [email protected].
METHOD TO MONITOR SOIL EROSION AND POLLUTION
goal of this study was to either develop a specific model or further modify the USLE such that it could be used to monitor NPSP, together with SE, in China. During the course of our previous studies (Bu et al., 1999; Xia et al., 2007; Ma et al., 2011), we developed an integrated method for quantitative monitoring of SE and NPSP based on the USLE. The modified method was applied to different areas in China (Shandong, Jiangsu, Yunnan, Fujian and Jiangxi provinces) with a total monitoring area of about 200 000 km2 . The results showed that the precision of soil loss monitoring using this method reaches 80% to 90% in nonintensive agricultural areas (Bu et al., 2005; Jiang et al., 2005). However, for the simultaneous monitoring of SE and NPSP in some intensive agricultural areas such as Nanjing area, this integrated method needs to be further developed. Nanjing is an important economic, cultural and transportation center in eastern China. Areas with intensive agricultural use surround the city center. Rapid urbanization during recent years caused deforestation and increased reclamation of sloping land for agriculture, which resulted in increased soil erosion and water pollution (Lin et al., 2008). Thus, the development of a quantitative monitoring and analysis system for SE and NPSP is required to provide a sound basis for a better management of city development and land use. The objectives of this study were: 1) to map SE
Fig. 1
675
and NPSP in the Nanjing area using the integrated quantitative model to assess SE and NPSP simultaneously, based on the meteorological data, a geographic information system (GIS) database and chemical analysis of soil and water samples collected from 2001 to 2010 in the Nanjing area, and 2) to identify the main factors contributing to NPSP and SE by calculating the NPSP and SE loads in different sub-areas. The findings are expected to yield a rational basis for future land management policy with respect to SE and NPSP in similar intensive agricultural areas. MATERIALS AND METHODS Study area description The Nanjing area (31◦ 14 –32◦ 36 N, 118◦ 32 –119◦ 14 E) is located in the Yangtze River Delta, eastern China and covers about 6 590 km2 . The mountains and down lands, plains and basins, and surface waters are comprise approximately 58.4%, 41.6% and 10.2% of the total Nanjing area, respectively. The region belongs to the north subtropical monsoon climatic zone, with an annual average temperature of 17.8 ◦ C. The sunshine duration is 1 687 h and the annual average rainfall is 1 034 mm. Four rivers of Yangtze River, Qinhuai River, Shuiyang River and Chuhe River traverse the region which also includes Gucheng Lake and Shijiu Lake, with the total water capacity of 428 million m3 (Lin et al., 2008) (Fig. 1a).
Water system (a) and soil sampling points (b) in Nanjing area, eastern China.
676
L. MA et al.
Quantitative monitoring method The proposed SE and NPSP quantitative monitoring method consists of nine parts, including SE and NPSP quantitative monitoring models, formulae and arithmetics of the factor models and related software, precision evaluation of digital elevation model (DEM), practical application of the fast update of GIS using global positioning system (GPS) measurement, and database building and inquiry (Bu et al., 1999, 2005; Ma et al., 2011). The types of models used in this integrated method are as the same as the USLE and revised USLE (RUSLE) models used in America (Wischmeier and Smith, 1978; Renard et al., 1997), but the formulae and arithmetics of each model factor were built based on large quantities of measured data and grid pixel calculation in the erosion area in China. This approach resulted in significantly different formulae, arithmetics and its related software from the USLE and RUSLE model. The proposed SE monitoring model (Eq. 1) retains the factors of the USLE to calculate annual sheet and rill erosion from a hillslope; however, changes have been made for some factors. Ai = f ·Ri ·Ki ·LS i ·CP i
(1)
where Ai (t km−2 year−1 ) is the amount of annual soil loss for a grid pixel i (i = 1, . . . , n) of the area under consideration having a certain soil type; Ri (MJ mm ha−1 h−1 year−1 ) is the rainfall erosivity factor of grid pixel i; Ki (t ha h ha−1 MJ−1 mm−1 ) is the soil erosion factor of grid pixel i; LS i (a dimensionless unit) is the product of slope length factor L and slope factor S of grid pixel i; CP i (also a dimensionless unit) is the product of vegetation cover factor C and soil conservation factor P of grid pixel i; and f is the conversion coefficient (normally it is 100, which transforms a unit of soil erosion from t ha−1 year−1 to t km−2 year−1 ). NPSP of the soil nutrient monitoring model is shown in Eq. 2: Wi,j = Ai ·cj
(2)
where Wi,j (kg km−2 ) is the annual increase in the amount of soil nutrient of the jth species (i.e., TN, TP, AN, and AP) for a grid pixel i; and cj (g kg−1 for TN and TP, and mg kg−1 for AN and AP) is the soil nutrient content of the jth species in the monitored soil use types. The main changes to the RUSLE approach incorporated into the proposed model are as follows (Bu et al., 1997):
R factor: R factor is the erosion power factor which was a major determinant of difference of the erosion amount in different regions or periods. Using the rainfall and runoff observation data for many years in various typical rainfall areas in China, a new arithmetic of R factor was obtained (Eq. 3): R = 2.1803Pf ·I30 − 2.6810I30
(3)
where Pf (mm) is the total rainfall of each month in the flood season; and I30 (cm h−1 ) is the annual value of the maximum rainfall intensity in 30 min. LS factor: LS factor is the acceleration factor of erosion power. As the slopes of farmlands in China are usually larger than those in USA, arithmetic of slope factor S in the USLE equation was not suitable for application in China. Using a large number of measured data for many years, we had made a large improvement on S factor formula (Eq. 4), and a series of formulae were incorporated into the software. To meet the requirements of quantitative remote sensing (RS) of soil erosion based on pixels, equations for slope factor S and slope length factor L of each pixel were made some improvements (Eqs. 5 and 6): S = 0.743 × 1.0595αi
(4)
αi = f (hi , hj , di )
(5)
Li = f (di , αi , mi )
(6)
where αi is the most appropriate slope value in the adjacent eight directions of the ith pixel; hi (hj ) is elevation value of the ith (jth) pixel; di is the side or diagonal length of the ith pixel; and mi is the index value changes with slope. CP factor: CP factor is the inhibiting factor of erosion power which reflects the effect of soil and water conservation. CP factor value is less than or equal 1. C factor was established using the observed data of the natural vegetation area. The equation is as follow: ⎧ ⎪ ⎨ 0.4149 − 0.0052C1 0.4399 − 0.0058C2 (7) C= ⎪ ⎩ 0.4500 − 0.0079C3 where C1 , C2 , and C3 (%) are the maximum, secondlargest, and average vegetation coverage, respectively, in a vegetation growth period. In generally, the value of P factor is l for the natural vegetation area and slope land, 0.01 for the level terrace, and 0.02 to 0.7 for the farmland with different control measures. The SE levels were evaluated based on the annual soil erosion module (t km−2 year−1 ) according
METHOD TO MONITOR SOIL EROSION AND POLLUTION
to the National Standard for Classification and Gradation of Soil Erosion (SL 190-2007) (Table I). The standard was also employed to assess NPSP. The R factor was obtained from rainfall data of various precipitation stations from 2001 to 2010. According to location, terrain and climate, four representative precipitation stations (Xiaguan, Luhe, Yuetang Reservoir and Tianshengqiao) were selected. The maps containing the administrative boundaries, DEM, and soil and land use types of Nanjing were extracted from the “Nanjing soil and water conservation geographic information system (NJSWCGIS)” and updated according to the latest boundaries of districts and towns, land use types and DEM map. Satellite data of the Nanjing region were obtained from a charge-coupled device received on April 3, 2008 and thematic mapper images received on April 17, 2000. TABLE I Soil erosion and non-point source pollution (NPSP) levels and classifications based on annual soil erosion module according to the National Standard for Classification and Gradation of Soil Erosion (SL 190-2007) Soil erosion module 103
km−2
× t < 0.5 0.5–2.5 2.5–5.0 5.0–8.0 8.0–15.0 > 15.0
Level
Classification
Insignificant Low Moderate High Severe Extremely severe
Class Class Class Class Class Class
year−1 I II III IV V VI
Sampling and analyses The soil samples were collected from 45 sampling sites (257 soil samples) distributed over the Nanjing area (Fig. 1b). Freshly collected samples were air-dried and the soil was passed through 20- and 100-mesh sieves after plant roots and other debris were removed. After that, the soil samples were stored in plastic bags for chemical analysis. Soil TN was determined using the Kjeldahl method (Bremner and Mulvaney, 1982). Soil TP was determined by digestion in H2 SO4 HClO4 and analyzed by colorimetry. Soil AN was determined using the alkaline hydrolysis diffusion method (Lu et al., 2000). Soil AP was determined by sample extraction with 0.5 mol L−1 NaHCO3 , and analyzed by colorimetry (Lu et al., 2000). The soil information was used as an input for the quantitative prediction of SE and NPSP in the Nanjing area. Statistical analysis Analysis of variance was conducted by SPSS ver-
677
sion 16.0 to process the quantitative monitoring data. Data fitting processes and graphs were prepared using the ArcGIS 9.2 and Excel 2010 software. RESULTS AND DISCUSSION SE in Nanjing area from 2001 to 2010 To investigate the spatial pattern of SE from 2001 to 2010 in the Nanjing area, the 10-year average of SE was analyzed and compared with the standards in Table I. The results showed that the SE of a region with the area of 896.68 km2 exceeded class II. This region corresponds to hill and plain with the area of 881.48 and 15.20 km2 , respectively. The amount of SE involved was 2.66 × 106 t, of which hill and plain area were 2.57 × 106 and 8.34 × 104 t, respectively. The overall SE module of this region was 2 441 t km−2 year−1 , which is at the upper end of the range for class II. The SE module of the hill area was 2 469 t km−2 year−1 and that of the plain 861 t km−2 year−1 . The results indicate clearly that SE in the Nanjing area occurs primarily in the hill area, as expected, representing about 22% of the total area investigated. Some suburbs with rapid economic development (such as Jiangning District) represented about 35% of the total SE amount. This is because the uncovered slopes of hills accelerate the movement of loose soil particles together with the nutrients bound to them, increasing soil losses (Rejani and Yadukumar, 2010) and NPSP. Moreover, the dredging activities during the industrial development in the Jiangning District also loosened soil structure, causing additional loss. NPSP in Nanjing area from 2001 to 2010 Quantitative monitoring of average soil TN, TP, AN, and AP pollution sources in the Nanjing area from 2001 to 2010 is summarized in Table II. The NPSP load is defined as the sum of TN and TP (Ongley et al., 2010; Wu et al., 2011). It is obvious that the NPSP occurred mainly in the hill area. The pollution modules of soil available nutrients (AN and AP) were small, but those of soil TN and TP were larger and therefore of more serious concern. In the hill area, the pollution module of soil total nutrients was classified to the moderate level (class III). This pollution module is equivalent to 21–30 kg ha−1 of N and P carried by soils entering into downstream water bodies. Temporal variation of NPSP during 2001–2010 Using the adapted model, we obtained annual changes of pollution load, module and area of TN, AN, TP and AP in the Nanjing area as a consequence of diffe-
678
L. MA et al.
TABLE II Average non-point source pollution (NPSP) area, load and module in the Nanjing area from 2001 to 2010 NPSPa)
Pollution area Total
Pollution load
Hill area
Plain area
Total
km2 TN AN TP AP a) TN
1 113.4 109.1 621.7 228.2
1 087.5 109.0 615.8 227.7
25.9 0.1 5.9 0.5
3 765.5 289.3 1 704.3 53.0
Pollution module
Hill area t 3 660.5 281.6 1 642.5 50.9
Plain area
Total
Hill area km−2
105.0 7.7 61.9 2.1
2 935.0 1 154.4 2 085.0 123.0
kg 2 984.0 1 154.5 2 097.0 123.1
Plain area
year−1 867.0 782.5 844.0 82.7
= total N; AN = available N; TP = total P; AP = available P.
rences in local climate and surface conditions over the decade of 2001–2010 (Figs. 2 and 3). In 2009 and even clearer in 2003, the annual nutrient load, the pollution module and the pollution area with erosion module exceeding class II were higher than the other years. The minima of these three variables were found in 2001. According to the monthly-distributed data, SE and NPSP mainly occurred from May to October in each year. This is the wet season of the study area, since April to October account for about 65% of the annual rainfall. The SE and NPSP show strong correlations with rainfall, as reported in the previous studies (Lee et al., 2010; Peng and Wang, 2012). Fig. 3 Changes of non-point source pollution (NPSP) area over low NPSP level (class II) in the Nanjing area from 2001 to 2010. TN = total N; TP = total P; AN = available N; AP = available P.
Fig. 2 Changes of non-point source pollution load and module in the Nanjing area from 2001 to 2010. TN = total N; TP = total P; AN = available N; AP = available P.
In the decade of 2001–2010, vegetation coverage in the Nanjing area has increased due to actions of the local authorities. Increasing vegetation cover has probably enlarged the buffer capacity of certain areas for block sediments, nutrients and other contaminants transported with runoff and SE (Duchemin and Hogue, 2009). Effective measures conducted by the local authorities to control agricultural NPSP include establishment of riparian vegetation buffers and restoration of forest on cultivated and sloping arable land. Both measures help to reduce nutrient flux to downstream waters (Tian et al., 2010; Zhang et al., 2011). In order to quantitatively assess the effect of vegetation management from 2001 onwards, we collected 4 years (1997–2000) of monitoring data before 2001. This 4year dataset can be regarded as the baseline. Figs. 4 and 5 show a comparison of the baseline with the data of 2001–2010. Focusing on N pollution, the maximum soil TN pollution load, module and area exceeding NPSP class II were identified in 2003, and then in 2009, with the minimum values in 2001. The average soil TN pollution
METHOD TO MONITOR SOIL EROSION AND POLLUTION
679
Fig. 4 Average values of non-point source pollution (NPSP) areas at different NPSP levels for the 4-year baseline (1997–2000) and 10-year study period (2001–2010) in the Nanjing area. TN = total N; TP = total P; AN = available N; AP = available P.
Fig. 5 Average values of non-point source pollution (NPSP) loads in the areas at different NPSP levels for the 4-year baseline (1997–2000) and 10-year study period (2001–2010) in the Nanjing area. TN = total N; TP = total P; AN = available N; AP = available P.
680
module for most years reached class III in most years, excluding 2001. The area with the average soil TN pollution exceeding class II was about 1 113 km2 , with 7 years of the study accounting for more than 1 000 km2 . Compared to the baseline period, the 10-year averages of soil TN pollution load (3 267.7 t), module (2 935 t km−2 year−1 ) and pollution area (1 113.4 km2 ) exceeding class II significantly decreased (P < 0.05). The soil TN pollution areas at low (class II) and high NPSP levels (class V) significantly decreased by 33.3% and 48.3%, respectively (P < 0.05). Similarly, the maxima of soil AN pollution load, module and area exceeding class II were found in 2003, and then in 2009, with the minimum values found in 2001. For each year of the study, the average soil AN pollution module mostly reached the low NPSP level (class II). The 10-year soil AN pollution load, module and pollution area exceeding class II significantly decreased from 231.26 to 125.90 t, 1 360 to 1 165 t km−2 year−1 and 170.08 to 109.07 km2 , respectively (P < 0.05). Soil TP and AP evolutions showed similar trends with the maxima occurring in 2003, and then again in 2009. The minima were found in 2001. The average soil TP pollution module for each year (excluding 2003) reached the low level (class II), corresponding to about 2 038.4 t km−2 year−1 . Compared with the 4-year baseline values, the 10-year average of soil TP pollution load (1 296.3 t), module (2 085 t km−2 year−1 ) and pollution area (621.69 km2 ) exceeding class II significantly decreased (P < 0.05). The 10-year soil TP pollution area at the high (class IV), severe (class V) and extremely severe (class VI) levels significantly decreased by 34.6%, 46.8% and 64.0%, respectively (P < 0.05) as compared to the baseline. The average soil AP pollution module for most years reached the insignificant level (class I), and the 10-year average of AP (123 t km−2 year−1 ) was also significantly lower than the 4year average (150 t km−2 year−1 ) (P < 0.05). Many studies showed that soil and water conservation measures demonstrate positive effects on soil fertility and soil nutrient loss reduction in China (Lin et al., 2009; Dur´an Zuazo et al., 2011; Nunes et al., 2011; Deng et al., 2012; Phan Ha et al., 2012). In this study, we could confirm the positive effect of vegetative cover on erosion reduction and, therefore, also on the NPSP decrease of N and P, even though the nutrient load has increased during the study period in many soil types of the Nanjing area. Spatial variation of NPSP during 2001–2010 NPSP not only changed in different years or months, but also significantly changed in different dis-
L. MA et al.
tricts of Nanjing. To analyze the spatial variation of NPSP in the Nanjing area, four variables were chosen, including soil pollution load, pollution module and the pollution areas exceeding classes II and IV. The concept of “management degree” was used to evaluate the effect of control measures on SE and NPSP between 2001 and 2010. The management degree is calculated for each of the four variables by taking the difference of average values of the variable in two periods (1997– 2000 and 2001–2010) and then dividing the difference by the average value in 1997–2000. By ranking the results for each area, the evolutions of SE and NPSP and the effectiveness of control measures can be seen for different districts of Nanjing area (Table III). The average management degrees of the pollution load and pollution areas exceeding classes II and IV for soil TN and AN exceeded 30% in the whole Nanjing area. However, the average management degree of the pollution area over class II for soil TP and AP was less than 30%. The effect of vegetative cover was thus stronger for N than for P. The largest effect of the measures taken to reduce SE and NPSP occurred in the Luhe District (data not shown). The management degree was in the order of Luhe District > Pukou District > Lishui District > Qixia District > Downtown District > Gaochun District > Jiangning District > Yuhua Distict. Validation of the simultaneous model To evaluate the reliability of the quantitative monitoring results of SE and NPSP, the model obtained in this study were validated in the field and tested using hydrological data collected in 2002 and 2003. The results showed that the monitoring precisions of TN in 2002 and 2003 were 71.6% and 84.7%, respectively. The average precision was 78.1%. The monitoring precision of TP in 2002 and 2003 were 92.5% and 83.1%, respectively. The average precision was 87.8%. Previous studies also show that the simultaneous determination of SE and NPSP by the proposed method is reliable and credible (Bu et al., 1999; Xia et al., 2007). The revised integrated monitoring method has better recognition ability for spatial variation of SE and NPSP with the support of “3S” (GIS, RS, GPS) technology. The new integrated method is also characterized by convenient, efficient, and quickly updated spatial data (Jiang et al., 2005). CONCLUSIONS The proposed quantitative method to assess NPSP and SE, simultaneously based on soil samples, DEM, land use and soil layers in a GIS environment, is trust-
METHOD TO MONITOR SOIL EROSION AND POLLUTION
681
TABLE III Effect of control measures on soil total nitrogen (TN) and total phosphorus (TP) pollution, evaluated by the concept of “management degree”, in different districts of Nanjing area Pollution
TN
TP
District
Luhe Pukou Yuhua Qixia Jiangning Downtown Lishui Gaochun Total Luhe Pukou Yuhua Qixia Jiangning Downtown Lishui Gaochun Total
Total area
km2 1 479.1 909.2 131.6 371.6 1 580.3 258.5 1 068.8 790.9 6 589.9 1 479.1 909.2 131.6 371.6 1 580.3 258.5 1 068.8 790.9 6 589.9
Management degree Pollution module
Pollution load
Pollution area over class II
Pollution area over class IV
3.9 8.3 1.3 16.2 4.1 42.3 23.3 5.8 7.9 7.6 14.3 9.1 19.1 5.5 38.9 23.8 16.7 11.5
36.8 45.4 7.3 28.8 24.2 54.7 45.3 37.2 34.8 38.3 46.8 6.6 28.7 22.1 49.6 43.7 37.2 34.7
% 36.6 42.7 5.8 13.0 21.8 27.2 34.8 31.5 30.9 35.4 40.2 −5.7 7.1 17.8 23.9 36.0 21.9 27.6
29.1 48.0 −15.0 1.2 18.8 44.6 61.4 43.0 31.0 45.1 59.2 20.9 30.5 25.2 74.0 73.3 59.5 43.6
worthy and provides scientific support for local SE and NPSP management in the Nanjing area. We confirmed in this study that the method has a high precision, reliability and credibility. The calculated TN, TP, AN, and AP levels in different types of soils allow assembling maps and temporal evolution of soil nutrient status in the Nanjing area. The positive effects of erosion control measures on SE and NPSP in the study area were significant over the decade of 2001–2010 (P < 0.05). The results of our study confirmed that the implementation of vegetative cover is an effective measure to control SE and NPSP. Therefore, simultaneous spatio-temporal estimation of SE and NPSP with the proposed integrated model can be applied for management of the areas with intensive agricultural use. REFERENCES Braskerud, B. C. 2002. Factors affecting phosphorus retention in small constructed wetlands treating agricultural non-point source pollution. Ecol. Eng. 19: 41–61. Bremner, J. M. and Mulvaney, C. S. Nitrogen—Total. 1982. In Page, A. L., Miller, R. H. and Keeney, D. R. (eds). Methods of Soil Analysis. Part 2. Chemical and Microbiological Properties. American Society of Agronomy, Madison. pp. 595–624. Bu, Z. H., Jiang, X. S., Yang, L. Z. and Zhang, Z. X. 2005. The experiment of optimum methods of renewing GIS’s data by GPS solid survey in the soil erosion fixed quantity monitoring. Acta Pedol. Sin. (in Chinese). 42: 712–719. Bu, Z. H., Sun, J. Z., Dong, Q. R. and Liu, S. Q. 1999. Study
on quantitative remote sensing method for soil erosion losses and its application in Shandong Province, China. Acta Pedol. Sin. (in Chinese). 36: 1–8. Bu, Z. H., Sun, J. Z., Zhou, F. J., Tang, W. L. and Xi, C. F. 1997. A study on quantitative remote sensing method of soil erosion and its application. Acta Pedol. Sin. (in Chinese). 34: 235–245. Deng, L., Shangguan, Z. P. and Rui, L. 2012. Effects of the grainfor-green program on soil erosion in China. Int. J. Sediment Res. 27: 120–127. Duchemin, M. and Hogue, R. 2009. Reduction in agricultural non-point source pollution in the first year following establishment of an integrated grass/tree filter strip system in southern Quebec (Canada). Agr. Ecosyst. Environ. 131: 85– 97. Dur´ an Zuazo, V. H., Rodr´ıguez Pleguezuelo, C. R., Martin Peinado, F. J., de Graaff, J., Francia Mart´ınez, J. R. and Flanagan, D. C. 2011. Environmental impact of introducing plant covers in the taluses of terraces: Implications for mitigating agricultural soil erosion and runoff. Catena. 84: 79– 88. Edwin, D. O., Zhang, X. L. and Yu, T. 2010. Current status of agricultural and rural non-point source pollution assessment in China. Environ. Pollut. 158: 1159–1168. Huang, M. X., Zhang, S., Tang, Y. J. and Chen, X. B. 2001. Nitrogen losses from farm runoff under simulated rainfall conditions. Soil Environ. Sci. (in Chinese). 10: 6–10. Jiang, X. S., Bu, Z. H., Yang, L. Z., Xia, L. Z. and Pan, J. J. 2005. Integrated quantitative monitoring of soil loss and water pollution—A case study of Fangbian valley in Jiangsu Province. Acta Pedol. Sin. (in Chinese). 42: 529–536. Lee, C. S., Chang, C. H., Wen, C. G. and Chang, S. P. 2010. Comprehensive nonpoint source pollution models for a freerange chicken farm in a rural watershed in Taiwan. Agr. Ecosyst. Environ. 139: 23–32. Leon, L. F., Soulis, E. D., Kouwen, N. and Farquhar, G. J. 2001.
682
Nonpoint source pollution: a distributed water quality modeling approach. Water Res. 35: 997–1007. Lin, C. W., Tu, S. H., Huang, J. J. and Chen, Y. B. 2009. The effect of plant hedgerows on the spatial distribution of soil erosion and soil fertility on sloping farmland in the purplesoil area of China. Soil Till. Res. 105: 307–312. Lin, J., Zhang, J. C., Wu, Y. M., Peng, G. L., Shi, K. F. and Li, H. D. 2008. The current situation and causes of water and soil losses in Nanjing and its conservation strategies. J. Nanjing Forest. U. (in Chinese). 32: 44–46. Lu, R. K. 2000. Analytical Methods of Soil Agrochemistry (in Chinese). China Agricultural Science and Technology Press, Beijing. Ma, L., Bu, Z. H., Peng, G. L., Xia, L. Z., Li, Y. D., Yang, L. Z., Jiang, X. S., Huang, R. and Wu, Y. M. 2011. Temporal and spatial variation of soil and water loss and its control in Nanjing area from 2001 to 2008. Acta Pedol. Sin. (in Chinese). 48: 683–692. Nunes, A. N., de Almeida, A. C. and Coelho, C. O. A. 2011. Impacts of land use and cover type on runoff and soil erosion in a marginal area of Portugal. Appl. Geogr. 31: 687–699. Ongley, E. D., Zhang, X. L. and Yu, T. 2010. Current status of agricultural and rural non-point source pollution assessment in China. Environ. Pollut. 158: 1159–1168. Ouyang, W., Skidmore, A. K., Toxopeus, A. G. and Hao, F. H. 2010. Long-term vegetation landscape pattern with nonpoint source nutrient pollution in upper stream of Yellow River basin. J. Hydrol. 389: 373–380. Peng, T. and Wang, S. J. 2012. Effects of land use, land cover and rainfall regimes on the surface runoff and soil loss on karst slopes in southwest China. Catena. 90: 53–62. Phan Ha, H. A., Huon, S., des Tureaux, T. H., Orange, D., Jouquet, P., Valentin, C., Rouw, A. D. and Duc, T. T. 2012. Impact of fodder cover on runoff and soil erosion at plot scale in a cultivated catchment of North Vietnam. Geoderma. 177178: 8–17. Rejani, R. and Yadukumar, N. 2010. Soil and water conservation techniques in cashew grown along steep hill slopes. Sci. Hortic. 126: 371–378. Renard, K. G., Foster, G. R., Weesies, G. A., Mccool, D. K. and Yoder, D. C. 1997. Predicting Soil Erosion by Water—A Guide to Conservation Planning with the Revised Soil Loss Equation (RUSLE). Agriculture Handbook No. 703. US Department of Agriculture, Washington, D.C.
L. MA et al.
Shen, Z. Y., Liao, Q., Hong, Q. and Gong, Y. W. 2012. An overview of research on agricultural non-point source pollution modeling in China. Sep. Purif. Technol. 84: 104–111. Tian, Y. W., Huang, Z. L. and Xiao, W. F. 2010. Reductions in non-point source pollution through different management practices for an agricultural watershed in the Three Gorges Reservoir Area. J. Environ. Sci. 22: 184–191. Wischmeier, W. H. and Smith, D. D. 1978. Predicting Rainfall Erosion Losses—A Guide to Conservation Planning. Agriculture Handbook No. 537. US Department of Agriculture, Washington, D.C. Wu, Y. H., Hu, Z. Y. and Yang, L. Z. 2008. Integrated technology to control non-point pollution in resident-farmland area of Dianchi Lake watershed, China. Ecol. Sci. (in Chinese). 27: 346–350. Wu, Y. H., Hu, Z. Y. and Yang, L. Z. 2010a. Hierarchical ecorestoration: A systematic approach to removal of COD and dissolved nutrients from an intensive agricultural area. Environ. Pollut. 158: 3123–3129. Wu, Y. H., Kerr, P. G., Hu, Z. Y. and Yang, L. Z. 2010b. Ecorestoration: Simultaneous nutrient removal from soil and water in a complex residential-cropland area. Environ. Pollut. 158: 2472–2477. Wu, Y. H., Xia, L. Z., Hu, Z. Y., Liu, S. Z., Liu, H. B., Nath, B., Zhang, N. M. and Yang, L. Z. 2011. The application of zerowater discharge system in treating diffuse village wastewater and its benefits in community afforestation. Environ. Pollut. 159: 2968–2973. Xia, L. Z., Bu, Z. H., Zhang, Z. X., Yang, L. Z., Yang, X. Y. and Jiang, X. S. 2007. Application of integrated method to assessment of soil and nutrient losses in Miyun Reservoir Basin. Adv. Earth Sci. (in Chinese). 22: 241–248. Yan, B. X. and Tang, J. 2005. Study on black soil erosion rate and the transformation of soil quality influenced by erosion. Geogr. Res. (in Chinese). 24: 499–506. Yong, R. A., Onstad, C. A., Bosch, D. D. and Anderson, W. P. 1989. AGNPS: A nonpoint-source pollution model for evaluating agricultural watersheds. J. Soil Water Conserv. 44: 168–173. Zhang, P., Liu, Y. H., Pan, Y. and Yu, Z. R. 2011. Land use pattern optimization based on CLUE-S and SWAT models for agricultural non-point source pollution control. Math. Comput. Model. 58: 588–595.
An Integrated Quantitative Method to Simultaneously Monitor Soil Erosion and Non-Point Source Pollution in an Intensive Agricultural Area∗1 MA Li1 , BU Zhao-Hong1,∗2 , WU Yong-Hong1 , P. G. KERR2 , S. GARRE3 , XIA Li-Zhong1 and YANG Lin-Zhang1,4 1 State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences, Nanjing 210008 (China) 2 School of Biomedical Sciences, Charles Sturt University, Wagga Wagga, NSW 2678 (Australia) 3 Universit´ e de Li` ege, Gembloux Agro-Bio Tech, AgricultureIsLife, Passage des d´ eport´ es 2, Gembloux 5030 (Belgium) 4 Institute of Agricultural Resources and Environmental Sciences, Jiangsu Academy of Agricultural Sciences, Nanjing 210014 (China)
(Received August 22, 2013; revised May 21, 2014)
ABSTRACT In China, some areas with intensive agricultural use are facing serious environmental problems caused by non-point source pollution (NPSP) as a consequence of soil erosion (SE). Until now, simultaneous monitoring of NPSP and SE is difficult due to the intertwined effects of crop type, topography and management in these areas. In this study, we developed a new integrated method to simultaneously monitor SE and NPSP in an intensive agricultural area (about 6 000 km2 ) of Nanjing in eastern China, based on meteorological data, a geographic information system database and soil and water samples, and identified the main factors contributing to NPSP and SE by calculating the NPSP and SE loads in different sub-areas. The levels of soil total nitrogen (TN), total phosphorus (TP), available nitrogen (AN) and available phosphorus (AP) could be used to assess and predict the extent of NPSP and SE status in the study area. The most SE and NPSP loads occurred between April to August. The most seriously affected area in terms of SE and NPSP was the Jiangning District, implying that the effective management of SE and NPSP in this area should be considered as a priority. The sub-regions with higher vegetation coverage contributed to less SE and NPSP, confirming the conclusions of previous studies, namely that vegetation is an effective factor controlling SE and NPSP. Our quantitative method has both high precision and reliability for the simultaneous monitoring of SE and NPSP occurring in intensive agricultural areas. Key Words:
nutrient load, soil particles, spatial variation, vegetation coverage, water runoff
Citation: Ma, L., Bu, Z. H., Wu, Y. H., Kerr, P. G., Garre, S., Xia, L. Z. and Yang, L. Z. 2014. An integrated quantitative method to simultaneously monitor soil erosion and non-point source pollution in an intensive agricultural area. Pedosphere. 24(5): 674–682.
INTRODUCTION Soil erosion (SE), nitrogen (N) and phosphorus (P) losses occur simultaneously in a continuous, dynamic process and are the main contributors to non-point source pollution (NPSP) (Huang et al., 2001; Wu et al., 2008; Shen et al., 2012). When N and P in soil particles move into downstream waters, some will be dissolved and become soluble N and P (Leon et al., 2001; Braskerud, 2002; Wu et al., 2010a, b). In recent years, the NPSP caused by SE in some intensive agricultural areas in China has become a serious environmental concern (Bu et al., 2005; Jiang et al., 2005; Wu et al., 2010a, b). Apart from the contributions of N and P carried by soil particles during SE, the N and P ∗1 Supported
carried by water runoff is the other major contributor to NPSP (Yan and Tang, 2005; Wu et al., 2011). To date, several quantitative methods have been applied to monitor SE and NPSP. For example, the universal soil loss equation (USLE) and soil and water assessment tool (SWAT) are the most widely used for monitoring SE (Wischmeier and Smith, 1978; Renard et al., 1997) and NPSP (Yong et al., 1989; Ouyang et al., 2010). The USLE is commonly used in China to monitor SE (Xia et al., 2007). However, there have been few studies focusing on the monitoring NPSP with the USLE or modified USLE in China. This is due to the differences in crop land type, topography and management practices between China and USA (Edwin et al., 2010; Ongley et al., 2010). Therefore, a
by the State Key Laboratory of Soil and Sustainable Agriculture, Institute of Soil Science, Chinese Academy of Sciences (No. 0812201210), the National Natural Science Foundation of China (No. 41301307), and the Knowledge Innovation Program of Chinese Academy of Sciences (No. ISSASIP1114). ∗2 Corresponding author. E-mail: [email protected].
METHOD TO MONITOR SOIL EROSION AND POLLUTION
goal of this study was to either develop a specific model or further modify the USLE such that it could be used to monitor NPSP, together with SE, in China. During the course of our previous studies (Bu et al., 1999; Xia et al., 2007; Ma et al., 2011), we developed an integrated method for quantitative monitoring of SE and NPSP based on the USLE. The modified method was applied to different areas in China (Shandong, Jiangsu, Yunnan, Fujian and Jiangxi provinces) with a total monitoring area of about 200 000 km2 . The results showed that the precision of soil loss monitoring using this method reaches 80% to 90% in nonintensive agricultural areas (Bu et al., 2005; Jiang et al., 2005). However, for the simultaneous monitoring of SE and NPSP in some intensive agricultural areas such as Nanjing area, this integrated method needs to be further developed. Nanjing is an important economic, cultural and transportation center in eastern China. Areas with intensive agricultural use surround the city center. Rapid urbanization during recent years caused deforestation and increased reclamation of sloping land for agriculture, which resulted in increased soil erosion and water pollution (Lin et al., 2008). Thus, the development of a quantitative monitoring and analysis system for SE and NPSP is required to provide a sound basis for a better management of city development and land use. The objectives of this study were: 1) to map SE
Fig. 1
675
and NPSP in the Nanjing area using the integrated quantitative model to assess SE and NPSP simultaneously, based on the meteorological data, a geographic information system (GIS) database and chemical analysis of soil and water samples collected from 2001 to 2010 in the Nanjing area, and 2) to identify the main factors contributing to NPSP and SE by calculating the NPSP and SE loads in different sub-areas. The findings are expected to yield a rational basis for future land management policy with respect to SE and NPSP in similar intensive agricultural areas. MATERIALS AND METHODS Study area description The Nanjing area (31◦ 14 –32◦ 36 N, 118◦ 32 –119◦ 14 E) is located in the Yangtze River Delta, eastern China and covers about 6 590 km2 . The mountains and down lands, plains and basins, and surface waters are comprise approximately 58.4%, 41.6% and 10.2% of the total Nanjing area, respectively. The region belongs to the north subtropical monsoon climatic zone, with an annual average temperature of 17.8 ◦ C. The sunshine duration is 1 687 h and the annual average rainfall is 1 034 mm. Four rivers of Yangtze River, Qinhuai River, Shuiyang River and Chuhe River traverse the region which also includes Gucheng Lake and Shijiu Lake, with the total water capacity of 428 million m3 (Lin et al., 2008) (Fig. 1a).
Water system (a) and soil sampling points (b) in Nanjing area, eastern China.
676
L. MA et al.
Quantitative monitoring method The proposed SE and NPSP quantitative monitoring method consists of nine parts, including SE and NPSP quantitative monitoring models, formulae and arithmetics of the factor models and related software, precision evaluation of digital elevation model (DEM), practical application of the fast update of GIS using global positioning system (GPS) measurement, and database building and inquiry (Bu et al., 1999, 2005; Ma et al., 2011). The types of models used in this integrated method are as the same as the USLE and revised USLE (RUSLE) models used in America (Wischmeier and Smith, 1978; Renard et al., 1997), but the formulae and arithmetics of each model factor were built based on large quantities of measured data and grid pixel calculation in the erosion area in China. This approach resulted in significantly different formulae, arithmetics and its related software from the USLE and RUSLE model. The proposed SE monitoring model (Eq. 1) retains the factors of the USLE to calculate annual sheet and rill erosion from a hillslope; however, changes have been made for some factors. Ai = f ·Ri ·Ki ·LS i ·CP i
(1)
where Ai (t km−2 year−1 ) is the amount of annual soil loss for a grid pixel i (i = 1, . . . , n) of the area under consideration having a certain soil type; Ri (MJ mm ha−1 h−1 year−1 ) is the rainfall erosivity factor of grid pixel i; Ki (t ha h ha−1 MJ−1 mm−1 ) is the soil erosion factor of grid pixel i; LS i (a dimensionless unit) is the product of slope length factor L and slope factor S of grid pixel i; CP i (also a dimensionless unit) is the product of vegetation cover factor C and soil conservation factor P of grid pixel i; and f is the conversion coefficient (normally it is 100, which transforms a unit of soil erosion from t ha−1 year−1 to t km−2 year−1 ). NPSP of the soil nutrient monitoring model is shown in Eq. 2: Wi,j = Ai ·cj
(2)
where Wi,j (kg km−2 ) is the annual increase in the amount of soil nutrient of the jth species (i.e., TN, TP, AN, and AP) for a grid pixel i; and cj (g kg−1 for TN and TP, and mg kg−1 for AN and AP) is the soil nutrient content of the jth species in the monitored soil use types. The main changes to the RUSLE approach incorporated into the proposed model are as follows (Bu et al., 1997):
R factor: R factor is the erosion power factor which was a major determinant of difference of the erosion amount in different regions or periods. Using the rainfall and runoff observation data for many years in various typical rainfall areas in China, a new arithmetic of R factor was obtained (Eq. 3): R = 2.1803Pf ·I30 − 2.6810I30
(3)
where Pf (mm) is the total rainfall of each month in the flood season; and I30 (cm h−1 ) is the annual value of the maximum rainfall intensity in 30 min. LS factor: LS factor is the acceleration factor of erosion power. As the slopes of farmlands in China are usually larger than those in USA, arithmetic of slope factor S in the USLE equation was not suitable for application in China. Using a large number of measured data for many years, we had made a large improvement on S factor formula (Eq. 4), and a series of formulae were incorporated into the software. To meet the requirements of quantitative remote sensing (RS) of soil erosion based on pixels, equations for slope factor S and slope length factor L of each pixel were made some improvements (Eqs. 5 and 6): S = 0.743 × 1.0595αi
(4)
αi = f (hi , hj , di )
(5)
Li = f (di , αi , mi )
(6)
where αi is the most appropriate slope value in the adjacent eight directions of the ith pixel; hi (hj ) is elevation value of the ith (jth) pixel; di is the side or diagonal length of the ith pixel; and mi is the index value changes with slope. CP factor: CP factor is the inhibiting factor of erosion power which reflects the effect of soil and water conservation. CP factor value is less than or equal 1. C factor was established using the observed data of the natural vegetation area. The equation is as follow: ⎧ ⎪ ⎨ 0.4149 − 0.0052C1 0.4399 − 0.0058C2 (7) C= ⎪ ⎩ 0.4500 − 0.0079C3 where C1 , C2 , and C3 (%) are the maximum, secondlargest, and average vegetation coverage, respectively, in a vegetation growth period. In generally, the value of P factor is l for the natural vegetation area and slope land, 0.01 for the level terrace, and 0.02 to 0.7 for the farmland with different control measures. The SE levels were evaluated based on the annual soil erosion module (t km−2 year−1 ) according
METHOD TO MONITOR SOIL EROSION AND POLLUTION
to the National Standard for Classification and Gradation of Soil Erosion (SL 190-2007) (Table I). The standard was also employed to assess NPSP. The R factor was obtained from rainfall data of various precipitation stations from 2001 to 2010. According to location, terrain and climate, four representative precipitation stations (Xiaguan, Luhe, Yuetang Reservoir and Tianshengqiao) were selected. The maps containing the administrative boundaries, DEM, and soil and land use types of Nanjing were extracted from the “Nanjing soil and water conservation geographic information system (NJSWCGIS)” and updated according to the latest boundaries of districts and towns, land use types and DEM map. Satellite data of the Nanjing region were obtained from a charge-coupled device received on April 3, 2008 and thematic mapper images received on April 17, 2000. TABLE I Soil erosion and non-point source pollution (NPSP) levels and classifications based on annual soil erosion module according to the National Standard for Classification and Gradation of Soil Erosion (SL 190-2007) Soil erosion module 103
km−2
× t < 0.5 0.5–2.5 2.5–5.0 5.0–8.0 8.0–15.0 > 15.0
Level
Classification
Insignificant Low Moderate High Severe Extremely severe
Class Class Class Class Class Class
year−1 I II III IV V VI
Sampling and analyses The soil samples were collected from 45 sampling sites (257 soil samples) distributed over the Nanjing area (Fig. 1b). Freshly collected samples were air-dried and the soil was passed through 20- and 100-mesh sieves after plant roots and other debris were removed. After that, the soil samples were stored in plastic bags for chemical analysis. Soil TN was determined using the Kjeldahl method (Bremner and Mulvaney, 1982). Soil TP was determined by digestion in H2 SO4 HClO4 and analyzed by colorimetry. Soil AN was determined using the alkaline hydrolysis diffusion method (Lu et al., 2000). Soil AP was determined by sample extraction with 0.5 mol L−1 NaHCO3 , and analyzed by colorimetry (Lu et al., 2000). The soil information was used as an input for the quantitative prediction of SE and NPSP in the Nanjing area. Statistical analysis Analysis of variance was conducted by SPSS ver-
677
sion 16.0 to process the quantitative monitoring data. Data fitting processes and graphs were prepared using the ArcGIS 9.2 and Excel 2010 software. RESULTS AND DISCUSSION SE in Nanjing area from 2001 to 2010 To investigate the spatial pattern of SE from 2001 to 2010 in the Nanjing area, the 10-year average of SE was analyzed and compared with the standards in Table I. The results showed that the SE of a region with the area of 896.68 km2 exceeded class II. This region corresponds to hill and plain with the area of 881.48 and 15.20 km2 , respectively. The amount of SE involved was 2.66 × 106 t, of which hill and plain area were 2.57 × 106 and 8.34 × 104 t, respectively. The overall SE module of this region was 2 441 t km−2 year−1 , which is at the upper end of the range for class II. The SE module of the hill area was 2 469 t km−2 year−1 and that of the plain 861 t km−2 year−1 . The results indicate clearly that SE in the Nanjing area occurs primarily in the hill area, as expected, representing about 22% of the total area investigated. Some suburbs with rapid economic development (such as Jiangning District) represented about 35% of the total SE amount. This is because the uncovered slopes of hills accelerate the movement of loose soil particles together with the nutrients bound to them, increasing soil losses (Rejani and Yadukumar, 2010) and NPSP. Moreover, the dredging activities during the industrial development in the Jiangning District also loosened soil structure, causing additional loss. NPSP in Nanjing area from 2001 to 2010 Quantitative monitoring of average soil TN, TP, AN, and AP pollution sources in the Nanjing area from 2001 to 2010 is summarized in Table II. The NPSP load is defined as the sum of TN and TP (Ongley et al., 2010; Wu et al., 2011). It is obvious that the NPSP occurred mainly in the hill area. The pollution modules of soil available nutrients (AN and AP) were small, but those of soil TN and TP were larger and therefore of more serious concern. In the hill area, the pollution module of soil total nutrients was classified to the moderate level (class III). This pollution module is equivalent to 21–30 kg ha−1 of N and P carried by soils entering into downstream water bodies. Temporal variation of NPSP during 2001–2010 Using the adapted model, we obtained annual changes of pollution load, module and area of TN, AN, TP and AP in the Nanjing area as a consequence of diffe-
678
L. MA et al.
TABLE II Average non-point source pollution (NPSP) area, load and module in the Nanjing area from 2001 to 2010 NPSPa)
Pollution area Total
Pollution load
Hill area
Plain area
Total
km2 TN AN TP AP a) TN
1 113.4 109.1 621.7 228.2
1 087.5 109.0 615.8 227.7
25.9 0.1 5.9 0.5
3 765.5 289.3 1 704.3 53.0
Pollution module
Hill area t 3 660.5 281.6 1 642.5 50.9
Plain area
Total
Hill area km−2
105.0 7.7 61.9 2.1
2 935.0 1 154.4 2 085.0 123.0
kg 2 984.0 1 154.5 2 097.0 123.1
Plain area
year−1 867.0 782.5 844.0 82.7
= total N; AN = available N; TP = total P; AP = available P.
rences in local climate and surface conditions over the decade of 2001–2010 (Figs. 2 and 3). In 2009 and even clearer in 2003, the annual nutrient load, the pollution module and the pollution area with erosion module exceeding class II were higher than the other years. The minima of these three variables were found in 2001. According to the monthly-distributed data, SE and NPSP mainly occurred from May to October in each year. This is the wet season of the study area, since April to October account for about 65% of the annual rainfall. The SE and NPSP show strong correlations with rainfall, as reported in the previous studies (Lee et al., 2010; Peng and Wang, 2012). Fig. 3 Changes of non-point source pollution (NPSP) area over low NPSP level (class II) in the Nanjing area from 2001 to 2010. TN = total N; TP = total P; AN = available N; AP = available P.
Fig. 2 Changes of non-point source pollution load and module in the Nanjing area from 2001 to 2010. TN = total N; TP = total P; AN = available N; AP = available P.
In the decade of 2001–2010, vegetation coverage in the Nanjing area has increased due to actions of the local authorities. Increasing vegetation cover has probably enlarged the buffer capacity of certain areas for block sediments, nutrients and other contaminants transported with runoff and SE (Duchemin and Hogue, 2009). Effective measures conducted by the local authorities to control agricultural NPSP include establishment of riparian vegetation buffers and restoration of forest on cultivated and sloping arable land. Both measures help to reduce nutrient flux to downstream waters (Tian et al., 2010; Zhang et al., 2011). In order to quantitatively assess the effect of vegetation management from 2001 onwards, we collected 4 years (1997–2000) of monitoring data before 2001. This 4year dataset can be regarded as the baseline. Figs. 4 and 5 show a comparison of the baseline with the data of 2001–2010. Focusing on N pollution, the maximum soil TN pollution load, module and area exceeding NPSP class II were identified in 2003, and then in 2009, with the minimum values in 2001. The average soil TN pollution
METHOD TO MONITOR SOIL EROSION AND POLLUTION
679
Fig. 4 Average values of non-point source pollution (NPSP) areas at different NPSP levels for the 4-year baseline (1997–2000) and 10-year study period (2001–2010) in the Nanjing area. TN = total N; TP = total P; AN = available N; AP = available P.
Fig. 5 Average values of non-point source pollution (NPSP) loads in the areas at different NPSP levels for the 4-year baseline (1997–2000) and 10-year study period (2001–2010) in the Nanjing area. TN = total N; TP = total P; AN = available N; AP = available P.
680
module for most years reached class III in most years, excluding 2001. The area with the average soil TN pollution exceeding class II was about 1 113 km2 , with 7 years of the study accounting for more than 1 000 km2 . Compared to the baseline period, the 10-year averages of soil TN pollution load (3 267.7 t), module (2 935 t km−2 year−1 ) and pollution area (1 113.4 km2 ) exceeding class II significantly decreased (P < 0.05). The soil TN pollution areas at low (class II) and high NPSP levels (class V) significantly decreased by 33.3% and 48.3%, respectively (P < 0.05). Similarly, the maxima of soil AN pollution load, module and area exceeding class II were found in 2003, and then in 2009, with the minimum values found in 2001. For each year of the study, the average soil AN pollution module mostly reached the low NPSP level (class II). The 10-year soil AN pollution load, module and pollution area exceeding class II significantly decreased from 231.26 to 125.90 t, 1 360 to 1 165 t km−2 year−1 and 170.08 to 109.07 km2 , respectively (P < 0.05). Soil TP and AP evolutions showed similar trends with the maxima occurring in 2003, and then again in 2009. The minima were found in 2001. The average soil TP pollution module for each year (excluding 2003) reached the low level (class II), corresponding to about 2 038.4 t km−2 year−1 . Compared with the 4-year baseline values, the 10-year average of soil TP pollution load (1 296.3 t), module (2 085 t km−2 year−1 ) and pollution area (621.69 km2 ) exceeding class II significantly decreased (P < 0.05). The 10-year soil TP pollution area at the high (class IV), severe (class V) and extremely severe (class VI) levels significantly decreased by 34.6%, 46.8% and 64.0%, respectively (P < 0.05) as compared to the baseline. The average soil AP pollution module for most years reached the insignificant level (class I), and the 10-year average of AP (123 t km−2 year−1 ) was also significantly lower than the 4year average (150 t km−2 year−1 ) (P < 0.05). Many studies showed that soil and water conservation measures demonstrate positive effects on soil fertility and soil nutrient loss reduction in China (Lin et al., 2009; Dur´an Zuazo et al., 2011; Nunes et al., 2011; Deng et al., 2012; Phan Ha et al., 2012). In this study, we could confirm the positive effect of vegetative cover on erosion reduction and, therefore, also on the NPSP decrease of N and P, even though the nutrient load has increased during the study period in many soil types of the Nanjing area. Spatial variation of NPSP during 2001–2010 NPSP not only changed in different years or months, but also significantly changed in different dis-
L. MA et al.
tricts of Nanjing. To analyze the spatial variation of NPSP in the Nanjing area, four variables were chosen, including soil pollution load, pollution module and the pollution areas exceeding classes II and IV. The concept of “management degree” was used to evaluate the effect of control measures on SE and NPSP between 2001 and 2010. The management degree is calculated for each of the four variables by taking the difference of average values of the variable in two periods (1997– 2000 and 2001–2010) and then dividing the difference by the average value in 1997–2000. By ranking the results for each area, the evolutions of SE and NPSP and the effectiveness of control measures can be seen for different districts of Nanjing area (Table III). The average management degrees of the pollution load and pollution areas exceeding classes II and IV for soil TN and AN exceeded 30% in the whole Nanjing area. However, the average management degree of the pollution area over class II for soil TP and AP was less than 30%. The effect of vegetative cover was thus stronger for N than for P. The largest effect of the measures taken to reduce SE and NPSP occurred in the Luhe District (data not shown). The management degree was in the order of Luhe District > Pukou District > Lishui District > Qixia District > Downtown District > Gaochun District > Jiangning District > Yuhua Distict. Validation of the simultaneous model To evaluate the reliability of the quantitative monitoring results of SE and NPSP, the model obtained in this study were validated in the field and tested using hydrological data collected in 2002 and 2003. The results showed that the monitoring precisions of TN in 2002 and 2003 were 71.6% and 84.7%, respectively. The average precision was 78.1%. The monitoring precision of TP in 2002 and 2003 were 92.5% and 83.1%, respectively. The average precision was 87.8%. Previous studies also show that the simultaneous determination of SE and NPSP by the proposed method is reliable and credible (Bu et al., 1999; Xia et al., 2007). The revised integrated monitoring method has better recognition ability for spatial variation of SE and NPSP with the support of “3S” (GIS, RS, GPS) technology. The new integrated method is also characterized by convenient, efficient, and quickly updated spatial data (Jiang et al., 2005). CONCLUSIONS The proposed quantitative method to assess NPSP and SE, simultaneously based on soil samples, DEM, land use and soil layers in a GIS environment, is trust-
METHOD TO MONITOR SOIL EROSION AND POLLUTION
681
TABLE III Effect of control measures on soil total nitrogen (TN) and total phosphorus (TP) pollution, evaluated by the concept of “management degree”, in different districts of Nanjing area Pollution
TN
TP
District
Luhe Pukou Yuhua Qixia Jiangning Downtown Lishui Gaochun Total Luhe Pukou Yuhua Qixia Jiangning Downtown Lishui Gaochun Total
Total area
km2 1 479.1 909.2 131.6 371.6 1 580.3 258.5 1 068.8 790.9 6 589.9 1 479.1 909.2 131.6 371.6 1 580.3 258.5 1 068.8 790.9 6 589.9
Management degree Pollution module
Pollution load
Pollution area over class II
Pollution area over class IV
3.9 8.3 1.3 16.2 4.1 42.3 23.3 5.8 7.9 7.6 14.3 9.1 19.1 5.5 38.9 23.8 16.7 11.5
36.8 45.4 7.3 28.8 24.2 54.7 45.3 37.2 34.8 38.3 46.8 6.6 28.7 22.1 49.6 43.7 37.2 34.7
% 36.6 42.7 5.8 13.0 21.8 27.2 34.8 31.5 30.9 35.4 40.2 −5.7 7.1 17.8 23.9 36.0 21.9 27.6
29.1 48.0 −15.0 1.2 18.8 44.6 61.4 43.0 31.0 45.1 59.2 20.9 30.5 25.2 74.0 73.3 59.5 43.6
worthy and provides scientific support for local SE and NPSP management in the Nanjing area. We confirmed in this study that the method has a high precision, reliability and credibility. The calculated TN, TP, AN, and AP levels in different types of soils allow assembling maps and temporal evolution of soil nutrient status in the Nanjing area. The positive effects of erosion control measures on SE and NPSP in the study area were significant over the decade of 2001–2010 (P < 0.05). The results of our study confirmed that the implementation of vegetative cover is an effective measure to control SE and NPSP. Therefore, simultaneous spatio-temporal estimation of SE and NPSP with the proposed integrated model can be applied for management of the areas with intensive agricultural use. REFERENCES Braskerud, B. C. 2002. Factors affecting phosphorus retention in small constructed wetlands treating agricultural non-point source pollution. Ecol. Eng. 19: 41–61. Bremner, J. M. and Mulvaney, C. S. Nitrogen—Total. 1982. In Page, A. L., Miller, R. H. and Keeney, D. R. (eds). Methods of Soil Analysis. Part 2. Chemical and Microbiological Properties. American Society of Agronomy, Madison. pp. 595–624. Bu, Z. H., Jiang, X. S., Yang, L. Z. and Zhang, Z. X. 2005. The experiment of optimum methods of renewing GIS’s data by GPS solid survey in the soil erosion fixed quantity monitoring. Acta Pedol. Sin. (in Chinese). 42: 712–719. Bu, Z. H., Sun, J. Z., Dong, Q. R. and Liu, S. Q. 1999. Study
on quantitative remote sensing method for soil erosion losses and its application in Shandong Province, China. Acta Pedol. Sin. (in Chinese). 36: 1–8. Bu, Z. H., Sun, J. Z., Zhou, F. J., Tang, W. L. and Xi, C. F. 1997. A study on quantitative remote sensing method of soil erosion and its application. Acta Pedol. Sin. (in Chinese). 34: 235–245. Deng, L., Shangguan, Z. P. and Rui, L. 2012. Effects of the grainfor-green program on soil erosion in China. Int. J. Sediment Res. 27: 120–127. Duchemin, M. and Hogue, R. 2009. Reduction in agricultural non-point source pollution in the first year following establishment of an integrated grass/tree filter strip system in southern Quebec (Canada). Agr. Ecosyst. Environ. 131: 85– 97. Dur´ an Zuazo, V. H., Rodr´ıguez Pleguezuelo, C. R., Martin Peinado, F. J., de Graaff, J., Francia Mart´ınez, J. R. and Flanagan, D. C. 2011. Environmental impact of introducing plant covers in the taluses of terraces: Implications for mitigating agricultural soil erosion and runoff. Catena. 84: 79– 88. Edwin, D. O., Zhang, X. L. and Yu, T. 2010. Current status of agricultural and rural non-point source pollution assessment in China. Environ. Pollut. 158: 1159–1168. Huang, M. X., Zhang, S., Tang, Y. J. and Chen, X. B. 2001. Nitrogen losses from farm runoff under simulated rainfall conditions. Soil Environ. Sci. (in Chinese). 10: 6–10. Jiang, X. S., Bu, Z. H., Yang, L. Z., Xia, L. Z. and Pan, J. J. 2005. Integrated quantitative monitoring of soil loss and water pollution—A case study of Fangbian valley in Jiangsu Province. Acta Pedol. Sin. (in Chinese). 42: 529–536. Lee, C. S., Chang, C. H., Wen, C. G. and Chang, S. P. 2010. Comprehensive nonpoint source pollution models for a freerange chicken farm in a rural watershed in Taiwan. Agr. Ecosyst. Environ. 139: 23–32. Leon, L. F., Soulis, E. D., Kouwen, N. and Farquhar, G. J. 2001.
682
Nonpoint source pollution: a distributed water quality modeling approach. Water Res. 35: 997–1007. Lin, C. W., Tu, S. H., Huang, J. J. and Chen, Y. B. 2009. The effect of plant hedgerows on the spatial distribution of soil erosion and soil fertility on sloping farmland in the purplesoil area of China. Soil Till. Res. 105: 307–312. Lin, J., Zhang, J. C., Wu, Y. M., Peng, G. L., Shi, K. F. and Li, H. D. 2008. The current situation and causes of water and soil losses in Nanjing and its conservation strategies. J. Nanjing Forest. U. (in Chinese). 32: 44–46. Lu, R. K. 2000. Analytical Methods of Soil Agrochemistry (in Chinese). China Agricultural Science and Technology Press, Beijing. Ma, L., Bu, Z. H., Peng, G. L., Xia, L. Z., Li, Y. D., Yang, L. Z., Jiang, X. S., Huang, R. and Wu, Y. M. 2011. Temporal and spatial variation of soil and water loss and its control in Nanjing area from 2001 to 2008. Acta Pedol. Sin. (in Chinese). 48: 683–692. Nunes, A. N., de Almeida, A. C. and Coelho, C. O. A. 2011. Impacts of land use and cover type on runoff and soil erosion in a marginal area of Portugal. Appl. Geogr. 31: 687–699. Ongley, E. D., Zhang, X. L. and Yu, T. 2010. Current status of agricultural and rural non-point source pollution assessment in China. Environ. Pollut. 158: 1159–1168. Ouyang, W., Skidmore, A. K., Toxopeus, A. G. and Hao, F. H. 2010. Long-term vegetation landscape pattern with nonpoint source nutrient pollution in upper stream of Yellow River basin. J. Hydrol. 389: 373–380. Peng, T. and Wang, S. J. 2012. Effects of land use, land cover and rainfall regimes on the surface runoff and soil loss on karst slopes in southwest China. Catena. 90: 53–62. Phan Ha, H. A., Huon, S., des Tureaux, T. H., Orange, D., Jouquet, P., Valentin, C., Rouw, A. D. and Duc, T. T. 2012. Impact of fodder cover on runoff and soil erosion at plot scale in a cultivated catchment of North Vietnam. Geoderma. 177178: 8–17. Rejani, R. and Yadukumar, N. 2010. Soil and water conservation techniques in cashew grown along steep hill slopes. Sci. Hortic. 126: 371–378. Renard, K. G., Foster, G. R., Weesies, G. A., Mccool, D. K. and Yoder, D. C. 1997. Predicting Soil Erosion by Water—A Guide to Conservation Planning with the Revised Soil Loss Equation (RUSLE). Agriculture Handbook No. 703. US Department of Agriculture, Washington, D.C.
L. MA et al.
Shen, Z. Y., Liao, Q., Hong, Q. and Gong, Y. W. 2012. An overview of research on agricultural non-point source pollution modeling in China. Sep. Purif. Technol. 84: 104–111. Tian, Y. W., Huang, Z. L. and Xiao, W. F. 2010. Reductions in non-point source pollution through different management practices for an agricultural watershed in the Three Gorges Reservoir Area. J. Environ. Sci. 22: 184–191. Wischmeier, W. H. and Smith, D. D. 1978. Predicting Rainfall Erosion Losses—A Guide to Conservation Planning. Agriculture Handbook No. 537. US Department of Agriculture, Washington, D.C. Wu, Y. H., Hu, Z. Y. and Yang, L. Z. 2008. Integrated technology to control non-point pollution in resident-farmland area of Dianchi Lake watershed, China. Ecol. Sci. (in Chinese). 27: 346–350. Wu, Y. H., Hu, Z. Y. and Yang, L. Z. 2010a. Hierarchical ecorestoration: A systematic approach to removal of COD and dissolved nutrients from an intensive agricultural area. Environ. Pollut. 158: 3123–3129. Wu, Y. H., Kerr, P. G., Hu, Z. Y. and Yang, L. Z. 2010b. Ecorestoration: Simultaneous nutrient removal from soil and water in a complex residential-cropland area. Environ. Pollut. 158: 2472–2477. Wu, Y. H., Xia, L. Z., Hu, Z. Y., Liu, S. Z., Liu, H. B., Nath, B., Zhang, N. M. and Yang, L. Z. 2011. The application of zerowater discharge system in treating diffuse village wastewater and its benefits in community afforestation. Environ. Pollut. 159: 2968–2973. Xia, L. Z., Bu, Z. H., Zhang, Z. X., Yang, L. Z., Yang, X. Y. and Jiang, X. S. 2007. Application of integrated method to assessment of soil and nutrient losses in Miyun Reservoir Basin. Adv. Earth Sci. (in Chinese). 22: 241–248. Yan, B. X. and Tang, J. 2005. Study on black soil erosion rate and the transformation of soil quality influenced by erosion. Geogr. Res. (in Chinese). 24: 499–506. Yong, R. A., Onstad, C. A., Bosch, D. D. and Anderson, W. P. 1989. AGNPS: A nonpoint-source pollution model for evaluating agricultural watersheds. J. Soil Water Conserv. 44: 168–173. Zhang, P., Liu, Y. H., Pan, Y. and Yu, Z. R. 2011. Land use pattern optimization based on CLUE-S and SWAT models for agricultural non-point source pollution control. Math. Comput. Model. 58: 588–595.