Impacts of agricultural land management on soil quality after 24 years: a case study in Zhangjiagang County, China

Abstract

Agricultural soil quality is drastically affected in modern societies by human activities. This paper evaluates the anthropogenic influence on agricultural soil quality variation in Zhangjiagang County, China from 1980 to 2004 based on indicator selection and standard scoring function (SSF). The results indicated that after 24 years of anthropogenic influence, soil organic matter (SOM), total nitrogen (TN), cation exchange capacity (CEC) and available phosphorus (av-P) increased significantly and total phosphorus (TP) and available potassium (av-K) decreased slightly. Soil pH increased slightly for Cambosols and decreased significantly for Anthrosols. According to analysis of an integrated soil quality index (SQI), both Cambosol and Anthrosol soil quality have improved. For Cambosols, class II and III soil increased by 4.4% and 5.7%, respectively; class IV soil decreased by 10.1%. For Anthrosols, Class II soil increased in area by 74.6%; class III soil decreased to 8.1%. ΔSQI showed the same variation trend for Cambosols and Anthrosols. Among the total selected indicators, SOM was the main driving factor and pH was the limiting factor of soil quality base on path analysis. Current anthropogenic influence on soil quality variation was double-edged, stakeholders must pay close attention to this tendency and closely monitor not only agricultural production yields but also crop safety, long-term soil quality and environmental quality indicators, and timely direct agricultural and economic activities to achieve the best economic performance while protecting natural resources.

Keywords:

• soil indicators variation
• anthropogenic influence
• soil quality evaluation
• limiting factor
• path analysis

Introduction

With increasing populations and the rapid development of global economies, the effects of human activities on natural ecosystems are increasingly critical to ecosystem sustainability (Dumanski & Pieri 2000). Modern agricultural soil quality, which is directly linked to soil productivity and the ability of humans to feed ourselves (Al-Kaisi et al. 2005), has been drastically affected by human activities (Huang et al. 2007). An increase in crop yields over the last three decades is due to the development of high-yield crop varieties and the increasing use of chemical fertilisers, pesticides, irrigation and mechanisation (Bindraban et al. 2000). In recent years, there has been growing concern about the effect of human activities on agricultural soil quality and its variation over time. Research indicates that appropriate agricultural practices such as tillage, irrigation, fertiliser and lime application, incorporation of crop residues into soil and conversion from dry land agriculture to rice paddy production have improved soil quality (Rasmussen & Parton 1994; Fischer et al. 2002; Karlen et al. 2006; Raiesi 2006; Huang et al. 2007). These factors are accelerating changes in soil properties, both directly and indirectly, to the degree that new trends reveal that human-induced variation of soil properties has surpassed natural variation (David et al. 2001).

Agricultural soil quality refers to the condition and capacity of land, including its soil, climate and biological properties, for purposes of production, conservation and environmental management (Pieri et al. 1995; Stamatiadis et al. 1999a). Positive and negative human activities on soil quality occur simultaneously, the extent of which depends on ecosystem resilience and disturbance feedback (Franzluebbers & Stuedemann 2006). Conversely, inappropriate human activities such as excessive inorganic and chemical fertiliser application and sewage sludge irrigation decrease soil quality (Edwards & Lofty 1982; Stamatiadis et al. 1999b; Ward 2001; Datta and de Jong 2002; Qi & Chang 2005). An estimated 40% of agricultural lands are affected by human induced land degradation (Oldeman et al. 1990). However, relatively little research has focused on the change of soil fertility over time and the links between soil fertility changes and anthropogenic influences.

There are many methods to evaluate soil quality, such as integrated soil quality indexes (SQIs) (Doran et al. 1994; Doran & Jones 1996), multi-variable indicator kriging (MVIK) (Nazzareno & Michele 2004) and soil quality dynamics (Larson & Pierce 1994), but no one uniform method has been well-accepted universally. Integrated SQIs are the most common methods to assess soil quality. The indicators, the weights of the indicators and the calculation method of the quality indexes are the most important considerations in SQI methods (Wang & Gong 1998).

Since the beginning of the reform and opening up of China in the 1980s, agriculture management decisions were transferred from government group leaders to individual farmers. Furthermore, small industries have multiplied in peri-urban areas of the Yangtze River Delta (YRD) region and are interspersed among small farms. These two trends have created a unique situation unlike that seen in developed countries. Intensive agricultural production methods such as crop rotations, crop residue incorporation, increased inorganic fertiliser application, and irrigation and drainage engineering, side by side with rapid industrialisation and urbanisation, have taken place in densely populated areas and significantly increased soil quality variations. These changes have created challenges to the long-term sustainability of these intensively cultivated agricultural systems and have raised questions about appropriate management practices.

A better understanding of soil quality variation as influenced by human activities is important in order to improve sustainable land use management practices (McGrath & Zhang 2003), provide early warning of adverse trends, identify problem areas (Bindraban et al. 2000) and provide a valuable base against which subsequent and future measurements can be evaluated. The objectives of this study were to evaluate soil quality variation in a typical agricultural ecosystem of the YRD region, from 1980 to 2004, using Zhangjiagang County as the study area, and to assess the effects of human activities on soil quality variation.

Materials and methods

Study area

Zhangjiagang County, Jiangsu Province is located on a diluvial plain in the northern YRD region (31°43′–32°01′N, 120°22′–120°49′E) and covers a total area of 999 km² with a population of 0.89 million in 2004 (Figure 1). Zhangjiagang County has a humid monsoon climate in the north subtropical zone, with four distinct seasons, plentiful precipitation, abundant sunlight and a long frost-free period. Average annual temperature and precipitation were 15.0 °C and 1045.9 mm in 1980 and 15.2 °C and 1039.3 mm in 2004 respectively. The county has predominantly flat topography, slightly elevated in the south and is situated along the Yangtze River.

Figure 1  Soil map of Zhangjiagang County with sample sites from 1980 to 2004.

Soils

Zhangjiagang County can be divided into two soil orders, Anthrosols and Cambosols (CRGCST 2001) (Fig. 1). However, the soil orders are affected by diverse soil parent materials. Anthrosols, with a clay texture, were developed from lacustrine deposits of alluvium in the south; Cambosols were developed from Yangtze River neo-alluvium in the north and have a sandy texture.

Human activities

In 1986, crop land administrative and management systems were transferred from cooperatives to family-contract systems that allowed individual farmers to rent land. Previously, agricultural land was managed by government leaders. This policy change boosted production as farmers were given the opportunity to decide which crops to cultivate and how to manage the land. Increasingly, significant differences in management practices proliferated, which induced fragmentary production modes very different from large farm operations in other countries (Huang et al. 2007).

Historically, Cambosols have been used to cultivate dry-land crops such as wheat in the spring, and corn, cotton and sweet potato in the summer. However, since the 1980s, these crops have continuously decreased due to increased investment in irrigation and mechanical cultivation and mass migration of the labour force to cities, resulting in increased rice cultivation. Rotation of rice and wheat has been the conventional production mode in Anthrosols. Rice cultivation rapidly increased from 13,930 ha in 1980 to 20,900 ha in 2004. Since then, the rice land area has been stable at almost 20,000 ha, which accounts for 88% of the total arable land (Agricultural Bureau of Zhangjiagang County, unpublished data).

Since the mid-1970s, incorporation of crop residues has been widespread, with the total crop residue incorporation area gradually increased and reaching a peak in 1995 (Gu et al. 2000). After 1995, however, crop residue incorporation rapidly decreased; to avoid extra labour, crop residues were increasingly burned. Currently, however, farmers are again returning crop residue to the soil. Mechanical harvesting and high crop yields are producing increasing amounts of crop residues to be incorporated into soils, which is increasing soil organic matter (SOM).

Since 1983, fertiliser application has been gradually increasing and reached a peak in 1998; since then it has decreased slightly, with nitrogen (N) fertiliser accounting for the largest fertiliser input. In 1983, 1998 and 2004, N application rates on rice were 187 kg ha⁻¹, 352 kg ha⁻¹ and 345 kg ha⁻¹, and on wheat were 150 kg ha⁻¹, 340kg ha⁻¹ and 280 kg ha⁻¹ respectively. K fertiliser rates had no temporal trend from 1983 to 2004, with mean annual rates of 50 kg ha⁻¹ yr⁻¹ and 51 kg ha⁻¹ yr⁻¹ for rice and wheat plantings respectively, but had relatively wide variation (30–80 kg ha⁻¹ yr⁻¹ for rice and 28–73 kg ha⁻¹ yr⁻¹ for wheat). Organic fertiliser application decreased drastically during the same time period. For rice, inorganic to organic fertiliser ratios went from 2:3 in 1983 to 19:1 in 2004; for wheat, organic fertiliser in 1983 accounted for 53% and was phased out of use by 2001.

Irrigation and drainage construction was another important development in Zhangjiagang. Before the 1970s, narrow and discontinuous ditches constituted the irrigation system and frequent storms and floods destroyed many farmlands and crops. Since the 1980s, much investment has been allocated to irrigation and drainage construction to satisfy paddy soil demands. In 2004, 8073 ditches were dug, creating 3890 km of new drainage. In addition, 5305 drainage stations with hydro-dams were established, with total output exceeding 37,000kW.

Soil sampling and laboratory analysis

In 1980, sites were randomly sampled throughout the county on agricultural land (n = 272), which included 177 Cambosol sites and 95 Anthrosol sites. A detailed description of each site and site location was recorded for future reference. Each soil sample was a composite of sub-samples taken from six locations within 350 m² of arable land and on surface disturbed by tillage (0–20 cm). Samples were placed in plastic sampling bags and brought back to the laboratory for analysis. In 2004, sites were again visited and samples were taken as described above. Sample sites were registered in 2004 using a hand-held global positioning system (GPS). All samples were taken in the fall, after harvest and before the next cropping season, in order to avoid the effect of direct fertilisation during the growing season. Soil samples were air-dried and sieved through a 2 mm sieve. SOM was determined by the dichromate-wet combustion method (Nelson & Sommers 1982) and total nitrogen (TN) was determined following the Kjeldahl method (Bremner & Mulvaney 1982). Total phosphorus (TP) was measured colorimetrically with ammonium molybdate after acid digestion. Soil pH was measured with a glass electrode in a 1:2.5 soil/water suspension. Cation exchange capacity (CEC) was measured by the sodium saturation method (Lu 2000). Soil available phosphorus (av-P) was extracted with 0.5 mol l-1 NaHCO₃ at pH 8.5, and P was determined colorimetrically using the molybdate method (Olsen et al. 1954). Soil available potassium (av-K) was extracted with 1 N ammonium acetate and then measured using an atomic absorption spectrometer (AAS) (Lu 2000). Soil available copper (av-Cu), zinc (av-Zn) and boron (av-B) were extracted using DTPA (Diethylene Triamine Pentacetate Acid), then measured using AAS (Lindsay & Norvell 1978).

Data processing and spatial analysis

Using GIS software, soil and administrative maps (1:5000) were digitised. Concentrations of all sample points in different years were interpolated using the block-kriging method to generate spatial distribution maps of soil quality (cell size = 100 m). Distribution maps of soil quality variation between the two years were generated by overlaying the spatial distribution maps from different years. Analysis of variance was performed using the general linear model (GLM). Mean separations were performed using Duncan's multiple range test to determine the effect of soil series and year on soil quality.

Evaluation of soil quality variation

A soil quality indicator is a measurable soil property that affects the capacity of a soil to perform a specified function (Karlen et al. 2006). For evaluation of soil quality, it is desirable to select indicators that are directly related to soil quality, but not identical for different research purposes and regions (Wang & Gong 1998). In this research, 10 indicators were selected for soil quality evaluation (SOM, TN, pH, CEC, TP, av-P, av-K, av-Cu, av-Zn and av-B). These indicators were chosen because the research purpose was focused on soil fertility quality. Nitrogen, P, K, av-Cu, av-Zn and av-B show the nutrient status of the soil for plants. SOM, CEC and pH influence the habitat for soil nutrients.

A standard scoring function (SSF) (Hussan 1997) was used to calculate the scores for all soil indicators except pH. The SSF is given by:

1

where x is the monitoring value of the indicator, f(x) is the score of the indicator, ranging from 0.1 to 1, and L and U are the lower and the upper threshold values respectively. The weight and values L and U for the indicators are determined by the effect of soil properties on plant growth in subtropical China (Sun et al. 1995; Wang & Gong 1998). L is the value under which plant growth is severely limited and U is the value at which plant growth is optimum. The values of L and U are listed in Table 1.

Table 1  Value of turning point of soil quality indicators in standard score function Equation (1).

	Indicator^2
Critical threshold^1	SOM	CEC	TN	TP	Av-P	Av-K	Av-Cu	Av-Zn	Av-B
L	15	10	0.75	0.4	5	40	2.0	1.5	0.5
U	30	20	1.5	1.0	15	100	4.0	3.0	1.0
^1 L is the value under which plant growth is severely limited; U is the value at which plant growth is optimum.
^2SOM is soil organic matter (g kg^−1), CEC is cation exchange capacity (cmol kg^-1), TN is total nitrogen (g kg^-1), TP is total phosphorus (g kg^-1), Av-P is available phosphorus (mg kg^-1), Av-K is available potassium (mg kg^-1), Av-Cu is available copper (µg kg^-1), Av-Zn is available zinc (µg kg^-1) and Av-B is available boron (µg kg^-1).

For soil pH, because there is an optimum range, the SSF is:

2

where L ₁ = pH 4.5,L ₂ = pH 5.5, L ₃ = pH 6.5 and L ₄ = pH 8.5.

There are many methods to assign indicator weights, such as experience, mathematical statistics and models (Wang & Gong 1998). In this paper, indicator weights were assigned using factor analysis. The given value and contribution of each principal factor was calculated, and then the commonality explained by each indicator based on the load matrix was calculated. The value of the commonality indicated the contribution of each soil indicator to soil quality and, on this basis, indicator weights were assigned (Table 2).

Table 2  Estimated communality and weight value of each soil quality indicator.

	Indicator
	SOM	pH	CEC	TN	TP	Av-P	Av-K	Av-Cu	Av-Zn	Av-B
Communality	0.879	0.791	0.711	0.783	0.802	0.604	0.533	0.675	0.560	0.308
Weight	0.132	0.119	0.107	0.118	0.121	0.091	0.080	0.102	0.084	0.046

The SQI was calculated using:

3

whereW _i is the weight and N _i the score of indicator i and n is the number of indicators.

The temporal changes in SQI (ΔSQI) were evaluated based on the difference between the two sampling years:

4

Soil quality was divided into four classes:

•	class I (high quality) is most suitable for plant growth with SQI >70
•	class II (medium quality) has slight limitations with SQI of 55–70
•	class III (low quality) has moderately severe limitations with SQI of 40–55
•	class IV has severe limitations, with SQI <40.

Changes in SQI were divided into six classes

•	extreme increase (>10)
•	moderate increase (5 to 10)
•	slight increase (0 to 5)
•	slight decrease (−5 to 0)
•	moderate decrease (−10 to −5)
•	extreme decrease (<-10).

Exploratory path analysis was used to determine the causal relationships between soil quality indicators and ΔSQI. Path analysis is a structural equation modelling technique that partitions correlations into direct and indirect effects (Bernatchez et al. 2008). Path analysis to determine the direct and indirect effects of each variable was conducted with multiple linear regression with sequential F-tests to discriminate between significant and non-significant variables using backward elimination. The standardised regression coefficients were used as path coefficients and included in the figures above the arrowed lines, representing their putative causal paths (Pamela et al. 2003). As such, it can be used to confirm or refute conceptual models describing mechanistic processes when model variables are inter-correlated.

Results

Changes in soil quality indicators

Comparison of soil quality indicators before and after 24 years of cultivation showed that Cambosols and Anthrosols varied considerably. In Cambosols, SOM, TN and CEC increased by 12.4%, 13.1% and 18.8% respectively and in Anthrosols 19%, 32.3% and 13% respectively (Table 3). Soil pH slightly increased for Cambosols from 7.9 to 8.1 and significantly decreased for Anthrosols from 7.4 to 6.4. Total P slightly decreased by 0.7% and 8.8%, but av-P increased significantly by 112.9% and 23.7% for Cambosols and Anthrosols respectively. Available K slightly decreased by 12.5% and 4.9% for Cambosols and Anthrosols respectively. Trace elements also increased during the two phases: av-Cu, av-Zn and av-B increased by 7.9%, 10.6% and 0.3% respectively for Cambosols and by 8.7%, 14.9% and 17.9% respectively for Anthrosols.

Table 3  Change of soil quality indicators in Cambosols and Anthrosols from 1980 to 2004.

	1980 mean±SD^1		2004 mean±SD		Increased% mean±SD
	Cambosols	Anthrosols	Cambosols	Anthrosols	Cambosols	Anthrosols
SOM (g kg^−1)	16.89±3.45d^2	22.24±3.94b	18.53±3.35c	25.35±4.20a	12.40±22.52	19.01±44.81
TN (g kg^−1)	1.08±0.18d	1.36±0.24b	1.19±0.24c	1.73±0.38a	13.11±24.75	32.29±44.56
pH	7.92±0.21b	7.40±0.48c	8.07±0.14a	6.38±0.81d	1.89 ±2.02	−13.78±15.69
CEC (cmol kg^−1)	10.17±2.17d	14.95±2.30b	11.75±2.33c	16.60±1.90a	18.79±26.60	12.96±18.37
TP (g kg^−1)	0.87±0.07a	0.67±0.14b	0.86±0.08a	0.60±0.12c	−0.66±12.93	−8.78±17.59
Av-P (mg kg^−1)	8.09±4.86c	9.54±4.96b	12.70±9.25a	9.28±4.49b	112.94±182.74	23.73±89.11
Av-K (mg kg^−1)	70.97±33.41a	76.69±30.68a	54.99±21.14b	73.89±26.88a	-4.89±75.15	12.46±61.71
Av-Cu (µg kg^−1)	5.38±0.70d	6.32±0.80b	5.72±1.70c	6.79±1.93a	7.92±36.79	8.66±31.95
Av-Zn (µg kg^−1)	0.71±0.13d	0.93±0.15b	0.76±0.43c	1.04±0.64a	10.55±61.34	14.89±76.83
Av-B (µg kg^−1)	0.23±0.03c	0.29±0.03b	0.23±0.09c	0.34±0.16a	0.26±41.60	17.86±57.40
^1SD is the standard deviation of each indicator.
^2Numbers followed by the same lowercase letter in each row are not significantly different at P = 0.05, using Duncan's multiple range test between soil quality indicators in different soil series and different years.

Field-scale variation is indicated by the coefficient of variation (CV) (Corwin et al. 2006). An interesting result was that the standard deviation (SD) of each soil quality indicator was higher than the mean value (Table 3), which means that the CV of each soil quality indicator was more than 100%. A high CV illustrates the heterogeneity of soil quality indicator variation from 1980 to 2004. This is a result of the variation of management practices on smallholding farm systems.

Soil quality variation

Generally, soil quality in Zhangjiagang increased from 1980 to 2004. In both years, class II and III were the dominant areas (Figure 2 and Figure 3). In 1980, there was no class I soil, and areas with class II, III and IV accounted for 36.2% (28,912 ha), 48.7% (38,903 ha) and 15.2% (12,119 ha) of the total soil area respectively. After 24 years of intense cultivation the amount of class III and IV areas decreased to 45.5% (36,331 ha) and 8.5% (6762 ha) respectively. Class II soil increased significantly to 40.2% (32,135 ha) and areas with class I accounted for 5.9% (4706 ha) of the total soil area.

Figure 2  Distribution of soil quality index of Zhangjiagang County in 1980 and 2004.

Figure 3  Soil quality variation of different soil orders of Zhangjiagang County in 1980 and 2004.

Of the two soil types, Anthrosols in the south had higher soil quality than Cambosols in the north, which may, in part, be related to soil texture derived from different parent materials (Ball et al. 2000; Mubarak et al. 2005) (Figure 2); however, overall soil quality improved in both soil orders. Cambosols in 1980 were predominantly class III (58.9%). By 2004, Cambosol quality improved slightly but still did not classify as class I soil. From 1980 to 2004, class II and III soils increased slightly by 4.4% and 5.7% respectively and class IV soil decreased by 10.1%. Compared to the slight improvement of Cambosols, Anthrosols improved dramatically. Anthrosols had higher soil quality in 1980, with 71.2% class II and 28.8% class III. After 24 years of cultivation, class II soil increased to 74.6%, class III soil decreased to 8.1% and 4680 ha of class I appeared.

Soil quality also varied between soil orders (Figure 4, Table 4). Table 4 indicates that 69.9% (36,967 ha) of Cambosols had positive ΔSQI values, and 26.7% (14,130 ha) had ΔSQI values >5. Where ΔSQI values decreased (30.1%, 15,889 ha), the change was primarily between -5 and 0. Of the total Anthrosol area, 93.1% (252.03 ha) had positive ΔSQI values, 70.4% (19,057 ha) had ΔSQI values >5 and 13.2% (3576 ha) had ΔSQI values >10. Only 6.9% area had a ΔSQI value between −5 and 0.

Figure 4  Evaluation map of soil quality variation of Zhangjiagang County in 1980 and 2004.

Table 4  Change percentage and area of different ΔSQI classes for different soil types from 1980 to 2004 in Zhangjiagang County.

Soil type	<−10	−10 to −5	−5 to 0	0 to 5	5 to 10	>10
Cambosols	nc^1 (0)^2	1.4% (726)	28.7% (15162)	43.2% (22837)	25.6% (13535)	1.1% (595)
Anthrosols	nc (0)	nc (0)	6.9% (1875)	22.7% (6146)	57.2% (15483)	13.2% (3576)
Total soil	nc (0)	0.9% (726)	21.3% (17037)	36.3% (28983)	36.3% (29.16)	5.2% (4171)
^1no change.
^2Numbers in parentheses are the soil areas (in ha) for different ΔSQI classes.

Relationship between soil indicators and soil quality changes

Changes in soil quality were controlled by a wide variety of soil quality indicators (Table 5). The results indicated that all soil quality indicators have significant correlation with ΔSQI except pH in Cambosols and B in Anthrosols, and also indicated that the change in all soil quality indicators, except pH, positively influenced soil quality. Acidification of the soil is having a negative effect on soil quality. The path coefficients between soil quality indicators and ΔSQI revealed the effect and contribution degree of the soil quality indicators and soil quality change. Path analysis indicated that SOM had the highest influence on soil quality changes, with path coefficients of 0.4 and 0.34 for Cambosols and Anthrosols respectively. Soil OM was the primary driving indicator of soil quality improvement in Zhangjiagang County from 1980 to 2004. The function sequence for soil quality improvement for Cambosols was SOM>av-P>CEC>av-K>TN>TP> av-Zn>av-Cu>av-B>pH; for Anthrosols it was SOM>av-K>av-P>CEC>TN>TP>av-Zn>av-Cu>av-B>pH. Available nutrients influenced ΔSQI more than total nutrients; that is, av-P and av-K were more significant than TN and TP, and trace nutrients did not significantly affect ΔSQI. Trace element variation did not significantly affect soil quality variation (Table 5).

Table 5  Relationship between soil quality indicators and ΔSQI in Zhangjiagang County.

	Correlation coefficient with ΔSQI		Path coefficient with ΔSQI
Indicator	Cambosols	Anthrosols	Cambosols	Anthrosols
pH	−0.1296	−0.3426^**	−0.1929	−0.1002
SOM	0.6978^** ^1	0.7380^**	0.3987	0.3362
CEC	0.4341^**	0.3743^**	0.2128	0.2035
TN	0.3405^**	0.5951^**	0.1609	0.1475
TP	0.5267^**	0.5066^**	0.1511	0.1054
Av-P	0.4395^**	0.5619^**	0.3390	0.2064
Av-K	0.4907^**	0.5997^**	0.2008	0.2409
Av-Cu	0.3014^**	0.2783^**	0.0466	0.0672
Av-Zn	0.4871^**	0.4172^**	0.1250	0.0302
Av-B	0.2932^**	0.1807	0.0093	0.0196
**Means correlation between change in soil parameter and ΔSQI is significant at P < 0.01.

Discussion

Positive effects of human activities on soil quality

Incorporation of crop residues usually improves SOM and TN directly, and induces accumulation of C and N (Rasmussen & Parton 1994; Fischer et al. 2002; Karlen et al. 2006; Raiesi 2006; Huang et al. 2007). Land use change from dry land to rice paddies also usually improves soil quality (Bhandari et al. 2002). Smallholder farm communities often improve soil quality indirectly by increasing the active participation of farmers (Mowo et al. 2006; Huang et al. 2007). The results of the current study support these findings. Organic and chemical fertilisers have increased, resulting in an increase in SOM, TN and av-P. Indicators such as SOM, TN, CEC, av-P and trace elements, to different degrees according to soil orders, caused increases in crop yields over the 24-year study period (Gu et al. 2007).

Negative effects of anthropogenic influences on soil quality

Soil acidification is an increasingly important topic in soil and environmental science and raises concerns about the sustainable use of soil resources. Unreasonable long-term application of ammonium-based fertiliser contributes to soil acidification (Gudmundsson et al. 2004). Long-term pesticide application also causes soil acidification (Rodriguez-Cruz et al. 2006). Acid deposition, containing high concentrations of sulphur dioxide, has been known to cause soil acidification in south China (Larssen et al. 1999), Hong Kong (Ayers & Yeung 1996) and north India (Satsangi et al. 2003). Sulphate and ammonium deposition caused forest soil pH to drop 0.1 to 0.3 units in northern Belgium (Schrijver et al. 2006). Sewage irrigation and industrial discharge containing large volumes of acidic mineral matter may also decrease soil pH markedly (Dubiková et al. 2002). In the current research, mean pH decreased from 7.4 in 1980 to 6.3 in 2004 and,in some localised areas,decreased by 2 units for Anthrosols.Furthermore, the decrease in Anthrosol pH has had a negative effect on soil quality (Table 4). According to Shao et al. (2006), heavy application of chemical fertilisers, acid deposition and industrial discharge could be responsible for soil pH decrease in Anthrosols.

Deficiency of av-K is a worldwide problem. The K status of European soils varies widely and, according to Johnston (2003), in many countries more than 25% of soils test ‘low’ and ‘very low’ in av-K. Long-term av-K trends show a decrease in many areas worldwide (Singh & Goulding 1997; Skinner & Todd 1998; Andrist-Rangel et al. 2007). Available K in Zhangjiagang County decreased from 1980 to 2004, and in 2004 was insufficient according to the national criteria for soil fertility grading (CGSF) (NSSO 1998). It is apparent that av-K is being depleted due to decreasing K fertiliser application and heavier dependence on N fertiliser application for higher yields.

Although Cu, Zn and B are important micronutrients for crop growth, their functions are double-edged. When their concentrations are excessively low, crop quality is affected; when their concentrations are too high, they are toxic to crops. Changes in soil pH and SOM can result in the transformation of micronutrients from non-available to available forms (Doelsch et al. 2006). The results of this work show that available trace nutrients have increased over the 24-year study period and this increase is well correlated with increasing SOM and decreasing pH (Table 2). For av-Cu, the levels exceeded the optimum for plant growth (1.8 mg kg^-1) and, for av-Zn and av-B, the average content fell below the optimal value(1.0 mg kg^-1)according to the CGSF (NSSO 1998). Excessive av-Cu may be caused by high background concentrations of Cu and by using factory waste water for irrigation, as noted in another study of the same area (Shao et al. 2008). The deficiency of av-Zn and av-B may be caused by many reasons, such as lower input of Zn and B fertilisers, the naturally deficiency of Zn and B in soil, removal of Zn and B in produce or leaching in drainage water for B.

Conclusions

Agricultural soil quality in Zhangjiagang County was significantly affected by human activities over the 24-year study period in two ways:

1.	an increase in soil nutrients, particularly SOM, TN, TP, av-P, and micronutrients positively influenced soil quality
2.	acidification, av-K depletion, excess av-Cu and deficient av-Zn and av-B negatively influenced soil quality.

We can conclude that agricultural ecosystems are increasingly affected by human activities and economic development, especially in rapidly developing regions of developing countries like China. In this process, soil fertility is highly monitored, but environmental concerns and crop quality are often neglected. Therefore, stakeholders must pay close attention to this tendency and closely monitor not only agricultural production yields but also crop safety, long-term soil quality, and environmental quality indicators and timely direct agricultural and economic activities to achieve the best economic performance while protecting natural resources.

Acknowledgements

The authors are grateful for funding from the National Key Basic Research Support Foundation of China (no. 2002CB410810), the Knowledge Innovation Project of the Chinese Academy of Sciences (no. KZCX3-SW-427) and the National Natural Foundation of China (no. 30872073).