East Asian Country

Paper Number 16

Fertilizer use, soil fertility and integrated nutrient management in China*

* This country report has not been formally edited and the designations and terminology used are those of the author.

Zhang F.S.¹, Fan M.S.¹, Zhao B.Q.², Chen X.P.¹,
Chen Q.¹, Li L.¹, Shen J.B.¹, Fen G.¹, Jiang R.F.¹,
Ma W.Q.³, Zhang W.F.¹, Cui, Z.L.¹, Fan X.L.⁴

1. Introduction

Since the establishment of the People’s Republic of China in 1949 the growth of agricultural production has been one of the main accomplishments of the country during its development. By 1999 China was successfully feeding 22 percent of the global population with only 9 percent of the world’s arable land, and per capita food availability reached developed country levels. Increasing the amount of inputs (e.g. fertilizer and water) has played a crucial role, accounting for about 50 percent of the yield increase.

Notwithstanding the achievements in agricultural production, there are still great challenges ahead. On the production side of the food sector, annual growth rates are gradually declining and grain production showed almost zero growth in 1996-2000. Furthermore, grain production declined from 508 mt in 1999 to 430 mt in 2003. On the demand side, China has to produce more food to feed an increasing population (predicted to plateau at 1.6 billion within 50 years), and improving living standards will also drive demand for high-value food products. On the resource utilization side, efficiencies (e.g. of nutrients) in crop production systems in China remain considerably lower than those in developed countries. Moreover, farmers are using increasing amounts of mineral fertilizers and only 47 percent of the agricultural land area continues to receive organic manures. The situation is further exacerbated by the loss of agricultural land at a rate of approximately one percent annually through rapid industrialization and urbanization; by the increased shortage of available water and fertilizer resources; and by the risk of environment pollution.

The Chinese government regards agriculture as the primary field of development of the national economy in the 21^st century. For China, the optimal agricultural developmental path is to improve the ratio of resource utilization and protect the environment while guaranteeing the grain supply. However, concerns about China’s food security and agricultural production have raised a number of questions. Will China continue to be able to feed its increasing population? What is the approach of improving grain production with efficient resource utilization and environment protection? What are the most effective policy measures to promote the technical aspects of improved grain production in terms of extension and dissemination?

This paper focuses on the approach of integrated nutrient management (INM) with the objectives of optimizing high yields, efficient resource utilization, and environmental protection. It first sketches fertilizer use and nutrient balance at the national level and presents general trends in soil fertility based on long-term fertilizer experiments. Current progress on technical measures for integrated nutrient management in major cropping systems in China is then addressed. Finally, policy effects and constraints that the country is facing in implementing integrated nutrient management are discussed.

2. Fertilizer use and nutrient balance in agro-ecosystems

2.1 Fertilizer use since the 1950’s

China has a long tradition over thousands of years of using organic materials to maintain relatively high yield levels and prevent soil fertility from declining. Before 1949 almost no inorganic fertilizer was utilized in China but the situation has changed greatly. The rapid increase in population and living standards has increased demands on agricultural production and the nutrients required outstripped the supply from organic manures.

Figure 2.1 Changes in fertilizer inputs in China from 1952 to 2004

Mineral fertilizers were therefore introduced in the 1950s and their use has increased rapidly. Calculated changes in fertilizer inputs in China from 1952 to 2004 are shown in Figure 2.1. The inputs of fertilizer N, P and K increased almost linearly from 8.9, 2.7 and 0.4 mt in 1980 to 24.8, 11.8 mt and 6.8 mt in 2004. The ratio of N:P₂O₅:K₂O in chemical fertilizers applied changed from 1:0.13:0.03 in 1980 to 1:0.21:0.23 in 2004, with an increased input of 180 percent, 340 percent and 1 767 percent for N, P and K over this 24-year period. Concomitantly, the contribution to total nutrient supply from organic manures decreased from almost 100 percent in 1949 to only 35 percent in 2001. For example, applied organic manures accounted for 18 percent of N, 28 percent of P and 75 percent of K overall in 2000.

The improper use of fertilizers since the 1950s, both over-use and under-application of fertilizers especially of N and P, has continuously occurred in different areas and cropping systems today. According to a recent survey by the Chinese Ministry of Agriculture, about one-third of farmers over-apply N, while one-third use deficiency levels of N on their crops (Ma Wenqi, personal communication). Evidence is mounting that both over-use and under-application of fertilizer can contribute to losses of crop yield, decreases in food quality and environment damage. It is therefore a great challenge to develop technological and political strategies and policies to further enhance grain yields while increasing nutrient use efficiency and protecting the environment.

2.2 N, P, K inputs, outputs and budgets in agro-ecosystems

Nutrient budgets are calculated as the difference between N, P and K inputs from fertilizers (both inorganic and organic) and from the environment (e.g. biological N₂-fixation, deposition and irrigation) and those nutrients removed from the farm in harvested products. The nutrient budgets in the agro-ecosystem in China from 1952 to 2004 at the national scale (Zhang, 2002; China Agriculture Yearbook, 2001-2005) are listed in Table 2.1. In the 1950s there was a small surplus of N with deficits of P and K. N and P became more balanced in the mid-1970s. The N and P budgets then show a surplus with K still in deficit. The nutrient surpluses of arable land reached 154 kg N/ha and 31 kg P/ha in 2004. The increasing surpluses of N and P can be attributed mainly to increasing fertilizer inputs and steady application of organic manures from 1980 to 2003. Although the K deficit decreased from -1.89 mt in 1979 to -1.34 mt in 2004, there was still a serious shortage of K. In addition, nutrient inputs from the environment, especially N inputs, have also contributed to the nutrient surpluses, accounting for 18.6 percent of N, 1.7 percent of P, and 12.7 percent of K of total inputs in 2004.

Nutrient balance varies even more strongly among crops. Typically, large volumes of inputs, either of chemical fertilizers, organic materials, or a mixture of both are very commonly applied in cash cropping systems and these systems involve large nutrient surpluses. Table 2 compares the differences in nitrogen balance in three cropping systems in Huiming county, Shandong province (Kou, 2004).

The total N input in greenhouse vegetable systems was 3 656 kg N/ha, which is 5.8 times that of wheat-maize systems (634 kg N/ha) and 4.2 times that of orchards (867 kg N/ha). N outputs in harvested products from these 3 systems were 329 kg N/ha, 280 kg N/ha and 121 kg N/ha, respectively. As a result, the N surplus in greenhouse systems (3 327 kg N/ha) was significantly higher than in orchards (746 kg N/ha) or wheat-maize systems (354 kg N/ha). Furthermore, an accumulation of up to 2 000 kg nitrate-N/ha in the top 2 m of the soil in orchards and 4 000 kg/N ha in the top 1 m in vegetable greenhouses has been found at various sites in north and northwest China (Ju et al., 2004).

Very little nitrogen from seed in plastic greenhouse vegetable systems was ignored Occupied by perennial apple trees, therefore no seed N input

3. Crop productivity and soil fertility in long-term fertilizer experiments

In China a network of experiments was established in 1990 to test the effects of fertilizer and manure treatments on the productivity of a variety of cropping systems in eight highly diverse regions of the country. Crop productivity and soil fertility in these experiments are summarized below.

3.1 Crop productivity

3.1.1 Wheat grain yields

Wheat was grown at six of the experimental sites. Similar yield responses to the various treatments were found at Beijing, Henan and Shanxi. Yields were poor on plots given no fertilizer, PK, NK or N. Yields increased markedly on plots where N, P and K were applied. Yields at Henan have been consistently high, averaging 6.5 t/ha since the start of the experiment. At Shanxi there has been a steady increase in yield from about 4 t/ha at the start of the experiment to 6.7 t/ha in recent years. In contrast, yields on the NPK treatment at Beijing have increased only in recent years, to ca. 5 t/ha, probably because of improved management.

Adding manure in addition to NPK tended to increase the yield of wheat at Shanxi and Beijing. However, the effects were small and unlikely to be significant. Yields did not increase at Henan but there were differences due to manure addition at three sites. In contrast to the Beijing site, fertilizer N was reduced by an amount equivalent to the total N added in the manures at the Shanxi and Henan sites when manure was applied. Adding straw in addition to NPK did not increase yields.

At Hunan wheat yields in fertilizer treatments have been compromised by soil acidity as a result of the acidifying effects of the Urea fertilizer. Adding manure has mitigated the effects of the Urea. As might be expected, yields of wheat were very poor on plots given fertilizers only. Improved, but still poor, yields were observed in plots where manures were applied.

Treatment effects at Chongqing have been similar to those at Beijing, Henan and Shanxi. However, the averaged wheat yield of the treatments was 3.5 t/ha, and addition of straw or manure with NPK did not further increase yields. Omission of K has resulted in a reduction in yield of about 0.8 t/ha compared to the NPK plots.

At Xinjiang winter wheat has been grown in six of the 15 years of the experiment. The yields on the control plots averaged 1.2 t/ha. Applying N only, PK or NK doubled the yield to 2.5 t/ha. On this site exchangeable K values are high even where no K has been applied for many years. Plots given NP gave a similar yield (5.1 t/ha) to those given NPK. Adding manure in addition to NPK did increase yields by about 0.5 t/ha, even though the amount of fertilizer N was reduced from 242 to 85 kg/ha.

3.1.2 Maize grain yields

At the three temperate semi-humid/humid sites (Beijing, Henan and Shanxi) yields on the control plots were poor. Yields on plots receiving NPK were very variable but has increased over time, averaging 6.5 t/ha at Beijing and 7.4 t/ha at both Henan and Shanxi. The effect of NPK plus manure was variable. At Beijing and Shanxi yields were increased by about 0.6 t/ha over the course of the experiment compared to NPK, but this was not significant. At Henan the effect was negligible. Given N only or NK, Henan and Shanxi crops yielded significantly more than the control plots in the first four or five years until available P in the soil became limiting. Where NP, but no K, was applied yields of maize did not decrease significantly compared to the NPK treatment at either Henan or Shanxi. At Beijing exchangeable K was also low and yields on the NP treatment, averaged over the 14 years, have been 0.7 t/ha less than those on the NPK treatment.

At Hunanyields of maize were severely constrained by increasing soil acidification, although not as severely as those of wheat. Yields on the control plots were poor at all times. Those on the N only and NK plots were initially better but then declined as P and/or pH became limiting. Yields on other treatments, although erratic, increased during the first six or seven years of the experiment. After that, yields on the NPK treatment declined steadily, presumably because of increasing acidity. As noted earlier, manure had a mitigating effect on soil pH. In recent years, yields on the NPK + M and on the 1.5 (NPK + M) treatments have remained fairly stable. Maize given manure only has yielded about 1.0 t/ha more than the NPK treatment in recent years.

Continuous maize was grown at Jilin. On average, yields of maize were greater here than at other sites which can possibly be attributed to longer growing period. On the control plots yields declined from about 5.0 to 2.5 t/ha during the experiment. Where NPK was applied, yields averaged 9.3 t/ha over the 15 years. Plots given N only, NK or NP have averaged 7.9, 8.8 and 9.0 t/ha respectively. Adding manure with NPK did not increase yields during most of the experiment, almost certainly because the amount of fertilizer N was reduced. Adding 1.5 (NPK + M) did increase yields by about 0.7 t/ha over the course of the experiment and the effect was more marked, i.e. 1.4/ha, in the subsequent years.

Maize was grown in rotation in 5 years on the desert soil at Xinjiang. Average yields on the control and PK plots were 4.2 and 5.6 t/ha respectively. Yields on plots given N or NK were slightly larger, ca. 6.5 t/ha. When N and P fertilizers were applied to the NP and NPK plots, it yielded to an average of ca. 7.1 t/ha. Adding manure together with NPK further increased the yields. Adding straw plus NPK also increased yields compared to plots given NPK.

3.1.3 Barley and rice grain yields

At Chongqing, rice was grown as the second crop after wheat. Yields on the control plots averaged 3.9 t/ha over the course of the experiment and have been maintained at a relatively high level, almost certainly because of nutrient inputs from the environment (e.g. irrigation, rain and biological N₂-fixation). Yields in NPK plots have been maintained at about 7.1 t/ha, higher than those in plots with N or P or K omitted. No residual effect was observed in plots in which manure only or manure plus NPK were given to the preceding wheat crop.

At Zhejiang, the cropping system was barley, early rice and late rice each year. Though the fertilization treatment gave higher yields than the control, it is doubtful whether this was significant. Yields of early rice averaged 3.7 t/ha on the control plots. Residues from manure alone added to the preceding barley crop did not increase yields. All other treatments increased yields on average by 1.1-1.6 t/ha compared to the control. Yields of the late rice crop were more variable than those of early rice. Yields on the control plots averaged 3.2 t/ha while yields in all other treatments averaged 3.7-4.3 t/ha and, in many cases, would not have differed significantly.

3.2 Soil fertility

3.2.1 Changes in soil organic C and total N

Table 3.1 shows the changes in soil organic carbon after about 10 years of the experiments at different sites. No fertilizer or manure inputs (control) over the long-term not only resulted in the lowest soil organic carbon values, but they also declined slightly after about 10 years compared with the initial contents at Henan, Jilin, Xinjiang and Chongqing. At Beijing, Shanxi, Hunan and Zhejing the soil organic carbon of control plots increased slightly. Long-term application of artificial fertilizers led to an increase or maintenance of soil organic carbon, with the exceptions of Jilin and Xinjiang. Artificial fertilizer combined with manure or straw can accelerate the accumulation of organic carbon at almost all of the experimental sites.

The changes in soil total N under different treatments at eight sites after about 10 years showed similar trends to the changes in soil organic carbon.

3.2.2 Changes in soil Olsen P and available K

Soil Olsen P decreased greatly compared with the initial values at the start of the experiment in those treatments without P fertilizer application, except for the Zhejiang (lowland) and Xinjiang sites. Long-term P inputs from fertilizer resulted in significantly increased Olsen P values which reached >20 mg/kg in almost all of the treatments. The values were higher in treatments with combined fertilizer and manure/straw, which can be ascribed to higher P inputs from the manure or straw.

Exchangeable K showed similar trends to Olsen P at some sites. Where K inputs were omitted, exchangeable K was prone to deletion at Henan, Shanxi, Hunan, Chongqing, Jiling and Xinjiang. Combined K fertilizer and manure/straw inputs resulted in higher exchangeable K than fertilizer K application only at Henan, Shanxi, Hunan, Jiling and Xinjiang. However, K fertilizer only or combined application of fertilizer K and manure/straw did not necessarily led to higher exchangeable K, as was found at Beijing, Chonging and Jiling.

¹ Average of the N, NP, NK and PK treatments
² M = manure
³ S = Straw

3.2.3 Changes in soil pH

Effects of fertilizer application on soil pH were found at the Hunan and Jiling sites. For example, on the NPK treatment pH declined from 5.7 at the start of the experiment to 4.5 after 12 years as a result of the acidifying effects of the Urea fertilizer. However, adding manure mitigated the effect of the Urea. Thus, in the NPK + M treatment soil pH was 5.9 after 12 or 13 years and was 6.7 in the manure-only treatment.

The present results show that fertilizer application plays an important role in maintaining or increasing crop productivity and soil fertility. Application of NPK led to higher crop grain yields and higher soil fertility. However, the effects of N application on soil acidity posed serious concern. Combined fertilizer and manure/straw inputs tended to produce greater benefits in crop yield and soil fertility than application of fertilizers alone. In general, as crop production remains a soil-based industry, maintenance or increases in productivity are unlikely to be obtained without ensuring that plants have an adequate and balanced supply of nutrients.

4. Current progress in integrated nutrient management

Sustainable agricultural production incorporates the idea that natural resources should be used to generate increased output without depleting the natural resource base. However, despite past achievements in crop production in China, there has still low resource efficiency and damage to the environment caused by both the over- and under-application of fertilizers, de-coupling between crop production and animal production, and poor management of resources.

The overall strategy therefore is to focus strongly on integrated nutrient management (INM). The strategy is expected to further increase crop yields to be able to feed the growing population and maintain the high yields in a sustainable way. This strategy also emphasizes the integrated use of nutrients from fertilizers, wastes (from both agriculture and industry), and from soil and environmental sources, such as atmospheric deposition and irrigation water. Nutrient management should also be integrated with sound soil management practices and other farming techniques such as high yielding cultivation systems (Zhang et al., 2005).

With the support of the Chinese Ministry of Agriculture and the National Natural Science Foundation of China, a large scale project has been carried out since 2002 which features integrated nutrient management systems for 12 cropping systems at 58 sites across the country (Figure 4.1). The N fertilizer applications have been split to match crop requirements at different growth stages based on the total fertilizer N rate required at the specific sites to minimize N losses from the soil-plant system. Fine-tuning to top-dressing was achieved using N-kit and SPAD/LCC. The fertilizer P or K management focuses on maintenance of adequate soil available P or K levels to ensure that P or K supply does not limit crop growth and N-use efficiency. Therefore, the maintenance fertilizer P or K rates are recommended through constant monitoring of soil nutrient supply capacity (Wang et al., 1995).

These newly adopted nutrient management systems could reduce N fertilizer inputs by 5-30 percent and increase crop N recoveries as well as grain yields by 10-15 percent as compared to traditional practices. The integrated nutrient management strategies for selected cropping systems are presented below.

Figure 4.1 Research on nutrient management in different cropping systems

4.1 Rice production systems

China has an area about 2.9 million ha of rice production, accounting for 20 percent of the world total rice production area. Rice production consumes about 37 percent of the world total N fertilizer used for rice production. The N application rate for single rice cropping is about 180 kg/ha, which is 75 percent higher than the world average. For example, the N fertilizer application rate is ~270-300 kg/ha in the Taihu area of Jiangsu province, and sometimes up to 350 kg/ha.

The recovery efficiency of applied N fertilizer is usually 30-35 percent in China (Zhu, 1997), which is lower than the world average. In Jiangsu province where excessive N was applied to rice, recovery efficiency (RE, i.e. uptake efficiency) of applied N fertilizer is only 19.9 percent. Low fertilizer-N use efficiency with high N loss has directly or indirectly led to a series of environmental problems such as groundwater pollution and eutrophication of surface waters. Low fertilizer-N use efficiency and poor N management lead to excessive N application and increased production costs while grain yields continues to decline, thus, reducing the profits of rice farmers. It is therefore becoming a large challenge for rice production, to increase fertilizer-N use efficiency by crops through improving rice nutrient management (Cassman et al., 1996). This is a very important step towards sustainable development of agricultural production and environmental protection.

4.1.1 Technical measures taken on INM for rice

Rice nutrient resource management has been developed and implemented throughout China in cooperation with IRRI since 2002. There are 7 rice-producing provinces taking part in the programme on rice nutrient resource management, including Jiangsu, Zhejiang, Hunan, Hubei, Guangdong, Heilongjiang and Hebei. The overall nutrient management strategy has focused on integrated soil nutrient utilization and optimizing fertilizer-N application rate and timing in combination with maintenance of P and K supply.

4.1.1.1 Determination of nutrient utilization and total fertilizer N application

The total fertilizer-N application rate for a rice field can be estimated by calculating the difference in N budget between N requirement and N supply from the soil (Dobermann and Fairhurst, 2000). On average, irrigated rice needs to absorb 17.5, 3.0 and 17.0 kg N, P and K, respectively, to produce each tonne of grain yield. The crop requirement can be calculated from a yield target selected as 75-80 percent of the climate yield potential. Soil nitrogen supply capacity is estimated using the yield of nutrient omission plots including nutrient inputs from atmospheric deposition (rainfall and dust), irrigation, floodwater, and sediments (dissolved and suspended nutrients) and biological N₂-fixation. The soil nitrogen supplied capacity in some areas of China are presented in Table 4.1.

Table 4.1 Soil N supply capacity in some areas of China (2003-2005, kg/ha)

4.1.1.2 Splitting fertilizer applications at different growth stages

Based on soil nitrogen supply capacity, the basal fertilizer N is recommended. Top-dressings are applied through SPAD/LCC fine-tuning. Therefore, the splitting pattern is in accordance with soil N supply capacity, crop growth stage, cropping season, the variety used, and the crop establishment method. Critical levels of SPAD and LCC and the corresponding fertilizer application rates have been developed for fertilizer-N distribution at different growth stages in China.

If SPAD/LCC is not available to farmers, a simplified distribution method for fertilizer N can be used according to the nutrient uptake patternns on nutrient management (e.g. 35-40 percent for basal fertilization, 20-25 percent for early tillering, 25-30 percent for panicle initiation and 0-10 percent for heading).

4.1.2 Impact and outlook of INM

The project on rice nutrient resource management has been carried out since 2002. The large-area results show that fertilizer N rates were reduced by 20-30 percent, input costs were reduced by 20 percent, and fertilizer-N agronomic use efficiency increased by 20-80 percent, but rice yields can be maintained at stable high levels (Table 4.2). The strategy of INM for rice is being adopted by more and more farmers and agricultural extension technicians. Nevertheless, it is very important for the further extension of INM to simplify the nutrient management technology and integrate high-yield cultivation practices, which correspond well with agronomic conditions.

Table 4.2 Effect of optimized nutrient resource management on rice production in 2004

AE = Agronomy efficiency

4.2 Vegetable production systems

Since the 1980’s smallholders have converted to vegetable production on a large scale from subsistence level to marketable sales and vegetable production has developed very rapidly. Until 2004, the total cultivated area was 17.6 million ha and vegetable production was 550.6 million tonnes (Ministry of Agriculture, PRC, 2005). In addition the area under greenhouse vegetables was over 250 million ha in 2003, which was about 350-fold higher than in 1980. Greenhouse production has become economically important in some regions, especially in Shandong and Hebei provinces. Vegetable planting and related commercial processes has probably been given the highest priority in agricultural enterprises and become the most important way to solve the problem of unemployment, especially for women, in many rural counties.

4.2.1 Overuse of fertilizers in vegetable production

Unlike cereal crops, most vegetable crops have high water and nutrient requirements due to very high biomass production over a relatively short growth period (Schenk, 1998). However, due to a lack of optimized management and technical assistance, large inputs of artificial fertilizer, organic materials, or a mixture of both, are very common (Liu, 2000; Wu, 2002). According to Chen et al. (2004), more than 35 percent of 132 greenhouses with tomato planting surveyed received >1 000 kg N/ha from organic and inorganic fertilizers in Beijing. Ju et al. (2004) reported that in Huimin, Shandong province, the average nitrogen application rate was as high as 2 848 kg N/ha per year, which was about 10-fold higher than crop demand. Organic manure, 50 percent as chicken manure, supplied over 50 percent of total NPK inputs for vegetable production in the survey (Chen, et al., 2004). Furrow and flood irrigation systems are still dominant in China and frequently used to apply at least 60-100 mm of water every one to two weeks, depending on the climatic conditions, which is a general rule of thumb to satisfy the expected peak demands.

Overuse of fertilizers is not only wasteful, but also damaging to both crop quality and the environment. For example, high inputs of nitrogen make vegetable field soils sensitive to accumulation of nitrate, until eventually nitrate leaching occurs when excessive irrigation water is applied or during extreme rainfall events. A survey in northern China showed that high nitrogen fertilization had led to high nitrate concentrations in groundwater (over 50 percent of investigated wells above the European Union drinking water threshold of 50 mg/L), and crop recoveries below 40 percent of applied N fertilizer (Zhang et al., 1996). In addition, greenhouse vegetable soils can become an important source of nitrous oxide emissions with increasing intensity of cultivation.

4.2.2 Technical measures taken on INM for vegetable production

N-expert system

One promising solution to optimize the supply of nitrogen in vegetable production is the N-expert system which was introduced by Fink and Scharpf (1993). Since 1998 the N-expert system with different irrigation regimes has been modified for a rotation of amaranth (Amaranthus tricolor L.), spinach (Spinacia oleracea L.) and cauliflower (Brassica oleracea L.) on the North China Plain. Some parameters, notably target yields and net mineralization rates, were modified according to local conditions, and most were referenced from the database of the N-expert system in the first cultivation year of the rotation. Parameters had been tested in earlier field experiments and were fine-tuned in later experiments. Three-year results showed that there is considerable potential for the use of the N-expert system for sustainable vegetable production on the North China Plain (Chen et al., 2005).

N recommendations based on Nmin supply in the rooting layer

To ensure crop-specific N demand, considering crop N uptake, soil Nmin buffer and inevitable N losses (including leaching, denitrification and NH₃ volatilization), the required N fertilizer is also calculated from the difference between target values of N supply and initial soil Nmin (or nitrate-N in dry land soil), and other N inputs, e.g. N through irrigation or deposition. Tang et al. (2004) suggested that the target value of N supply for N top-dressing was 300 kg N/ha for each event of fertilizer N application within the growing season of greenhouse tomato at a fruit yield level of 70 t/ha with conventional furrow irrigation in Shouguang, Shandong province. The potential for reducing N fertilizer use target value of N supply was 62-81 percent, with conventional organic manure application and irrigation regime, in a tomato cropping system from 2002 to 2005 in Shouguang (He et al., 2005). The soil nitrate content at 0-0.9 m soil depth in the recommended treatment was much lower than that in the conventional N treatment within the seasons investigated. The cumulative emission of nitrous oxide in the optimal treatments was reduced by 38 percent compared with traditional farming practice throughout the three seasons.

Synchronizing water and nutrient supply techniques

Using a balance strategy to maintain the optimum nutrient and water levels in the root zone is necessary, for example 60 kg N/ha were saved for spinach growth in open fields by using sprinkler irrigation because of a reduction in nitrate leaching from the root zone compared with conventional furrow irrigation (Li et al., 2004). Compared to furrow irrigation, fertigation techniques could save 40-50 percent of applied fertilizer without any reduction in yield or quality (Sui et al., 2001).

In the present study, the potential to reduce nitrogen applications has been seen, with the objective of environment protection and maintaining crop yields in integrated nutrient management systems in vegetable production. However, the technical measures mentioned above were only adopted on a limited scale by farmers for vegetable production. Technical measures should be simplified so that farmers can easily use them with less expensive equipment. Furthermore, to convince decision-makers as well as farmers and extension advisers to accept the idea of optimized N fertilizer management and optimized irrigation, demonstration experiments or demonstration farms at various locations in the region have to be installed and combined with teaching in pilot projects.

4.3 Wheat-maize rotation systems

The North China Plain (NCP), located in the northeast of the country, covers on area of 0.444 × 10 ⁶ km ² and is one of the most important crop production areas in China. Winter wheat-summer maize is the main cropping rotation on the North China Plain. Usually, winter wheat is seeded at the beginning of October and harvested at the beginning of June in the following year, and afterwards summer maize is immediately sown and is harvested at the end of September. This rotation system produces about 40 percent of wheat and 28 percent maize on a national basis (Yang, 1991).

4.3.1 Excess N fertilizer application and NO₃-N accumulation

However, the survey showed that farmers use excessive N fertilizer in this cropping system. For example, in high-yielding regions of the NCP, N fertilizer application is usually over 500 kg N/ha for wheat and maize together. According to Cui’s survey, the average N application rates were 424 kg N/ha for winter wheat (59 kg manure N/ha, n = 370) and 249 kg N/ha for summer maize (n = 368) in Huiming county, Shandong province, which was about 2-fold higher than the total nitrogen demand of wheat and maize (Cui et al., unpublished data). Overuse of fertilizers is not only wasteful, but also damaging to the environment. For example, soil NO₃-N is prone to accumulate under large N fertilizer applications. As shown in Table 4.3, the average accumulation in soil to 90 cm depth in farmers’ fields were 233 kg N/ha after the wheat harvest and 292 kg N/ha after the maize harvest. If the accumulated NO₃-N is not utilized by crops it is eventually lost through leaching and/or denitrification. Thus, N fertilizer recommendation techniques, which take into account NO₃-N accumulation in soil, are urgently required in NCP.

4.3.2 N fertilizer recommendations based on soil testing

For N fertilizer recommendations that take adequate account of soil N supply, a Nmin method was developed that utilized a test of soil Nmin and gave N fertilizer recommendations for two or three different growth stages during the crop growing season (Chen, 2003). For example, the growth of summer maize was divided into three periods: from sowing to the three-leaf stage, three leaves to the ten-leaf stage and the ten-leaf stage to harvest. The N rate in each growth stage was dependent on target N demand and measured soil Nmin. In this method, the effects of N mineralization, immobilization and N losses on plant available N during the preceding growth stage are included in the results of the next soil Nmin analysis as well as in the next N fertilization recommendation. Therefore, synchronization of soil N supply (Nmin in the rooting layer), fertilizer N application and subsequent crop demand for N can be attained to a certain degree.

The present results show that low N rate and high N-use efficiency could be achieved without any reduction in grain yield by optimized N management. During the winter wheat growing season, N fertilizer application rate in the optimum N treatment was saved by 69 percent from 369 to 98 kg N/ha compared to the conventional N treatment, and N fertilizer efficiency was improved from 15 percent to 44 percent efficiency of recovery and from 3 to 10 kg/kg agronomic efficiency. Similar results were also found for summer maize.

N fertilizer saving and N fertilizer efficiency improvement can be partly ascribed to adequate crop utilization of soil accumulated N as well as mineralized soil N and N inputs from environmental sources (e.g. irrigation and rainfall) during the crop growing stage. As shown in Table 4.4, during the winter wheat growing season, initial soil Nmin supply (soil Nmin after harvest – soil Nmin before sowing) and apparent N mineralization rate were 48 kg N/ha and 28 kg N/ha, which account for 32 percent and 18 percent of crop N uptake. During the summer maize growing season, apparent N mineralization rate is 100 kg N/ha and is 56 percent of crop N uptake.

This N recommendation method has been widely tested the in north of China. It is one solution for good synchronization of soil N supply, fertilizer N application and subsequent crop N demand and reduced residual soil nitrate-N accumulation in the rooting layer after harvest.

4.4 Rice-wheat rotation systems

Rotations of rice and other crops are practiced widely along the Yangtze River Basin where they occupy a total area of about 13 million hectares (Timsina and Connor, 2001) and they account for 30 percent of cereal crop production in China (Fan, 2005). However, a developing water crisis and environmental pollution arising from improper nutrient management threaten the sustainability of these crop rotations and a trend of declining or stagnating yields has been observed (Bouman, 2001; Ladha et al., 2003). Improvements are therefore required in the management of nutrient resources, soil, water and straw. Since 2000, integrated nutrient management (INM) has being developed in rice-wheat cropping systems (R-W) in southwest China. The strategy of INM in R-W and its impacts on farming livelihoods are presented below.

4.4.1 Strategy of INM in R-W

Non-flooded straw/plastic film mulching cultivation

Non-flooded mulching cultivation, a new rice water-saving cultivation technique, was introduced into China in the late 1980’s. Lowland rice fields are irrigated and a shallow water layer is maintained before transplanting. The soil surface is then covered with plastic film (0.005-0.010 mm thick), straw or paper and the soil is maintained at 70-90 percent of water holding capacity or rain fed during the rice development stage, depending on rainfall and the underground water level (Fan et al., 2005). After the rice harvest each year, the plastic film is moved from the field by hand and wheat is sown directed into the zero tillage soil.

Several long-term field studies indicated that non-flooded mulching cultivations may be alternative options for farmers using rice-wheat rotations for saving water, enhancement or maintenance of system productivity (Table 4.5) and soil fertility (Fan et al., 2005). At present crop yield levels (rice, 6.6-7.6 t/ha; wheat, 4.8-5.5 t/ha), plastic film mulching cultivation for rice can maintain stable soil and crop productivity over time. Non-flooded straw mulching cultivation resulted in decreased rice yields, but system productivity could be maintained. Non-flooded straw mulching cultivation led to increases in soil organic matter, soil total nitrogen, Olsen P, and exchangeable K in the topsoil (Fan et al., 2005).

^a Within each column, values with the same letter are not significantly different by LSD at the 5 percent level across all cultivation systems.
  ^b Combined rice and wheat grain yields.

4.4.2 Combination of cultivation techniques and nutrient management

In the traditional approach to nutrient management the emphasis was solely on fertilizer application, with no regard for the efficient utilization of indigenous nutrients or integration with other agricultural techniques. The following describe the strategy of nutrient management integrated with cultivation techniques in R-W.

In the rice season, this cultivation technique involves planting 3 young seedlings per hill in a triangular pattern with 10-12 cm spacing between the plants. The hills are planted in a staggered 50 × 50 cm grid. In the wheat season, the wheat seeds are sown directly into the zero tillage soil at two spacing of 10 cm × 15 cm and 10 cm × 25 cm. As for the nutrient management systems, a description has been presented in part 4.1 of this paper. Briefly, N is applied in split dressings according to crop development stage. Fertilizer P and K rates are recommended using the constant monitoring soil nutrient supply capacity method.

Contrasting experiments in different farmers’ fields in Sichuan province and Chongqing showed that integrated systems could solve the problems of insufficient or excessive N fertilizer application.

Compared with the average N application rate (156 kg N/ha) and the average rice yield (6 890 kg/ha) in farmers’ fields (n = 12), integrated nutrient management saved N fertilizer application by 24 percent while increasing rice yields by 24 percent. In the wheat season, integrated nutrient management decreased N fertilizer inputs by 30 percent, without any decline in wheat grain yield (Fan, 2005). This indicates that integrated system is one of the feasible solutions for harmonization of nutrient inputs, crop production and environmental protection in R-W.

4.4.3 Dissemination of INM and its impacts on farming incomes

So far, these systems have been used widely by farmers in some areas of Sichuan province, southwest China. The farmers’ association and local government played crucial roles in the dissemination of INM. For example, in Dongxi town, Jianyang city the government will give certain subsidy to those farmers who use integrated nutrient management systems in the first year. By joining the farmers’ association, farmers can receive INM technical training and share experience and information with each other. This has greatly facilitated the promotion of INM in Dongxi town.

Current evidence indicates that INM offers benefits to farmers (Li Xiaoyun personal communication) and this is why farmers were willing to continue to adopt integrated nutrient management systems in spite of tha lack of a local government subsidy during the second year. As shown in Table 4.6, the gross margin per ha for rice production with integrated nutrient management systems in 2003 was 6.93 times that with conventional methods when all imputed labour costs are included, and 2.11 times that when the costs of labour are not considered. These factors are higher than those in 2004, which confirms the value of integrated nutrient management to farmers in a ‘bad’ year.

¹ Cost of labour included
² Cost of labour included
³ Integrated nutrient management system
⁴ Ratio of SRI margin compared with conventional one

4.5 Cotton production systems

China is one of the largest cotton producing and consuming countries worldwide. Textile and garment industries make a significant contribution to the national economy. In more than 100 major cotton-growing counties, the cotton industry makes a sixty-percent contribution to local GDP. Moreover, there are 150 000 000 people directly or indirectly involving in the cotton industry. Cotton growing regions in China can be divided into three main geographical regions, the Yangtze river cotton belt, the Yellow river cotton belt and the northwestern cotton belt.

Xinjiang, one of the main production areas, occupies first position in both total yield and yield per unit area. The cotton planting industry produces major incomes for local farmers. However, there are many problems in nutrient management in cotton production, especially the over application of nitrogen. It is therefore essential to develop a sustainable nutrient management system based on a better understanding of nutrient demand pattern and dry matter accumulation of cotton.

4.5.1 Technical measures taken on INM for cotton

The integrated nutrient management system for cotton involved the optimized nitrogen management technique based on the soil Nmin test and nitrogen nutrition diagnosis index system. Phosphorus and potassium fertilizers were recommended through constant monitoring as shown in this paper.

4.5.2 Soil Nmin test for total N fertilizer recommendations

The source of the nutrient includes dry and wet deposition, biological nitrogen fixation, irrigation, artificial fertilizers and organic fertilizers. However, nitrogen sources for cotton also include the soil mineral nitrogen (Nmin) pool which exists in soil solution in nitrate and ammonium forms. The total amount of soil mineral nitrogen in the rooting layer can be used to recommend soil nitrogen supply. The soil Nmin was considered as available as nitrogen fertilizer, with both sources contributin to yield formation in cotton. The quantified model of the soil nitrogen supply for some target yields (namely nitrogen demand target value) could be achieved:

Nitrogen supply for target yield = N from environment (dry and wet depositions, biological nitrogen fixation, irrigation) + mineral nitrogen in the rooting layer (Nmin) + fertilizer nitrogen

Because the content of N from the environment would be relatively low, the formula could be simplified as:

Nitrogen supply for target yield = inorganic nitrogen in rooting layer (Nmin) + fertilizer nitrogen

Fertilizer nitrogen = nitrogen supply for target yield – the mineral nitrogen in rooting layer (Nmin)

Studies have shown that the nitrogen supplying capacity of the soil (i.e. the sum of the inorganic nitrogen in the rooting layer and the nitrogen fertilizer) was highly correlated with crop yield. Regression equations between the different nitrogen supply values of the soil and the corresponding cotton yields are established. Based on this correlation, the soil nitrogen supply corresponding to the highest yield is calculated as the optimum nitrogen supply. The formula can be quantified as:

Fertilizer nitrogen = optimum nitrogen supply – mineral nitrogen in the rooting layer (Nmin)

4.5.3 Split applications according to cotton growth and development and N uptake

Cotton is a high temperature and light-loving fiber crop with a long life cycle. In general, the development stage is 145-175 days which can be divided into four discrete stages, namely seedling, square, flowering and boll-forming, and boll-opening. Cotton demands much mineral nutrient for gro by the rice crop at a specific site with fixed distribution rates based on previous investigatiowth and development. On average, the nutrient demands for producing 100 kg lint are 12-15 kg N, 5-6 kg P₂O₅ and 12-15 kg K₂O. Numerous studies have shown that cotton requires little nutrient at seedling stage, just about 1 percent of the total nutrient demand, until square appearance the nutrient consumed is 3 percent, from square appearance to flowering 27 percent, from bloom appearance to late boll production takes about 60 percent and the boll-opening stage takes 9 percent of the total nutrient. Thus, flowering and boll-forming are the key stages for cotton to be supplied with fertilizers. Table 4.7 shows the N fertilizer application rate and timing for different target yield levels in Xinjiang province.

Both field experiments and demonstration results indicate that integrated nutrient management enhanced the nitrogen utilization efficiency greatly and increased cotton yields in the Xinjiang cotton belt. In 2004 in a 3 300-ha demonstration area in Xinjiang, pre-sowing N application ratio was adjusted from 60 percent of total N fertilizer to 40 percent based on soil test values and the N demand of the cotton. Compared with the traditional N application method, the optimized N method reduced N application rate by 9 percent, but the lint yield was retained. Net income increased by 650 Yuan RMB per ha.

4.6 Intercropping systems

Intercropping continues to be widely employed in both tropical and temperate regions (Vandermeer, 1989). Both wheat/maize and faba bean/maize strip intercropping are long established practices in major grain production systems in northwest China, especially in irrigated areas that are limited to only one cropping season annually due to temperature constraints. The area under wheat/maize intercropping in 1995 was 75 100 ha in Ningxia province, producing 43 percent of the total grain yield for the area. In Gansu province 200 000 ha are intercropped annually. Faba bean/maize intercropping occupies a smaller area than wheat/maize intercropping. However, little was known about the relationship between interspecific interactions (competition and facilitation) and efficient nutrient utilization in these intercropping systems by agronomists and farmers. Farmers applied more and more chemical fertilizers in the systems, which resulted in low fertilizer efficiency and high risk of environmental pollution. We have studied interspecific interactions between intercropped species for many years, and have used our results in agricultural production for several years.

4.6.1 Influence of interspecific facilitation on nutrient utilization and nutrient
management practices in cereal/legume intercropping

Nutrients mobilized by efficient species contribute to inefficient species maize

Interspecific root interactions between intercropped faba bean and maize played an important role in the yield advantage and nitrogen and phosphorus acquisition by the intercropping system (Li et al., 1999; 2003a). Faba bean showed a greater capacity in mobilizing insoluble phosphorus than maize and an interspecific facilitation of P uptake was also observed. Phosphorus uptake by maize from Al-P, Fe-P, or Ca-P was 86 percent, 49 percent and 19 percent higher when their roots were intermingled than when their roots were separated. One mechanism behind the facilitation was that faba bean had a greater ability to acidify its rhizosphere than maize. Biomass and P uptake of wheat were significantly increased when the roots of intercropped wheat and chickpea intermingled compared to treatments with either a solid root barrier or nylon mesh, regardless of whether P was supplied as phytate or FePO₄ . Wheat is less able to use organic P than inorganic P, whereas chickpea can use both P sources equally effectively. The mechanism behind this facilitation was that chickpea had greater rhizosphere acidification ability and more activity of phosphatase released from the roots.

Both field and greenhouse experiments showed that the improvement in the Fe nutrition of peanut intercropped with maize was mainly caused by rhizosphere interactions between the two crops. In the treatment where the roots of the two species were physically separated, the peanut plants were chlorotic, with most of the peanut plants in the row closest to maize showing slight signs of chlorosis. In contrast, in the treatment where a 37-|im nylon mesh was inserted into the ground between the two crop species to prevent direct root contact but allow interactions through mass flow and diffusion, the peanut plants in rows 1 and 2 remained completely green, while those in rows 3-10 were chlorotic to varying extents (Zuo et al., 2000).

Nutrient management practices - reducing rates of nitrogen and phosphorus in cereal/ legume intercropping

The above studies showed that nitrogen fixation was increased by cereal/legume intercropping (i.e. maize/faba bean intercropping), compared to sole faba bean. At the same time, phosphorus utilization by cereals can be enhanced by associated legumes (i.e. faba bean or chickpea). Therefore, the rate of nitrogen fertilizer application can be lowered to 150-200 kg N/ha by inoculating the legumes with selected rhizobium strains, from the conventional 250-300 kg N/ha for cereal/legume intercropping in this area, which saves substantial amounts of nitrogen fertilizer and reduces the risk of environmental pollution. Field experiments showed that phosphorus fertilizer could also be reduced to 50-75 kg P/ha from the conventional 120-150 kg P/ha.

4.6.2 Competitive-recovery production principle and nutrient management in cereal/
cereal intercropping

Yield advantage of dominant species from interspecific competition

In the wheat/maize or wheat/soybean intercropping systems in northern China, there is a 70- to 80-day overlapping growth period that causes intense interspecific interactions between the intercropped species. The yield advantage in the border row of intercropped wheat probably derived from the differences in interspecific competitiveness, and wheat was more competitive relative to maize and soybean during overlapping growth periods in the wheat/soybean and wheat/maize intercropping systems (Li et al., 2001a).

Recovery or compensation of subordinate species at a late growth stage

There is a recovery of nutrient uptake and growth after harvest of the earlier-maturing species, which makes the later-maturing species compensate for impaired early growth once the early-maturing species has been harvested (Li et al., 2001b).

Nutrient management practices based on the ‘competition-recovery principle’ in cereal/ cereal intercropping

Farmers have usually applied more than 350 kg N/ha, and sometimes up to 600 kg N/ha in cereal/cereal (i.e. wheat/maize and barley/maize) intercropping to get more than 12 t/ha of grain yield in irrigated areas of northwest China. Based on our proposed ‘competition-recovery principle’, wheat or barley, which had much more competitive ability for soil nutrients at the co-growth stage of the two crops, can utilize more soil nutrients, whereas maize growth and yield are more dependent on fertilization. Therefore, we emphasized that the key point of nutrient management lay in the subordinate species, for instance maize, at late growth stages in this type of intercropping. We recommend that a total of 230-300 kg N/ha is distributed as 120 kg N/ha basal application for both wheat and maize, 80 kg N/ha at elongation state for maize only, and 100 kg N/ha for maize only at the pretasseling stage. The application rate of nitrogen fertilizer was greatly reduced by these recommendations compared to conventional rates of more than 350 kg N/ha. All in all, intercropping plays an important role in efficient nutrient management if a suitable crop combination is chosen.

4.7 Lychee production systems

Tropical fruit trees such as lychee are important cash crops which occupy an area of about 800 000 ha in south China (Jiang, 1997). However, in traditional farming practice, nutrient management does not always fit lychee production. Secondly, in order to promote earlier production, girdling has been a common measure after fertilization. This does not only result in losses of fertilizer but is also harmful to the lychee tree itself. Thirdly, lychee orchard soils are hard, especially around the trunk, resulting in the uneven crown of lychee trees in the orchard. As a result one can often observe problems such as poor fruit setting rate, lower yields, and alternation of high and low yield harvest years caused by malnutrition (Menzel, 1987) and by poor understanding of NPK nutrition during the annual growth cycle of the tree. Furthermore, poor nutrient management in lychee production systems (Lin, 2001 and Dai et al., 1995) results in damage to the environment as well as lost income for farmers. An integrated nutrient management system was required for this production system.

4.7.1 Technical measures taken on INM for lychee

The overall nutrient management strategy for lychee focused on nutrient requirements and consumption based on understanding the growth pattern, and NPK nutrient requirement pattern and yield response to set of specialty compounds.

In this context, the annual growth cycle of lychee is divided into three phases. The first is from last harvest to the autumn shoot maturation stage (ASMS), the second from ASMS to the flowering stage (FS) and the third from FS to the next harvest. The nutritional state of each phase is characterized by its physiological properties. During the first stage, the nutrient supply should match the recovery of the body of the tree and developing fruiting shoots. Then, development of flower buds and flowering as well as fruit setting will be the main physiological change in the second phase. In the last stage the main physiological property is fruit enlargement and maturation.

Fertilizer is required in the different growth phases. The first fertilizer application with high nitrogen compound fertilizers (e.g. 20-10-10 or 22-6-12) should be made 1 to 2 weeks after harvest and the dosage is half of the total fertilizer required (TFR) in an annual growth cycle. The second application of high P and K compound fertilizers (e.g. 10-20-20 or 8-16-16) should be made at autumn shoot maturation stage and the suggested dosage is a quarter of the TFR. The last application is recommended to be a high K formulation (13-10-21 or 10-6-24) at the fruit setting stage and the dosage is another quarter of the TFR. The amount o compound fertilizer used annually is around 3 kg per tree every growth cycle for an 8 to 15-year-old tree.

5. Major issues in dissemination and implementation of INM

5.1 Analyse the policy measure effects on INM dissemination and implementation

Policy has played an important role in the development of China’s agriculture, particularly over the past two and a half decades. A few recently implemented agricultural policies that may have significant implications for dissemination and implementation of INM are briefly stated and analysed below.

5.1.1 The No. 1 centre document

There are multiple pressures stemming from resource limitations, environmental pollution and population growth for China’s “grain security”. The Chinese government regards agriculture as the primary development field of the national economy in the 21^st century. The policy goal is focused on raising the agricultural product yields, increasing farming incomes and securing the development of rural society. Such issues can be reflected in the No. 1 centre documents which were released in 2004 and 2005. The new concept involving the agricultural production chain should be adapted to obtain high production, product quality and environmental friendliness simultaneously. Therefore, integrated nutrient management technology would be one of the core technologies of this concept.

5.1.2 Increasing subsidies for grain production

Agricultural production has been promoted by agricultural subsidies from central government and most of the provincial governments. For example, in 2004 the subsidy for grain production was 11.6 billion Yuan. The scope of the subsidy was extended from soybean to a number of crops including wheat, maize and rice, ranging from 150 Yuan/ha to 225 Yuan/ha. The central government also supplied 70 million Yuan to help some farmers to buy large farming machinery, which drove the local governments to supply more than 400 million Yuan for the same purpose. The measure of the subsidy for grain production and maintenance of high selling prices for grain products stimulated farmers to produce grain. Farmers were encouraged to adopt new technologies with the objectives of increasing yields and high input efficiency.

5.1.3 Abolish agricultural taxes - farmers’ initiative

In 2004, agricultural production was promoted by abolition of agricultural taxes by central government and most provincial governments. The central government abolished the tax for special products except for tobacco. In 2005, except for three provinces in the main provinces of China in which agricultural taxes decreased to 2 percent, agricultural taxes have been abolished. This has save the farmers 22 billion Yuan.

5.1.4 Increasing agricultural and rural investment

In 2004, the investment of central government to agriculture, farmers and rural society increased 22.5 percent over 2003. The agricultural investment accounted for 34.2 percent of national debt investment. The expenditure for supporting rural production, capital construction and science and technology promotion increased by 49 percent, 7 percent and 6 percent, respectively. In 2005, significant further investment was made to agriculture, farmers and rural society. Centre government also plans to promote rural construction in seven aspects including excellent seed systems, technological innovation and application systems, animal and plant protection systems, agricultural products safety systems, marketing information systems, resources and environmental protection systems, and management and services systems. The increasing investment and basic construction will increase the influence of technology on every aspect of agriculture.

In general, these policy measures have led to favourable crop production outputs in 2004 after several years of decreasing outputs. Total grain production was 469.47 mt, which increased by 9.0 percent compared to 2003. Grain yield was 4 620.5 kg/ha, which was the highest recorded and increased by 288 kg and 6.6 percent over 2003. However, 75 percent of the increased grain production came from changes in area, with just 25 percent coming from changes in yield. Considering the limited potential to increase grain area in the future, it is becoming apparent that technology and policy measures should be given more importance to help improve integrated productivity of agriculture and increase grain yield per unit area of arable land.

5.2 Constraints in dissemination and implementation of INM

5.2.1 Low efficiency of agricultural technology extension services

China Agricultural extension systems are organized in administrative regions, following an order from the central authority to provincial and municipal levels, and to counties and villages. However, since the 1980s the efficiency of the agricultural technology extension system had became low and there were serious difficulties such as lack of investment, and poor training of technicians (Research Centre of Rural Economy, Ministry of Agriculture, 2005). According to a survey, farmers obtained more agricultural knowledge and experience from their neighbours than from the agricultural technology extension technicians. The low efficiency of the extension services also contributed substantially to the low contribution rate (34.1 percent during the ‘Eight Five’ Plan from 1990 to 1995) of science and technology to agricultural development (Fan and Guo, 1999). In recent years, despite rapid development of non-governmental agro-technological service, dissemination of agricultural technology is still very limited. For example, even with the various agro-chemical services set up by numerous fertilizer enterprises, there has still low technology transfer since their services have focused mainly on fertilizer sales rather than integrated nutrient management techniques.

5.2.2 Limitation of small holdings

According to the statistical data (National Bureau of Statistics of China, 2005), there were about 250 million rural households with an average of 0.6 hectares of sown area per rural household in 2004. The small-holding farms restricted nutrient management technical extension to a certain extent because the ratio of products for commercial goal is very low ascribe to its small area (He, 2000). On the other hand, 17 000 staff engaged in soil and fertilizer management in the extension system could not meet the technological demands of 250 million small-holders. Hu and Li (2004) indicated that in 2001 the ratio of agro-technological extension technicians to rural labourers was about 1/800 which is higher than in Europe (1/431) or north America (1/325). Furthermore, the agro-technological extension technicians allocated less than 30 percent of their working time to visiting farmers in villages (Hu et al., 2004). It was therefore difficult for most farmers to gain access to agricultural technicians and technology transfer.

5.2.3 Low level of education and insufficient trainings to improve agricultural
knowledge of farmers

Low level of education and insufficient trainings to improve agricultural knowledge of farmers in China is another constraint for extension of nutrient management technology. From a survey conducted by the National Bureau of Statistics of China the proportion of the population educated to a level above senior high school is only 20 percent. The rate is said to be even lower for rural people. Low level of education and lack of training has limited farmers’ understanding of the essence of fertilizer management and its implications on the environment. For example, when farmers’ education level was above senior, 74 percent of the farmers knew that excessive fertilizers would produce negative effects on groundwater. However, this number will decrease significantly when the farmers are more poorly educated (Ma and Zhang, 2001). Though about 70-80 percent of farmers know that fertilizer application rates should accord with soil fertility and target yield levels, most of them did not know how to determine the fertilizer rate and application time satisfactorily. Improvement of education, whether formal or informal, and updated technology training to farmers will make an important contribution to improve the provision of agricultural technical extension programmes.

References

Bouman B.A.M., 2001. Water-efficient management strategies in rice production. International Rice Research Notes (IRRN), 26, 17-22.

Cassman, K.G., G.C. Gines, M.A. Dizon, M.I. Samson, and J.M. Alcantara, 1996. Nitrogen-use efficiency in tropical lowland rice systems: contributions from indigenous and applied nitrogen. Field Crops Res. 47: 1-12.

Chen Q., Zhang X.S., Zhang H.Y., Christie P., Li X.L. 2004. Evaluation of current fertilizer practice and soil fertility in vegetable production in the Beijing region. Nutr. Cycl. Agroecosys. 69: 51-58.

Chen, Q, H.Y. Zhang, X.L. Li, P. Christie, D. Horlacher, H.-P. Liebig, 2005. Use of a modified N-expert system for vegetable production in Beijing region. J. Plant Nutri. (In press)

Dai Liangzhao, Guo Yangchu and Xie Xiyao, 1995. State of mineral nutrition of Lychee chinensis and effect of chemicals on controlling its nutrition condition. J. of Fujian Academy of Agricultural Science. 10(1): 48~53.

Dobermann A. and Fairhurst T. 2000. Rice: Nutrient Disorders & Nutrient Management. PPI, PPIC and IRRI.

Fan M.S., Jiang R.F., Liu X.J., Zhang F.S., Lu S.H., Zeng X.Z. and Christie P. 2005. Interactions between non-flooded mulching cultivation and varying N inputs in rice-wheat rotations. Field Crops Research, 91, 307-318.

Fan M.S., Liu X.J., Jiang R.F., Zhang F.S., Lu S.H., Zeng X.Z. and Christie P. 2005. Crop yields, internal nutrient use efficiency, and changes in soil properties in rice-wheat rotations under non-flooded mulching cultivation. Plant and Soil. 277, 265-276.

Fan Qizhou and Guo Youhuan, 1999. The rank of limitation factors for agro-technology extension in China. Agriculture Technological Economics, (23): 12-15.

Fink, M. and H.C. Scharpf, 1993. N-expert – A decision support system for vegetable fertilization in the field, Acta Hort. 339, 67-74.

He Weijin, 2000. Difficulties and countermeasures in agro-technology extension of China. Journal of Beijing Agricultural Administration College, 14(4): 10-11.

He, F.F., W.L. Xiao, J.L. Li, Q. Chen, R.F. Jiang, F.S. Zhang, 2006. Application of integrated nitrogen management in greenhouse tomato production. Plant Nutrition and Fertilizer Science (Beijing). 12 (In press).

Hu Ruifa and Huang Jikun, 2001. Current situation of investment of agro-technology extension and its effect in China. Strategy and Management, (3), 25-31.

Hu Ruifa and Li Liqiu, 2004. compare of agro-technology extension in China to other countries. Science & Technology Review, (1): 26-29.

Hu Ruifa, Sun Dingqiang and Dong Xiaoxia, 2004. Behaviour of agro-technology extension technicians as going to countryside and its affected factors in China. China Rural Economics. (11): 29-35.

Huang Jikun, Scott Rozelle, Carl Pray and Wang Qinfang, 2002. Plant Biotechnology in China. Science, 295: 674-677.

Jiang Xuelin, 1997. Potassium nutrition and potassium fertilization for Lychee Chinensis Sonn J. Guangxi Agricultural Science. No. 1: 23-26.

Ju X.T., Liu X.J., Zhang F.S., Roelcke M. 2004. Nitrogen fertilization, soil nitrate accumulation, and policy recommendations. Ambio. 33(6): 300-305.

Ju, X.T., Liu, X.J., Zhang, F.S. and Roelcke, M., nitrogen fertilization, soil nitrate accumulation, and policy recommendations in several agricultural regions of China. Ambio, 2004, 33(6), 300-305.

Kou, C.L. 2004. Effect of nitrogen fertilization of different intensive cropping systems on environment in North China Plain. Ph.D. Dissertation of China Agriculture University.

Ladha J.K., D. Dawe, H. Pathak, A.T. Padre, R.L. Yadav, Bijay Singh, Yadvinder Singh, Y. Singh, P. Singh, A.L. Kundu, R. Sakal, N. Ram, A.P. Regmi, S.K. Gami, A.L. Bhandari, R. Amin, C.R. Yadav, E.M. Bhattarai, S. Das, H.P. Aggarwal, R.K. Gupta, P.R. Hobbs, 2003. How extensive are yield declines in long-term rice-wheat experiments in Asia? Field Crops Research, 81, 159-180.

Li L., Sun J.H., Zhang F.S., Li X.L., Rengel Z. and Yang S.C. 2001b. Wheat/maize or soybean strip intercropping. II. Recovery or compensation of maize and soybean after wheat harvesting. Field Crops Research 71, 173-181.

Li L., Sun J.H., Zhang F.S., Li X.L., Yang S.C. and Rengel Z. 2001a. Wheat/maize or wheat/soybean strip intercropping. I. Yield advantage and interspecific interactions on nutrients. Field Crops Research 71, 123-137.

Li L., Tang C.X., Rengel Z. and Zhang F.S. 2003. Chickpea facilitates phosphorus uptake by intercropped wheat from an organic phosphorus source. Plant and Soil 248(1-2), 297-303.

Li L., Yang S.C., Li X.L., Zhang F.S., Christie P. 1999. Interspecific complementary and competitive interactions between intercropped maize and faba bean. Plant and Soil 212, 205-214.

Li, J.L., H.D. Liu, X.S. Zhang, X.F. Wang, Q. Chen, 2004. Effects of different irrigation patterns on growth and nitrogen utilization of spinach under open field cultivation, Plant Nutrition and Fertilizer Science (Beijing). 10(4), 398-402.

Lin xiong, Zhang Xiaoqi, Li Chongyang et.al. 2001. Study on the physiology of nutrition in autumn shoots of lychee. Journal of South China normal University (Natural Science Edition). No. 4: 98~100.

Liu X.J., Ai Y.W., Zhang F.S., Lu S.H., Zeng X.Z. and Fan M.S. 2005. Crop production, nitrogen recovery and water use efficiency in rice-wheat rotation as affected by non-flooded mulching cultivation (NFMC). Nutrient Cycling in Agro-ecosystems, 71, 289-299.

Liu, Z.H. 2000. Characteristics of greenhouse vegetable soil and reasonable fertilization in Shandong province. Ph.D. thesis of 2000, China Agricultural University, Beijing (In Chinese).

Ma Wenqi and Zhang Lingmin, 2001. The farmers’ fertilizer application and the awareness of farmers on agricultural environment protection in China. UNESCO INRULED. 49-61.

Menzel, C.M. 1987. Lychee nutrition: a review. Scietia Hort. 3(3/4): 195-224.

Ministry of Agriculture, PRC. 2005. 2004 China Agriculture Statistical Report. China Agriculture Press. Beijing.

National Bureau of Statistics of China, 2005. China Statistical Yearbook 2005, China Statistics Press.

National Bureau of Statistics of China. 1949-2005. China Agriculture Yearbook. Beijing: China Agriculture Press.

Qiao Fangbin, Zhang Linxiu and Hu Ruifa, 1999. Analyses of behaviour of agro-technology extension technicians in China. Agriculture Technological Economics, (3): 12-15.

Research Centre of Rural Economy, Ministry of Agriculture, 2005. Survey and reform strategy of agro-technology extension system in China. China Rural Economics. (2), 46-54.

Schenk, M.K. 1998. Nitrogen use in vegetable crops in temperate climates. Horticultural Reviews. 22: 185-223.

Sui F.G., Wang Y.H., Chang Y.C., Makato Nogotomo, Naoya Chrishiky, Wunimuren, Shunji Inanaga. 2001. The efect of fertigation system on yield of sweet pepper and soil nitrate in greenhouse culture. Journal of Huazhong Agricultural University. 20(4): 358-362 (In Chinese).

Tang L.L., Chen Q., Zhang F.S., Li X.L. 2004. Nitrogen topdressing and feedback control in greenhouse tomato. Plant Nutrition and Fertilizer Science (Beijing). 10(4): 391-397 (in Chinese).

Timsina J. and D.J. Connor. 2001. Productivity and management of rice-wheat cropping systems: issues and challenges. Field Crops Research, 69: 93-132.

Wang Xinren Cao Yiping Zhang Fusuo Chen Xinping, 1995. Feasibility of a fertilizization method for keeping constant application rate of phosphorus by monitoring available phosphorus in the soil. Plant nutrition and fertilizer sciences, 1(3-4), 58-63 (In Chinese).

Wu, J.F. 2002. Study on diagnosis fertilization and environmental effect for non-polluted vegetables in Beijing. Ph.D. Dissertation. Huazhong Agricultural University, China (In Chinese).

Zhang F.S., Ma W.Q., Zhang, W.F. and Fan M.S., 2005. Nutrient management in China: From production systems to food chain. In: Li Chunjian et al. (eds.) Plant Nutrition for Food Security, Human Health and Environmental Protection. Tsinghua University Press, Beijing, China, pp. 618-619.

Zhang W.L., Tian Z.X., Zhang N., Li X.Q. 1996. Nitrate pollution of groundwater in northern China. Agric. Ecosyst. Environ. 59(3): 223-231.

Zhang, R.Q. Nutrient balance in agriculture fields of China. Ph.D. Dissertation, China Agriculture University, Beijing.

Zhu, Z. 1997. Fate and management of fertilizer nitrogen in agro-ecosystems. In: Z. Zhu, Q. Wen, and J.R. Freney (ed.) Nitrogen in Soils of China. pp. 239-279. Kluwer Academic Publishers, Dordrecht, the Netherlands.

Zuo Y.M., Zhang F.S., Li X.L. and Cao Y.P. 2000. Studies on the improvement in iron nutrition of peanut by intercropping with maize on a calcareous soil. Plant and Soil 220, 13-25.

¹ College of Resource and Environmental Sciences, China Agricultural University; Key Laboratory of Plant-Soil Interaction, Ministry of Education, Beijing 100094 China.

² Institute Agricultural Resources and planning, Chinese Academy of Agricultural Sciences (CAAS), Beijing 100081 China.

³ College of Resource and Environmental Science, Agricultural University of Heibei, Baoding 071001 China.

⁴ Fertilizer and Balance Fertilizer Laboratory, South China Agricultural University, Guangdong, 510642 China.