Farmer behavior and perceptions to alternative scenarios in a highly intensive agricultural region, south central China

Journal of Integrative Agriculture 2017, 16(8): 1852–1864 Available online at www.sciencedirect.com

ScienceDirect

RESEARCH ARTICLE

Farmer behavior and perceptions to alternative scenarios in a highly intensive agricultural region, south central China LI Hong-qing, ZHENG Fei, ZHAO Yao-yang Institute of National Land Resources Management, Hohai University, Nanjing 210098, P.R.China

Abstract Intensive agriculture has caused unintended environmental consequences, such as water quality degradation. It is necessary for policymakers to make proper planning of sustainable agricultural development. Using a Pressure-State-Response (PSR) framework, we conducted surveys focused on farmer behavior toward agriculture and environmental protection in 2009 and 2011. The surveys indicated that farmer behavior was complex and contradictory, and caused some environmental effects. Therefore, we used normative landscape scenario method to develop two scenarios. Both scenarios emphasized on stable economic growth along with water quality improvement and presented good effects. A feedback survey was organized in 2013 to interpret farmers’ perceptions of the alternative scenarios. The results indicate Scenario I is likely to be accepted by farmers; however, the beautiful rural landscape in Scenario II represents what farmers want, and Scenario I or II can be achieved by changing farm behavior in the future. By logistic regression model analysis, increasing agriculture benefits and new technology popularization were key factors affecting farmer behavior. Relevant policy implications on farmers were proposed. This paper showed how important to understand farmer behavior and perceptions to agricultural development, and a description of the alternative scenarios and policy implications are meaningful for policymakers to manage nature resources. Keywords: farmer behavior, alternative scenarios, PSR framework, water quality

to be a significant challenge to improve and protect water

1. Introduction Agricultural non-point source (NPS) pollution has been recognized as a major contributor to water quality degradation in highly intensive agricultural areas. NPS pollution from agricultural land use is a very complex problem. It continues

quality worldwide (Lam et al. 2011; Liu et al. 2013). Therefore, agricultural policies and farmland practices impact intensive agricultural land use, which in turn affects water quality, greenhouse gases, and biodiversity. Policymakers try to find effective policy instruments to alleviate water pollution or protect the environment (Wescoat and Johnston 2007; Blackstock et al. 2010; Stuart et al. 2014). These policies may limit some farmer behaviors, or encourage the farmer to adopt a new practice to mitigate water pollution.

Received 25 July, 2016 Accepted 23 November, 2016 Correspondence LI Hong-qing, Tel: +86-25-83786381, E-mail: [email protected] © 2017, CAAS. Publishing services by Elsevier B. V. All rights reserved. doi: 10.1016/S2095-3119(16)61547-2

However, sometimes farmers do not respond to policies as expected (Sattler and Nagel 2010; Willy et al. 2014). The study of farmers’ perceptions and decision-making behavior can play a significant role in the development of policies. Policymakers should greatly consider the farmer behavior

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

and perceptions during the policy-making process. Many conservation practices have been implemented for environmental protection, water quality, or biodiversity all over the world. To test the efficacy of these measures and recommend further scientific and stakeholder consultations, a review of farmer adoption decision behavior is necessary. Many empirical studies have been conducted to analyze the relationship between the adoption of conservation practices and farmers (Ryan et al. 2003; Lemke et al. 2010; Sattler and Nagel 2010), such as farmer’s adoption of the Nitrate Reduction Programme in Greece (Giovanopoulou et al. 2011), the Rural Environment Protection Scheme in Ireland (Murphy et al. 2014), pesticide control in Vietnam (Hoai et al. 2011), and water quality trading in the U.S. (Mariola 2012). These conservation practices were proposed and examined by analyzing farmers’ perceptions. Research about farmer attitudes and motivations has a more extensive application, such as farmers’ perceptions about biodiversity (Malawska et al. 2014) and the abstract ecosystem services concept (Smith and Sullivan 2014). In an effort to better address the different effects of agricultural policies and practices, the methods for understanding farmers’ perceptions include: face-to-face interviews, interviews by e-mail, telephone and questionnaires by mail, and workshops. There are many farmer-specific variables to choose from, such as age, location, education, and income (Mbaga-Semgalawe and Folmer 2000). Ingram (2008) focused on a farmers’ knowledge about soil and its sustainable management. Farmers’ perceptions are complex, and can constantly change in the context of internal and external factors. Cross et al. (2011) showed how economic dependence is diminishing farmer motivation to participate in conservation programs, but Kvakkestad et al. (2015) found that income was less important than other factors for responders. BaumgartGetz et al. (2012) provided a quantitative summary of 46 studies from 1982–2007 addressing the adoption of agricultural best management practices in the U.S. and found that farmer adoption decisions were always changing. Knowler and Bradshaw (2007) summarized a universal variable from the world’s empirical studies, but Prokopy et al. (2008) concluded that there was no single factor that consistently affected farmer decisions. Statistical analysis is a common method for learning the relationship between farmers’ perceptions and agricultural policies. Farmer typology and the neighborhood effect were also reviewed in some studies (Baerenklau 2005; Briggeman et al. 2007). From the literature reviews, a strong understanding of farmers’ perceptions is very helpful in making and implementing policies, as well as choosing useful farmer-specific variables that may generate further information. In China, the large population of farmer supplies more opportunities. There are lots of studies focusing on farmer

1853

behavior from different perspectives; for example, farmer attitudes toward China’s Sloping Land Conversion Program (Grosjean and Kontoleon 2009; Komarek et al. 2014), and the Grain for Green Program (Cao et al. 2009; Liang et al. 2012). Especially in the primary production areas, agricultural intensification has caused many serious environmental problems. Managers are seeking to find effective policies and practices to balance agriculture and the environment. Farmers as direct operators play a significant role in the process (Zhen et al. 2011). Facing food security and environmental problems, a study of Chinese farmers’ perceptions could provide useful information and set an example for developing countries. Although the above comprehensive studies provide helpful insights about farmers for policy-making and environmental improvement, most of them only explain farmer attitudes, motivations or decision-making about one specific policy or practices, which is just one part of the process of policy-making. They often ignore farmer participation before policy-making or obtain feedback after policy implementation. Farmers’ perceptions are very complex and can be affected by many factors at the same time, therefore continuous study is required. To explore future agricultural environmental protection and guide relevant policy-making, we investigated farmers from 2009 to 2013. Based on the first two investigations, we understood farmer’s behavior and assessed environmental effect. And using a normative landscape scenario method, we designed two scenarios to illustrate future agricultural land use with different water quality improvement. Then we discussed the farmers’ perceptions to alternative scenarios and analyzed feasibility and acceptability of scenarios. With the help of logistic regression model, we found the key factors and proposed agricultural policy to change farmer’s behavior and perceptions to meet the hypothetical scenarios. Farmers were involved in each process, including: target setting, design, and feedback. Both designing rules of alternative scenarios and a detailed analysis of farmer’s behavior and perceptions could provide recommendations for policymakers.

2. Methodology 2.1. Study area Jinjing Town (27°55´–28°40´ N, 112°56´–113°30´ E) (Fig. 1) is a primary grain-producing area located in Dongting Lake Basin in south central China. It covers approximately 13 440 ha, 65% woodland and approximately 27% arable farming land of the landscape. Double-cropped rice is the main crop. Jinjing is comprised of 14 villages and 2 communities. The population was approximately 42 000, and net

1854

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

income of farmer was 6 110 RMB, which was a little higher than the national average in 2010.

Jinjing Town is just a 1-h distance to the provincial capital Changsha, Hunan Province of China, and is a typical subN N

0

600 1 200 1 800 2 400 km

2010

Scenario I

Scenario II

Tea garden

Paddy land

Road

River

Buffer area

Residential area

Forest

Reserviors

Vegetable farm

Wetland

0

1

2

3

4

km

Fig. 1 Land-use map and alternative scenarios of Jinjing Town, Hunan Province, China. Scenario I, maximizing agricultural production; Scenario II, improving water quality.

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

urban agriculture area with characteristics of high input, high yield, and high environmental risk. The main agricultural economic activity for farmers is smallholder farming, livestock breeding and tea gardening. Now they are very different from ten years ago. More and more people like to work in the city, leaving the older farmers still working on the farms. The people living in the deep mountain areas have moved out, and their farmland and houses have been abandoned. At the same time, lots of commercial vegetable companies were brought in by government and the tea garden expanded rapidly. These caused more paddy land use change and serious environmental problems. Water pollution, soil degradation, and biodiversity loss have been an important environmental risk in the Jinjing Watershed. Since 2010, the government of Jinjing Town has administered certain policies and land-use planning to protect environment. For example, the Livestock Forbidden Area Project is helpful for improving river water quality and the environment efficiently. Therefore, Jinjing Town still faces a great challenge of how to balance economic development and environmental protection like other rapid developing country area in China. The policies will affect the agriculture landscape and landuse trend, influencing the environment directly or indirectly, and of course, farmer behavior will change. Farmers and stakeholders also exhibit different attitudes and change their practices to adapt to the policy. Fortunately, they are very interested in expressing their ideas about agriculture, the environment and future prospects in Jinjing Town. This provides a good opportunity to understand their perceptions through this study. Therefore, we would like to discuss farmer’s attitude to the alternative scenarios, environmental protection and agriculture development in future. This study is expected to set a good example of how to solve relationship between agricultural development and environmental improvement from farmer’s perspective for the similar high intensive agricultural area.

2.2. Data collection Developing effective agricultural policies necessitates a better understanding of the behavior and perceptions of farmer. This research is based on data from three original surveys, conducted during 2009, 2011 and 2013. All of the farmers were interviewed face-to-face and were selected randomly. The first survey occurred in 2009. This was the basis for our study. A total of 782 farmers from the town were interviewed. The interviews mainly focused on collecting basic information about households, agricultural production, and agricultural practices. We obtained a general overview of Jinjing from this original survey. The second survey was undertaken in 2011. A total of

1855

124 farmers were randomly selected from five villages with different development situation: Xishan, Huinong, Jinlong, Longquan and Guanjia. The main components included household characteristics, individual farmland practices and attitudes about agricultural services. This survey provided useful detailed information and suggestions from diverse farmers for us to understand farmer behavior and design feasible future agricultural scenarios with specific goals. The last survey was organized in 2013. Fifty-five farmers participated in this investigation. They were also from five villages similar to the second survey, except that Bamaotian replaced Longquan. The farmers’ perceptions about the future of agriculture and the environment, and their preference for landscape alternative scenarios were obtained, which constitutes the farmer’s feedback on two scenarios. We continued to focus on farmers’ perceptions from 2009–2013, which permeated our entire research approach. We learned that a better understanding of farmer’s perception was very helpful for us to build future landscape patterns and environmental protection, and can also provide farmer suggestions and consultations to policymakers.

2.3. Framework of Pressure-State-Response (PSR) PSR (Fig. 2) framework is a common tool for environmental quality assessments. It provides a systematic mechanism to monitor the status of an environment or the sustainable development of natural resources and environmental ecology. The PSR model was divided into three factors in terms of the “Pressure”, “State” and “Response”. “Pressure” on a farmer leads to agricultural land use changes and finally leads to a certain “State”, and this “State” produces positive or negative economic, environmental and social effects, which provoke “responses” by policymakers and/or society with a change to the “Pressure”. PSR provides a strong explanation of the interactions between human activities and the environment. In our study, we employed a PSR framework to analyze farmers’ perceptions to agricultural and environmental development. As Fig. 2 shows, this study is comprised of three sections: present state descriptions, alternative scenario analyses, and response implications. The upper section is the present PSR model analysis. We conducted two surveys about farmers’ behavior, attitudes and motivations toward agriculture, the environment and the future in 2009 and 2011, and we had a strong understanding of “States”: the present land use state and environmental effects. The lower section represents the alternative scenarios analysis. Based on an analysis of the present state, we hypothesized different designing rules and adjustments of farmer behavior to develop two scenarios, including maximizing agricultural production and improving water quality, which

1856

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

Farmer behavior Pressure

Agricultural land use State Farmers’ behavioral adjustments

Alternative scenarios Future state

Agricultural outcomes Effects

Policy Response

New goals

Scenarios assessment Future effects

Farmers’ perceptions Farmers’ responses

Fig. 2 Flow chart of the Pressure-State-Response (PSR) framework. Solid rectangle, the present PSR procedure; dotted circle, the future PSR procedure.

were all based on water quality management. This section discussed what “Pressure” should be taken on farmers, what the “future State” of agricultural land use would be like, and what the “future effects” of these scenarios would be. The “policy” section represents the policy implications. We conducted a survey about farmer responses to two alternative scenarios in 2013, and analyzed the obstructions and feasibility comparing with farmer’s behavior. Reviewing the future agriculture scenarios designing and an analysis of farmers’ perceptions, policy implications were illustrated for policy-making and water quality improvement. An emphasis on farmer participation was an important feature of our study. A PSR framework not only indicated the present PSR process but also identified future PSR scenarios. Understanding farmer behavior, design of future agricultural scenario, and policy implication is very meaningful for nature resources management in future.

3. Farmer behavior and environmental effects Most farmers who participated in the interviews were interested in expressing their attitudes about the present state of agriculture. A mass of information was produced from the interviews, and we would like to analyze farmer behavior from farmer characteristics and individual agricultural practices.

usually has two or three laborers. Most farmers were older than 40 years during the survey. We found that a few young people had stayed home, and it was a little difficult to talk with them about agriculture. They usually declined the interview saying: “I don’t know how to do farming work.” Especially in the hilly area in the northern part of Jinjing, homesteads were even abandoned. Due to data corruption of farmer annual incomes, the family net income and agricultural income were inaccurate; however, we were convinced that off-farm work has become the main source of family income. 78% of households chose off-farm work or part-time farming. They did not like to spend a significant amount of time working on the farm. The attitudes about satisfaction with the family annual income included one-third satisfied, dissatisfied and alright. Education was another vital factor affecting farmer decision-making; 83% of farmers had primary or junior middle school levels of education, but none had bachelor’s degree. A lack of knowledge limited farmer acceptance of new technologies and environmental conservation practices. It is important to mention that we found one quarter of households raised pigs, and several of them directly discharged wastewater into rivers without treatment in 2009. Farming is still a primary occupation for many households in the region. However, more and more people would like to do off-farm work to obtain a stable profit. The age and education of farmers are important factors driving farmers’ perceptions in the following analysis.

3.1. Farmer characteristics 3.2. Individual agricultural practices In the study area, the average number of farmers is 4.4 persons in each household (Table 1). Each household

The principal crop in Jinjing is double-cropped rice, and

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

single-cropped rice is usually distributed on low soil quality farmland (Table 1). Some of the farmers cultivated small dry lands for maize and sweet potato in the hilly areas. For the sake of spending more time on off-farm work, some farmers would like to choose single-cropped rice rather than double-cropped rice. Each household only owned approximately 0.28 ha farmland, and the early rice, late rice and single-cropped rice yields were 4 830, 6 160, and 6 820 kg ha–1, respectively. Agricultural inputs included the following: the cost of machinery, fertilizers, pesticides, herbicides, and seeds. Farmers helped each other during the busy season for free, so the labor input was excluded. 55% of farmers adopted agricultural machinery for plowing, seeding, and harvesting, but only 38% used it for plowing and seeding. Heavy

Table 1 Farmer characteristics and behavior Description1) Statistics value No. of farmers interviewed in the following villages Guanjia 28 Jinlong 21 Longquan 26 Huinong 25 Xishan 24 Age structure (%) Age≤40 years old 17.7 4060 years old 25.0 Education level (%) Primary level 37.9 Junior high school level 45.2 Senior high school level 12.9 None education 4.0 Average number of family 4.4 Farm size (ha) 0.28 Satisfaction with the family annual income (%) Satisfied 24.2 Dissatisfied 36.3 Alright 39.5 Cropping system (%) Double-cropped rice 56.5 Single-cropped rice 34.7 Both 8.9 Grain yield (kg ha–1) Early rice 4 830 Late rice 6 160 6 820 Single-cropped rice Agricultural machinery (%) Part machinery 54.8 Full machinery 37.9 Manpower or animal power 7.3 6 750 Agricultural input (RMB ha-1 yr-1) 285 Fertilizer input (Pure N, kg ha–1) 1)

Village number is ranked by agricultural intensity level from low to high. Guanjia, Jinlong, Longquan, Huinong, and Xishan are villages in Jinjing, Hunan Province, China. Data source: the second survey of 124 farmers in 2011.

1857

chemical fertilizer application is still a traditional method for farmers wishing to improve agricultural production. According to our investigative data, approximately 285 kg ha-1 of pure nitrogen (N) was applied above the recommendation. The decisions about how to apply fertilizers, pesticides and herbicides mainly depended on experience and advices from agricultural service stations. Along with the increased cost of agricultural inputs, 73% of farmers thought the production yield was steady over the last ten years. These may be negative for agricultural development and environment.

3.3. Environmental effects Jinjing is a typical suburban agricultural area that is very near the provincial capital of Changsha. The features of local agriculture include high inputs, high yields and high environmental risks. Agriculture is a major contributor to water pollution, soil degradation, and biodiversity loss similar to other high-intensity agricultural areas in China. Agricultural NPS pollution is a serious environmental problem for the local government and stakeholders. The annual average concentration of total N was 4.70 mg L-1, which largely surpassed the national surface water quality standard (2.00 mg L-1). Every year, approximately 431 tons of total N was discharged from the watershed, including 170 tons of NO3–-N and 143 tons of NH4+-N. Excessive N has been recognized as an important environmental risk factor in the Jinjing watershed. Exacerbating the problem, some agricultural companies are developing rapidly, and the tea industry is also expanding. There is concern that more woodland will transfer to paddies or tea gardens, and water quality is becoming worse. Therefore, how to make a proper planning of agricultural development and environmental improvement in future is an urgent problem to be solved.

4. Design of alternative agricultural scenarios 4.1. Establishment of alternative scenarios The normative landscape scenario is one of scenario methods. It describes future landscape scenarios using specific goals. Normative scenarios may not exist, but such scenarios depict a plausible future (Nassauer and Corry 2004; Nassauer et al. 2007). This method heavily emphasizes the importance of farmer participation. Aimed at sustainable agricultural development and water quality improvement in the study area, we designed two scenarios: Scenario I, maximizing agricultural production; and Scenario II , improving water quality (Li and Liu 2016). Both scenarios focus on the improvement of water quality.

1858

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

The goal of Scenario I was to increase agricultural production over a short time, and the expected effects were as follows: maximizing agricultural production, significantly increasing the net income of farmers, and improving water quality over the 2010 baseline. The landscape in Scenario II was designed to improve water quality, and its expected effects were as follows: to control water quality within the national standard and to ensure the net income of farmers is higher than the baseline. According to our previous studies, the source of water pollution mainly arises from three aspects. Apparently, agricultural activities are a major contributor, accounting for approximately 80%; the other two factors are domestic sewage from daily household life and wastewater from livestock breeding, accounting for 5 and 10%, respectively. Therefore, the rules of design based on these three aspects. After iterative process of goals selection, qualitative description and quantitative design, two future scenarios were developed and achieved the hypothetical goals. The final landscape pattern of the two scenarios was displayed in Fig. 1.

agricultural management practices designed in Scenario I and Scenario II could lead to substantial improvements in water quality and farmer’s income in the future.

4.2. Assessment of alternative scenario effects

Farmers were expected to possess approximately 0.3 ha of farmland per household in the future. This may be different from our assumptions. The reason was that, in the survey results, the main function of agriculture was to provide enough food for the family, not to produce high economic returns. Furthermore, farmers would like to continue to operate smallholder farms, and the appropriate farm size was approximately 0.3 ha, which was the size favored by 86% of farmers. In Scenario I, we hypothesized that more enterprises would vigorously develop in the study area; 93% of farmers would consider renting their farmland to enterprises. In fact, in conversation with farmers, we recognized that if the rent was more than 9 000 RMB ha-1, which is equal to the average family expenditures on food, farmers were willing to rent their farmland immediately. They also mentioned the environment: all of the farmers like appealing agricultural landscapes, rather than high-output, high-pollution agricultural patterns. 70% of farmers will not sacrifice the environment to profitability

Both scenarios focused on water quality improvement and agriculture economic development. Table 2 shows the alternative scenario effects. Obviously, Scenario I indicates some improvement in water quality, as the values are slightly lower than baseline; the water quality in Scenario II is controlled very well and kept at a very low concentration level. Correspondingly, in both scenarios total N export is lower than baseline. Compared with the baseline, biodiversity including plant and animal diversity is of the best quality under the landscape resulting from Scenario II. Thus, both of them fulfill the assumption. On economic aspect, Scenario II has the best ecosystem service value than the other scenario and baseline, Scenario I has the largest direct benefits from agricultural production, and the highest farmer’s income from agriculture. Overall, the results shows that the changes in land use and

5. Assessments of acceptability and feasibility of alternative scenarios As previously mentioned, after a serial of changes in land use and agricultural management practices, scenarios of maximizing agricultural production and improving water quality can be realized. In achieving these different goals, the farmer is the direct participator and a very significant factor. Farmers’ perceptions of acceptability related to the success of failure of future scenarios. Furthermore, we conducted a survey on farmers’ perceptions about the agricultural development, environment protection and future landscape preferences in 2013, and assessed the feasibility and acceptability comparing with the rules proposed in scenarios.

5.1. Farmers’ perceptions of alternative scenarios

Table 2 Assessment of alternative scenario effects Description1) Overall landscape Water quality (total N, mg L-1) Total N export (t yr-1) Biodiversity (CONTAG, %; PRD, pieces100 ha-1) Output value of agriculture (million RMB) Economic service value (million RMB) Farmer’s income from agriculture (RMB) 1)

Baseline 2010 Boring 4.08 431 57.21; 0.0520 406.43 1 825.29 3 595

Scenario I Boring 3.45 343 56.81; 0.0595 672.35 1 723.41 4 971

Scenario II Appealing 1.69 175 67.39; 0.0744 439.35 2 033.58 4 766

CONTAG, contagion index one of landscape pattern metrics, which usually means higher values may result from landscapes with a few large, contiguous patches; PRD, patch richness density, often correlates well with species richness. The prices are based on 2010 price.

1859

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

Table 3 Comparison of farmers’ perceptions and alternative scenarios1) Description Agricultural practices Total farm acreage Farm size Income source Agriculture benefits Whether considering rent farms Practices decision Attitude of new technology Satisfaction of subsidies Consider sacrificing the environment to profitability (high fertilization) Treatment of domestic sewage Decreased amount of livestock breeding 1)

Perception

Scenario I

Scenario II

2 313 ha 0.3 ha Off-farm 83%

Large-scale farms None Tea garden Horticulture Substantial increase income

Smallholder farms None Agritourism

Balance 52% Increased income Profitable 29% Yes 92% High-quality farms will be rent Low-quality farms will be No 8% expropriated Personal experience Full and good agricultural service As Scenario I 91% and technology support Pioneer 21% New technologies: machinery, As Scenario I Wait and see 61% materials, seed Yes 75% Machinery, seed subsidy; Machinery, seed, ecological, Agricultural insurance livestock subsidy. Yes 30% High intensive agriculture Precision agriculture No 70% 60% 90% 50% 90%

Data source: the third survey of 55 farmers in 2013. The table only lists the agricultural activities related to farmers. Additionally, there are many other rules, such as land allocation mapping rules and water quality prediction models. In this study, we only discuss the relationships between farmers’ perceptions and alternative scenarios.

Table 4 Farmer preferences for the alternative landscape scenarios1) Description of Perception landscape Agricultural field I: 98% landscape II: 2%

Primary ditch design

I: 80% II: 20%

Secondary ditch design

I: 98% II: 2%

Riparian zone design

I: 22% II: 78%

1)

Scenario I

Scenario II

Data source: the third survey data in 2013.

through high agricultural output in the future (Table 3). Based on the alternative scenarios, we developed digital images of particular views of future landscapes. The photographs had different features, such as fields, ditches, buffer strips, and they were presented to farmers (Table 4). The results indicated that most farmers preferred large fields resulting from land consolidation. It looks more tidy, beautiful, and easy to cultivate. Instead of environmentally friendly ditches, 80% of farmers chose cement ditches. They explained that rainwater can be quickly drained away from farmland in this way. In spite of introducing ecological function to them, they still maintained their opinions. Rela-

tive to ecological ditches, they preferred other ecological measures: riparian buffers along the river and buffer strips along the roads. 78% of farmers accepted this measure for improving water quality. Farmers’ perceptions of agricultural landscapes suggested that economic benefits alone do not account for their landscape preferences. In fact, farmers in Jinjing have generally been found to be concerned about the environmental effects of agriculture, such as the land use type, water quality and rural environment. However, due to the lack of knowledge, they do not know what to do and how to protect the environment. Based on their perceptions, Scenario I is likely to be accepted by farmers because it increases production with its related assumed increases in economic returns. However, Scenario II could be an appealing, environmentally sustainable, and beautiful rural landscape that is exactly what farmer want.

5.2. Feasibility of alternative scenarios Maintaining stable economic growth along with environmental improvement is the ideal and sustainable approach for the future. Both scenarios illustrated the approaches for managers, and farmers’ perceptions are positive for future agriculture development. Comparing with farmer behavior, we discussed whether the scenarios could be achieved. Agricultural land use pattern When we designed the future scenario and assumed that farmer decisions were mostly driven by economic rationality, we found that economic

1860

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

value was not the most important factor affecting farmer decision-making. Although the output value of agriculture was very low compared with other industries, most of farmers would still like to continue farming. In their opinion, continued farming could supply them with enough food security and reduce expenditures in their daily life. What they consider more seriously is continuing to support their family by farming in case they lose their work in the city. However, they would like to rent to other farmers for free, rather than sell directly. In scenarios, the land cover is very different from present, but as long as paying enough rent for farmers, agriculture land use will be easily changed. Farmer behavior change The investigation results show that 91% of farmers choose the agricultural practices according to their personal experience rather than agricultural service station. Farmers reflected that they obtained information about local agricultural services only through the instructions for fertilizer, pesticide, and seed use methods. There was no technological guidance provided for agricultural facilities, soil quality, or planting techniques. In terms of the willingness to adopt new technologies, only 21% of farmers were pleased to be pioneers of innovation implementation; most of the rational farmers exhibited a wait-and-see attitude. The new agricultural technology extension faced great challenges. Heavy fertilizers application for increasing agricultural production is still rooted in farmers’ perceptions. This situation is very hard to change. Chemical fertilizers, pesticides, and herbicides were still applied in the traditional way in Scenario I, but suggestions for Scenario II included farmers adopting precision agricultural practices. For example, the application of fertilizers needed to be reduced to 185 kg ha-1, which is still a large challenge for managers. Treatment of domestic sewage and waste water We didn’t investigate these aspects in our surveys, but the goals in both scenarios can be achieved by legislation and subsidies. For example, the Livestock Forbidden Area Project encourages farmers who live near reservoirs or upper streams to abandon breeding swine; wastewater and manure had to be recycled based on legislation. And Grain for Green Program provides subsidy to farmers who willing to transfer the dryland to forest. These measures are helpful for improving water quality and the environment. Thus proper policy can meet the requirements in design scenarios. The results show that Scenario I or Scenario II can be acceptable by farmers and is likely to be achieved in future, although there are still some difficulties to be faced. Farmer behavior and perceptions are comprehensive and could be changed or influenced by internal and exterior factors. Farmer’s perception on alternative scenarios implies that

the farmer still concerns about agricultural economic profit. Thus agricultural income is the most important driving force for farmer behavior change. Above all, making a proper agricultural and environmental planning is necessary, but increasing farmer’s income and protecting their benefit are the core things to achieve these aims. That means the government should finance farmers to change their behavior, such as the use of pesticides and nitrates. Policy intervention is a powerful tool to guide farmer change, so we proposed some policy implication for farmers in the following.

6. Discussion and policy implications We know that agricultural policy, technology, and farmer decisions have driven agricultural land use and environment changes. We developed two scenarios of future agricultural landscape patterns with different policy goals. Meanwhile, we also conducted in-depth interviews with farmers to research their motivations, beliefs and perceptions about agriculture and the environment. The results show scenarios are likely to be achieved and farmers play an important role in this process. In order to regulate farmer behavior and perceptions closed to the requirement of Scenario I or Scenario II, we discussed future agricultural policies for farmers.

6.1. Correlation analysis of farmer behavior Farmer is a comprehensive and rational social man. Their behaviors and perceptions continually change in accordance with the changes of environments and polices. For example, land characteristics (plot size, distance, and quality) affect farmers’ soil fertility management behaviors (Tan 2014). Also policy of Improved Nitrogen Management training in China can significantly reduce farmer’s N fertilizer use behavior, but not sufficient to change farmer’s practice significantly (Jia et al. 2013). In our three surveys, we collected a dense mass of information of farmers. It is necessary to find the key indicators affecting farmer behavior. We chose “Whether willing to continue to farm in future” (Yes 0; No 1) as independent variable, which reflected farmer’s attitude to future agricultural development directly. Then we selected agricultural intensity level of village, age, literacy level, number of laborer per household, farmland size per household, agricultural machinery use level, agricultural input per unit, subsidy per year, purpose of farming, attitude to new technology, and effectiveness evaluation of new technology extension as variables. After standardization process of data, we employed logistic regression model. Here we just analyzed correlation rather than simulated the

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

Table 5 The result of logistic regression model1) Variables B SE Wlad Sig. Exp(B) Village –0.338 0.157 4.642 0.031 0.713 Age 0.380 0.208 3.340 0.068 1.462 Purpose of farming –0.470 0.231 4.140 0.042 0.625 Attitude to new technology –0.468 0.292 2.565 0.109 0.626 Constant 1.168 0.930 1.578 0.209 3.215 1)

B, regression coefficient; SE, standard error; Wlad, Chi-square value; Sig., significance; Exp(B), odds ratio.

equation model. Table 5 shows the result. The result shows that agricultural intensity level of village, age, purpose of farming and attitude to new technology were entered the equation. Village variable indicates the farmer who lives in high-quality farmland area is more likely to continue to farm. More and more farmer would like to farm with age increase. If purpose of farming is for sale agricultural production, farmers won’t choose farm, that is to say, agricultural benefit is too low to attract farmers in future. Attitude to new technology variable indicates that farmers don’t like adopt new technology to agriculture. Therefore, obtaining good benefits and increasing acceptability of new technology are the significant factors to adjust farmer behavior in future. Based on the results of surveys data and model analysis, we proposed environmental improvement measures from economic and policy intervention.

6.2. Production prices According to previous analyses, growth from agricultural benefits cannot maintain the increase in agricultural inputs. Farmers are aware that agriculture cannot bring substantial economic benefits. They spend less time and attention on farming. Standard economic theory assumes that producers maximize profits and that their behavior may be changed by economic incentives. This means that if the crop purchase price is high enough, farmers will invest more in agriculture and may greatly reduce household willingness to convert farmland to other uses. The interviews prove that rising commodity production prices and cutting down input costs can encourage farming. The minimum crop purchase price has been implemented for many years. Particularly in high-intensity agricultural areas, farmer’s net incomes are more sensitive with crop purchase prices. To ensure food security in China and to protect farmer benefits, price support mechanisms are an effective way to stimulate agricultural development. Furthermore, the company competition mechanism proposed in Scenario I is also a feasible measure. These enterprises adopt a farmer-enterprise mode or hire farmers

1861

to increase local farm incomes. Farmers showed great interest and were very pleased to rent farmland or join the company.

6.3. Agricultural subsidies Agricultural subsidies can reduce farmer agricultural input costs in the form of direct cash payment. In Jinjing, several agricultural subsidies were issued recent years, but the effects varied. Our analysis of the survey data indicated that the cultivated land subsidy and superior seed subsidy had little power to affect farmer decisions. Subsidies to reduce livestock breeding entitle farmers who give up animal husbandry in “livestock forbidden areas” to obtain subsidy payments. This subsidy has a strong effect, causing many farmers not to breed pigs any more. However, the biogas subsidy does not work well. Only a few households have built biogas digesters to obtain the payments. Furthermore, all agriculture taxes were abolished in China from 2006 to alleviate the financial burden on agricultural producers. In the study area, some farmers would like to lease their farmland to others for free, but the subsidies still belong to them. If these types of subsidies are cancelled, they may fully abandon their farmland. Although agricultural subsidies do not play a significant role in farmers’ perceptions, they have a strong relevance to other agricultural decision-making. Policymakers should provide specific subsidies as required in the future, for example an agricultural machinery subsidy.

6.4. Eco-compensation and ecological engineering In Scenario II, farmers were encouraged to transfer poor quality farmland to forests and nature reserves in northern Jinjing. These measures all required farmers to cooperate. Economic incentives are an effective method to achieve this goal. China’s sloping land conversion program is one of the world’s largest programs offering payments for environmental services. To encourage participation, households are offered annual payments in cash. The results are remarkable and respected. This is a really good example for policymakers. We suggest that if future policies emphasize environmental protection, as in Scenario II, eco-compensation mechanisms are an essential requirement. Several ecological conservation practices, such as riparian buffers, natural reserves and off-channel wetlands, were also developed in Scenario II. These ecological engineering methods are widely used all over the word and have been proven to provide high ecological functions. Scenario II exhibits the same results: substantially improved water quality and biodiversity. Therefore, the government can

1862

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

choose these conservation practices to improve environmental quality. No matter what the eco-compensation or ecological engineering method, these programs need funding from the government. They require the government to guarantee enough investment in agricultural environment programs in the future. They are also a feasible way to establish cooperative relationships among governments, enterprises and farmers. Their responsibility is as a guide and to build and maintain. Another challenge for policymakers is that farmers are unaware of the importance of ecological engineering. The adoption of ecological practices is very limited. Therefore, to help farmers better understand the environment is a new task for policymakers.

6.5. Agricultural services and technology extension According to the survey data, agricultural services just included fertilizers, pesticides, and herbicides, and the main information channel was agricultural materials suppliers. Obviously these services and extensions stimulate farmer consumption. In fact, some books and instructions of how to farm scientifically can be found in agricultural service stations and websites. However, a lack of communication and access to the Internet blocks these information channels. When a new technology comes out, most people observe implementation from other farmers. One reason for non-adoption of a new technology is uncertainty about the outcomes of adoption. Trust helps new technologies succeed because it makes the information more credible to farmers. Access to a variety of agricultural information is crucial. Farmers need information that will allow them to estimate the costs and benefits of available alternatives. A regulation should be made and it should be clear that agricultural service staff and technicians help farmers solve related problems, rather than promote agricultural products. To build farmer trust and encourage farmer adoption of new technologies, the government can promise to supply free technical assistance and agricultural insurance.

7. Conclusion The effect of agricultural activities on water quality largely depends on the behavior of farmers. Environmental and agricultural policies directly influence how farmers to manage their farms. Understanding farmer behavior and perceptions toward agriculture is essential for policymakers to manage nature resources. Therefore, based on our analysis of the survey data and goals, we developed two scenarios to illustrate future agricultural landscapes. One scenario

emphasized maximizing agricultural production, the other one focused on improving water quality. The assessment of the two scenarios indicates that Scenario I can provide high agricultural economic benefits while sometimes improving water quality; and that Scenario II was the attractive landscape and sustainable environment, which generated substantial improvements in water quality and biodiversity. Meanwhile, some protective farming practices, rules for landscape change, and farmer land use behaviors were proposed in both scenarios. These two scenarios illustrated alternative scenarios with rules, criterion, and mapping. It is very helpful for policymakers to imagine what a plausible future landscape will be under different agricultural policy goals and to determine what “Pressures” might be implemented to achieve these expected results. The three farmer surveys were conducted in 2009, 2011, and 2013, respectively. Interpreting and analyzing the data indicated that famers are complex individuals. Farmers can change their behaviors to pursue the largest benefits according to their ideas. Meanwhile, the farmer is a contradiction. They are concerned about the agricultural environment, but they rely on heavy fertilizers to improve production. They are full of hope for future agricultural development, but they are worried about how new technology bring economic losses. Scenario I is likely to be accepted by farmers because it increases agricultural commodity products. However, the appealing landscape, sustainable environment, and beautiful rural landscape in scenario II are exactly what farmers want. The results also show that: Although there are still some difficulties in changing farmer behavior, Scenario I or II can be accepted by farmers and is likely to be achieved in future. Finally, we proposed some policy implication on obtaining well benefits and on increasing acceptability of new technology for encouraging farmers to adjust behaviors and protect environment. A description of alternative scenarios and an analysis of farmers’ perceptions indicated available choices and constructively suggested effective policies. Several measures were proposed for sustainable agricultural development and water quality improvement. Proper polices are helpful to lead farmers toward positive attitudes in implementation. We have made progress in environmental and economic programs, but managers still face challenges in the future.

Acknowledgements This research was supported by the Fundamental Research Funds for the Central Universities, China (2015B13614), and the National Natural Science Foundation of China (41130526). We would like to thank the experts and farmers who participated in the interviews.

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

References Baerenklau K A. 2005. Toward an understanding of technology adoption: Risk, learning, and neighborhood effects. Landscape Ecology, 81, 1–19. Baumgart-Getz A, Prokopy L S, Floress K. 2012. Why farmers adopt best management practice in the United States: A meta-analysis of the adoption literature. Journal of Environmental Management, 96, 17–25. Blackstock K L, Ingram J, Burton R, Brown K M, Slee B. 2010. Understanding and influencing behavior change by farmers to improve water quality. Science of the Total Environment, 408, 5631–5638. Briggenman B C, Gray A W, Morehart M J, Baker T G, Wilson C A. 2007. A new U.S. farm household typology: Implications for agricultural policy. Applied Economic Perspective and Policy, 29, 765–782. Cao S, Xu C, Chen L, Wang X. 2009. Attitudes of farmers in China’s northern Shanxi Province towards the land-use changes required under the Grain for Green Project, and implications for the project’s success. Land Use Policy, 26, 1182–1194. Cross J E, Keske C M, Lacy M G, Hoag D L K, Bastian C T. 2011. Adoption of conservation easements among agricultural landowners in Colorado and Wyoming: The role of economic dependence and sense of place. Landscape Urban Planning, 101, 75–83. Giovanopoulou E, Nastis S A, Papanagiotou E. 2011. Modeling farmer participation in agri-environmental nitrate pollution reducing schemes. Ecological Economics, 70, 2175–2180. Grosjean P, Kontoleon A. 2009. How sustainable are sustainable development programs? The case of the sloping land conversion program in China. World Development, 37, 268–285. Hoai P M, Sebesvari Z, Tu B M, Pham H V, Renaud F G. 2011. Pesticide pollution in agricultural areas of Northern Vietnam: Case study in Hoang Liet and Minh Dai communes. Environmental Pollution, 159, 3344–3350. Ingram J. 2008. Are farmers in England equipped to meet the knowledge challenge of sustainable soil management? An analysis of farmer and advisor views. Journal of Environmental Management, 86, 214–228. Knowler D, Bradshaw B. 2007. Farmers’ adoption of conservation agriculture: A review and synthesis of recent research. Food Policy, 32, 25–48. Komarek A M, Shi X, Heerink N. 2014. Household-level effects of China’s sloping land conversion program under price and policy shifts. Land Use Policy, 40, 36–44. Kvakkestad V, Rorstad P K, Vatn A. 2015. Norwegian farmers’ perspectives on agriculture and agricultural payments: Between productivism and cultural landscapes. Land Use Policy, 42, 83–92. Jia X P, Huang J K, Xiang C, Bergmann H, Zhang F S. 2013. Farmer’s adoption of improved nitrogen management strategies in maize production in China: An experimental

1863

knowledge training. Journal of Integrative Agriculture, 12, 364–373. Lam Q D, Schmalz B, Fohrer N. 2011. The impact of agricultural best management practices on water quality in a North German lowland catchments. Environmental Monitoring and Assessment, 183, 351–379. Lemke A M, Lindenbaum T T, Perry W L, Herbert M E, Tear T H, Herkert J R. 2010. Effects of outreach on the awareness and adoption of conservation practices by farmers in two agricultural watersheds of the Mackinaw River, Illinois. Journal of Soil and Water Conservation, 65, 304–315. Li H Q, Liu L M. 2016. Normative landscape scenario and its application in planning of new countryside construction. Transactions of the Chinese Society of Agricultural Engineering, 32, 305–311. (in Chinese) Liang Y, Li S, Feldman M W, Daily G C. 2012. Does household composition matter? The impact of the Grain for Green Program on rural livelihoods in China. Ecological Economics, 75, 152–160. Liu R, Zhang P, Wang X, Chen Y, Shen Z. 2013. Assessment of effects of best management practices on agricultural non-point source pollution in Xiangxi River watershed. Agricultural Water Management, 117, 9–18. Malawska A, Topping C J, Nielsen H. 2014. Why do we need to integrate farmer decision making and wildlife models for policy evaluation? Land Use Policy, 38, 732–740. Mariola M J. 2012. Farmers, trust, and the market solution to water pollution: The role of social embeddedness in water quality trading. Journal of Rural Studies, 28, 577–589. Mbaga-Semgalawe Z, Folmer H. 2000. Household adoption behavior of improved soil conservation: The case of the North Pare and West Usambara Mountains of Tanzania. Land Use Policy, 17, 321–336. Murphy G, Hynes S, Murphy E, O’Donoghue C. 2014. An investigation into the type of farmer who chose to participate in Rural Environment Protection Scheme (REPS) and the role of institutional change in influencing scheme effectiveness. Land Use Policy, 39, 199–210. Nassauer J I, Corry R C. 2004. Using normative scenarios in landscape ecology. Landscape Ecology, 19, 343–356. Nassauer J I, Santelmann M V, Scavia D. 2007. From the Corn Belt to the Gulf: Societal and Environmental Implications of Alternative Agricultural Futures. Resources for Future, Washington, D.C., USA. Prokopy L S, Floress K, Klotthor-Weinkauf D, Baumgart-Getz A. 2008. Determinants of agricultural best management practice adoption: Evidence from the literature. Journal of Soil and Water Conservation, 63, 300–311. Ryan R, Erickson D, De Young R. 2003. Farmers’ motivations for adopting conservation practices along riparian zones in a mid-western agricultural watershed. Journal of Environmental Planning and Management, 46, 19–37. Sattler C, Nagel U J. 2010. Factors affecting farmers’ acceptance of conservation measures-a case study from north-eastern Germany. Land Use Policy, 27, 70–77.

1864

LI Hong-qing et al. Journal of Integrative Agriculture 2017, 16(8): 1852–1864

Smith H F, Sullivan C A. 2014. Ecosystem service within agricultural landscapes - Farmers’ perceptions. Ecological Economics, 98, 72–80. Stuart S, Schewe R L, McDermott M. 2014. Reducing nitrogen fertilizer application as a climate change mitigation strategy: Understanding farmer decision-making and potential barriers to change in the US. Land Use Policy, 36, 210–218. Tan S H. 2014. Do land characteristics affect farmers’ soil fertility management? Journal of Integrative Agriculture, 13, 2546–2557. Wescoat J L, Johnston D M. 2007. Political Economies

of Landscape Change. Springer Press, Dordrecht, Netherlands. Willy D K, Zhunusova E, Holm-Muller K. 2014. Estimating the joint effect of multiple soil conservation practices: A case study of smallholder farmers in the Lake Naivasha basin, Kenya. Land Use Policy, 39, 177–187. Zhen L, Li F, Huang H, Dilly O, Liu J. 2011. Households’ willingness to reduce pollution threat in the Poyang Lake region, southern China. Journal of Geochemical Exploration, 110, 15–22. (Managing editor SHI Hong-liang)