Exploring the role of environmental literacy and social norms in farmers ’ LMTT adoption: evidence from China

Purpose – Climatic changes caused by greenhouse gases (GHGs) emissions are an urgent challenge for all regions around the globe while the livestock sector is an important source of GHGs emissions. The adoption of low-carbon manure treatment technology (LMTT) by farmers is emerging as an effective remedy to neutralize the carbon emissions of livestock. This paper aims to incorporate environmental literacy and social © Songqing Li, Xuexi Huo, Ruishi Si, Xueqian Zhang, Yumeng Yao and Li Dong. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode This work was supported by the Modern agricultural industrial technology system construction project “ Research on the Economic Development of China ’ s Apple Industry ” (Grant No.CARS-28). The authors would like to extend their appreciation and gratitude to the agricultural departments of Hebei, Henan and Hubei provinces, China, for providing with related data. Retraction notice : The publisher of the International Journal of Climate Change Strategies and Management wishes to retract the article by Songqing Li, Xuexi Huo, Ruishi Si, Xueqian Zhang, Yumeng Yao, Li Dong (2022), “ Exploring the role of environmental literacy and social norms in farmers ’ LMTT adoption: evidence from China ” , published in the International Journal


Introduction 1.1 The relationship between global carbon emissions and livestock manure
The livestock sector is a key part of the morden agricultural industry and plays an essential role in meeting the growing demand for meat-derived food, improving the dietary structure of residents and promoting the adjustment of the modern agricultural structure (Oreggioni et al., 2021).Also, it has become an engine for some developing countries to eliminate poverty traps, increase family income and improve their welfare (Faisal et al., 2021).However, the large amount of livestock manure simultaneously imposes numerous environmental challenges such as greenhouse gas (GHG) emissions, which further contributes to global climate changes and extreme disasters such as persistent droughts and severe floods (Jahangir et al., 2022;Steinfeld et al., 2006).According to the reports of the United Nations, the GHGs emitted annually from the livestock sector represent 29% of the emissions induced by the agricultural sector globally (Rehman et al., 2021).Many scholars have confirmed the causal relationship between livestock manure and GHGs emissions, especially carbon emissions (Awasthi et al., 2022;Wang et al., 2021).Just as Zubair et al. (2020) reported that GHGs such as CO 2 , CH 4 and N 2 O are mainly produced from the livestock industry during livestock manure treatment.In this vein, reducing carbon emissions from the livestock industry by strengthening the environment's supervision and enhancing low-carbon manure treatment have become a consensus of all countries (Sakadevan and Nguyen, 2017).

China's carbon emissions and low-carbon manure treatment
Over the past 40 years, together with urbanization and industrialization in China, the increase in GHGs has exacerbated climate effects, such as an increase in extreme precipitation and abnormal high temperature (Zhang and Maroulis, 2021).In 2020, China's carbon emissions will reach 9.899 billion tons, accounting for 30.7% of global carbon emissions, and China has become the world's largest carbon emitter.Therefore, strengthening China's carbon emission IJCCSM R e t r a c t e d governance is conducive to compressing global carbon emissions and alleviating global warming and climate affects (Dong et al., 2018).As the world's primary meat consumer country, China's livestock industry has become an essential source of GHG emissions (Piwowar, 2019).In 2019, the total carbon emissions of China's livestock sector have exceeded 14 million tons, accounting for 50% of agricultural carbon emissions (Yao et al., 2020).Consequently, since 2012, the government has successively implemented regulatory policies such as legal penalties, financial subsidies and technical guidance to restrict or motivate farmers to adopt a series of low-carbon manure treatment technology (LMTT) to reduce manure carbon emissions (Wang et al., 2022).Unfortunately, the rate of farmers' LMTT adoption is still low.
1.3 Literature review on the low-carbon manure treatment technology and influencing factors of farmers' low-carbon manure treatment technology adoption Existing studies have confirmed that increasing farmers' adoption of LMTT plays a vital role in reducing manure-induced carbon emissions (Wang et al., 2018).LMTT mainly includes energy technology, such as biogas fermentation, and resource technology, such as compost fermentation.The advantage of LMTT is to reduce carbon emissions by turning livestock manure into renewable energy resources.It ultimately drives shifts the livestock economy toward a low carbon emission and climate-adaptive society (Roubík and Mazancov a, 2019).Specifically, the advantages of biogas fermentation are the production of clean biogas fuel, the reduction of foul smell and the decrease of GHG emissions induced by manure (Rajendran et al., 2012).According to Molino et al. (2013), it is found that biogas produced by anaerobic digestion of small biogas digesters can provide a clean, efficient and low-cost renewable energy source.In a recent study, Awasthi et al. (2019) demonstrated that compost fermentation, a biological treatment technology, can stimulate the response of aerobic microorganisms and reduce the unfavorable effects of livestock manure, that is, foul odor, carbon emissions and transmission of bacteria or viruses.
Although the previous literature did not directly analyze the influencing factors of farmers' adoption of LMTT, many scholars have explored the driving factors of farmers choosing manure biogas or compost fermentation.The empirically identified influencing factors mainly include individual characteristics, such as gender, age, education level and political identity (He et al., 2022); cognitive characteristics, such as environmental attitude, risk preference, risk awareness and health perception (Goldfarb et al., 2022); family and business characteristics, such as population size, family labor, farmland area, breeding scale, livestock housing area, livestock market price and cooperative participation (Zhang et al., 2022); social characteristics, such as relationship network, peer effect, social supervision and group pressure (Kreidenweis et al., 2021); policy conditions, such as government supervision, financial support, credit rationing and technical services (Rehman et al., 2021).Farmers' manure biogas and compost fermentation treatment are environment-friendly behaviors and have typical public goods attributes, which probably cause farmers' adverse selection and moral hazard (Spielmeyer, 2018).Although previous studies have focused on the impact of government regulations on the farmers' composting or fermentation treatment, there is little reasons recognized and accepted widely, which still hinder the increase in the rate of farmers' LMTT adoption.
1.4 Environmental literacy and social norms offer new ideas to solve the issue The farmer's environmentally friendly behavior is the combined action of internal and external factors (Unay-Gailhard and Bojnec, 2021).Previous studies have confirmed that environmental literacy has expanded the farmer's 'behavioral attitude' within the framework of the theory of planned behavior and played an essential role in driving farmers' green production (Guo et al., 2020).Meanwhile, accompanied by the diversified development of the social governance system, Environmental literacy and social norms R e t r a c t e d social norms have also become an essential supplement to government regulation and played a key role in guiding farmers' green production with low-cost and high-efficiency (Quan et al., 2022).Consequently, the academic community has conducted many survey analyses on the role of environmental literacy and social norms in promoting the pro-environmental behavior of farmers (relevant literature in Table 1).This research used 1,023 Chinese households' survey data, explored farmer environmental-friendly behaviors, including organic fertilizer application, and stated that environmental literacy, such as environmental responsibility, knowledge and skill, positively and significantly influences farmer environmental-friendly behaviors Guo et al. (2020) This research used data from rural inhabitants in the state of North Kordofan, Sudan and Chiang Mai Province and the local people believe that they are aware of their environment by using the concept of environmental literacy that may help to develop sustainable environmental management practices Hares and Eskonheimo (2006) This research used survey data collected from Zambia and reported that environmental literacy positively and significantly influences farmers' decisionmaking ability to lessen environmental degradation; The more farmers understand environmental degeneration and its consequences, the more likely they are to take certain measures to reduce the intensity of grazing Wu and Mweemba (2010) The study is based on a combination of survey and census data from 105 farmers in Switzerland and found that there is a significant impact of noncognitive skills on farmers' adoption of climate change mitigation measures Kreft et al. (2021) This research analyzed data from a randomized controlled trial (RCT) conducted with 1287 small-scale tea farmers in Vietnam and believed that significant impacts of the 50% price subsidy and experience sharing on farmers' adoption of organic fertilizer  (2020) argued that environmental cognition has a significant impact on individual ecological consumption.In a recent study by Lu et al. (2020), it is found that environmental cognition can encourage farmers to adopt the crop straw recycling behavior.Consequently, it is hypothesized that if farmers' environmental cognition level is higher, they will hold a positive attitude towards environmental issues and tend to adopt the LMTT.
Similarly, the other dimension, such as environmental skill, is also considered a key determinant in modifying the environmental behaviors of farmers.With regard to the relationship between farmer skill and behavior, Wang et al. (2017) found that skill training and guidance from the government and even relatives significantly influence farmers' behavior of using pesticides.Similarly, Pinzone et al. (2019) believe that providing green-related skills for employees helps improve an organization's environmental performance.In this study, environmental skill refers to the ability of farmers to mitigate and lessen manure-induced pollution.In livestock and poultry breeding, the higher the level of environmental skill, the more skilled farmers will adopt environmental protection approaches, that is, the stronger the farmers' ability to solve manureinduced environmental pollution, the greater the possibility of adopting LMTT.
Lastly, environmental responsibility means the concentrated expression of individuals' attitudes, views, beliefs and values on environmental issues (Yang et al., 2021).Individuals with strong environmental responsibility will have altruistic behavior tendencies toward negative environmental externalities (Hines and Hungerford, 1986;Pawaskar et al., 2018).Ding et al. (2017) believed that the higher level of residents' environmental responsibility led individuals to implement energy-saving technologies.In agricultural farming, farmers are inclined to invest time and money to adopt the LMTT to reduce the adverse effects of production activities on the environment.It infers that farmers with a higher sense of environmental responsibility take economic benefits when making production decisions, pay close attention to the impact of self-behaviors on the public environment and tend to adopt the LMTT.On the basis of the above analysis, the following hypotheses are proposed.
H1. Environmental literacy exerts a positive and significant influence on farmers' adoption of LMTT.
H1a. Environmental cognition can promote the adoption of LMTT by farmers.

R e t r a c t e d
H1b. Environmental skill can drive the adoption of LMTT by farmers.
H1c. Environmental responsibility can encourage the adoption of LMTT by farmers.

2.2
The impact of social norms on farmers' low-carbon manure treatment technology adoption Social norms are critical variables of reasoned action theory and planned behavior theory (Ajzen, 1991).These theories hold that social norms play a vital role in shaping behavioral intentions and affecting individual behaviors.Czajkowski et al. (2017) argued that morality and intrinsic motivation induced by social norms significantly influence the waste disposal behavior of households.Kim and Seock (2019) pointed out that altruism values affect the purchase of environmentally friendly products by individuals.In a recent study by Zeng et al. (2020), it is found that external incentives and social norms are significant determinants for rice farmers to implement reduced use of fertilizers and pesticides.
According to Cialdini et al. (1990), social norms are divided into descriptive and imperative norms.Descriptive social norms refer to what people actually do, that is, specific behaviors that have been implemented or are being implemented by most people in a group.Imperative social norms refer to what people think they should or should not do, such as the particular behavior that people agree or oppose in the social group.Descriptive social norms often affect people's behavior unconsciously; an individual's behavior is usually affected by the behavior of most people around them, but he or she doesn't notice it (Cialdini et al., 2007).Imperative social norms strengthen the right and wrong judgment of specific behavior by most people in the group and further impact individual behavior choices.When descriptive and imperative norms favor individual behavior, then the probability of the individuals' behavior becomes the strongest.
Based on descriptive social norms, the present study infers that if some farmers without LMTT adoption live together with other farmers with LMTT adoption, they will follow the trend of adopting LMTT.The impact of imperative social norms on farmers is manifested by social pressure; farmer behavior is affected by public opinion around them.For example, when relatives, friends or neighbors believe that LMTT should be adopted, they will adopt it under public opinion pressure.On the basis of the above analysis, the following hypotheses are proposed.
H2. Social norms have a positive and significant impact on farmers' LMTT adoption.
H2a. Descriptive social norms have a positive impact on farmers' LMTT adoption.
H2b. Imperative social norms have a positive impact on farmers' LMTT adoption.

The moderating effect of social norms in environmental literacy affecting farmers' lowcarbon manure treatment technology adoption
In China, farmers mainly exchange information related to agricultural production through social networks (Beaman and Dillon, 2018).Social norms generally act as a function of mutual trust and information exchange between individuals (Li et al., 2021).When farmers have stronger social norms that are environmentally friendly, the relationship network is filled with more pro-environmental signs.Therefore, pro-environmental signs may moderate the impact of environmental literacy on the environmental behavior of farmers.
This article hypothesizes that farmers can also make behavioral adjustments based on the intensity of social norms when environmental literacy drives farmers to IJCCSM R e t r a c t e d adopt the LMTT.Firstly, as for social norms of pro-environmental behaviors, environmental protection signs can be easily diffuse among farmers, so their judgment regarding environmental behaviors may alter according to other group members, who are also conducive to enhance farmers' LMTT awareness and prompt them to adopt LMTT actively.Second, social norms that transmit pro-environmental signs through the relationship network also disseminate environmental skill and interaction between environmental behaviors.When most farmers' environmental skills and environmental behaviors are harmonized, their environmental skills will be further improved and encouraged toward adopting the LMTT.Finally, with the rise of manure-induced pollution and the strengthening of environmental regulations, farmers have realized the environmental costs caused by opting for conventional production methods.When some farmers have a high sense of environmental responsibility, they also lead other farmers to shoulder environmental responsibility and adhere to the LMTT.On the basis of the above discussion, the following hypotheses are proposed in the current study.Besides, Figure 1 shows the theoretical analysis framework of this paper.
H3. Social norms have a positive moderating effect on the relationship between environmental literacy and farmers' LMTT adoption; H3a. Social norms have a positive moderating effect on the relationship between environmental cognition and farmers' LMTT adoption; H3b. Social norms have a positive moderating effect on the relationship between environmental skill and farmers' LMTT adoption; H3c. Social norms have a positive moderating effect on the relationship between environmental responsibility and farmers' LMTT adoption.
The theoretical framework operationalize in the current study is shown in Figure 1.

Sample selection
Data are obtained through a field survey from three provinces of China, i.e.Hebei, Henan and Hubei, from July to August 2018 and March 2019 (as shown in Figure 2).The sample This paper used a combined sampling approach such as stratified and random sampling.The selection is made in the following steps astwo to four counties from each province are selected, thenthree to five towns from each county are chosen.And finally, farmers are randomly selected from -five to seven villages for each town.Around 1,000 questionnaires are distributed in the study areas, and finally 941 valid samples were obtained after eliminating 59 invalid questionnaires, accounting for 94.10% of the total sample.According to the sample size, 320 farm households from Hebei, 314 from Henan, and 307 from Hubei were found to reflect the distribution of samples in the study regions evenly.

Outcome variable
The outcome variable in the current study is farmers' LMTT adoption, a discrete binary variable.If farmers adopt the LMTT, the value is 1; otherwise, the value is 0. In practice, due to the heterogeneity of government technology promotion and farmers' resource

Explanatory variables
The core explanatory variables are environmental literacy (environmental cognition, skill and responsibility) and social norms (descriptive and imperative social norms)the fivelevel Likert scale method is used for the measurement of the variables.Meanwhile, drawing on the views of Sorkun (2018), we used the arithmetic average of variables for simple weighting to measure the level of environmental literacy and intensity of social norms.The descriptive statistics of these indicators are depicted in Table 2.

R e t r a c t e d
In addition, this article adopts an independent sample T test to analyze the difference in the explanatory variables core between adopters and non-adopters to preliminarily judge the relationship between the explanatory variables core (environmental literacy and social norms) and the explained variable (farmers' LMTT adoption).From Table 3, the results show that the T-test rejects the null hypothesis and concludes that there are significant differences in environmental literacy and social norms between the adopters and nonadopters.Compared to nonadopters, the overall level of environmental literacy of adopters and the intensity of social norms are higher.Specifically, for energy technology, the mean differences of adopters' environmental skill, environmental responsibility and descriptive social norms are 0.773, 0.506 and 0.426, respectively.Similarly, for resource technology, the mean differences in the environmental skill of adopters' environmental skill, environmental responsibility and imperative social norms are 0.411, 0.330 and 0.489, respectively.

Control variables
Given that other variables can also influence the adoption of LMTT by farmers and using relevant research by  4.
Table 4 shows that male heads represent 82.16% of the sample and males are still the main decision-makers of the family.About 50.23% of household heads belong to the age group between 41 to 60 years old, and the overall average age of the sample is 46.75 years old.Meanwhile, 67.43% of rural households have an educational level of less than nine years, and most of them belong to primary and middle-level education.The average education level of rural households is 7.80 years.About 77.32% of households have a net income of less than US$10,000, and the overall average income is around US$8,545.Approximately 83.64% of the breeding scale is less than 500, and the breeding scales are mainly composed of free-range (<50 heads) and professional breeding (50-500 heads).Furthermore, 87.41% of households have less than five workers and 94.35% of households have farmland areas of less than 3 hm 2 .
3.5 Model specification 3.5.1 Logit and IV-Logit.Given that the adoption of LMTT (energy and resource technologies) by farmers is a discrete binary variable.Therefore, drawing on previous research such as Khan and Habib (2020), the current study used the Logit model to analyze environmental literacy and Compared to the Probit model, also widely used for microeconomic econometric analysis, the Logit model does not require the survey data to completely obey the normal distribution and has the consistency of the estimated results with the maximum utility theory (Babiker et al., 2021).The Logit model formula is as follows: (1) In formula (1), where p represents the probability (0-1) of farmers' LMTT adoption.b 0 is regression intercept (constant term).x i is the influencing factors of the adoption of LMTT by Previous research has confirmed a reverse causal relationship between the level of agricultural skills and the adoption of farmers' technologies (Xue et al., 2021).As the intensity of the adoption of LMTT by farmers increases, the environmental skill of farmers also tends to increase synchronously.Therefore, there may also be a reverse causal relationship between environmental skill and the adoption of LMTT by farmers, which further results in an endogenous problem and ultimately leads to bias in the logit model estimation results.Furthermore, omitted control variables can affect both the core explanatory variables and explained variables, thus exacerbating the endogeneity issues.Therefore, the current study chooses variables such as "the nearest distance between barn and livestock department" as an instrumental variable and applies the IV-logit model to amend the endogeneity issue.The main reasons for this instrumental variable chosen are as follows: On the one hand, the adoption of LMTT by farmers is not subject to the imperative restrictions of the livestock department, and the adoption of LMTT depends on the wishes of farmers.Consequently, "the nearest distance between the barn and the livestock department" is not directly related to the adoption of LMTT by farmers, so the variable is considered as exogenous variables.On the other hand, the closer the pig barn is to the livestock department, the more convenient it is to accept the livestock department's environmental technical guidance.In this vein, farmers have higher environmental skill, which meets the correlation between instrumental variables and endogenous variables.Thus, "the nearest distance between the barn and livestock department" is regarded an appropriate instrumental variable in the current study.
3.5.2Moderating effect model.If the influence of the explanatory variable X on the explained variable Y changes with the change of the third variable M, then the variable M will play a moderating role in the relationship between X and Y (Weise et al., 2020).When both X and Y are continuous variables, the interaction term of X and Y can be introduced into the model to test the moderating effect.If the interaction term is significant, the variable M will have a significant moderating impact (S anchez-Infante Hern andez et al., 2020).This paper introduces the interactive items of environmental cognition, environmental skill, environmental responsibility and social norm, respectively, to test the possible moderating effects of social norms on the impact of environmental literacy on the adoption of LMTT by farmers.The specific model is as follows: IJCCSM

R e t r a c t e d
The literacy Â norm represents interaction terms, and b 3 is the estimated value vector of the interaction term.The connotation of other variables is the same as formula (1).

Model fitting results
As shown in Models 1 and 4 in Table 5, we used the logit model to analyze the influence of environmental literacy and social norms on the adoption of energy and resource technologies by farmers.Furthermore, we explored the impact of different dimensions of environmental literacy and social norms on explained variables, as shown in Models 2 and 5.The results in Table 5 show that the LRx 2 values of equations ( 1), ( 2), ( 4) and ( 5) at 1% significance level are 45.27, 44.05, 45.29 and 44.01, respectively, indicating that models fit the data.
To fix the endogeneity issue, we applied the IV-Logit model, as shown in Table 5.The findings show that the Waldx 2 values of equations ( 3) and ( 6) at a 5% significance level are 40.28 and 41.06, respectively.The Durbin-Wu-Hausman (DWH) test values are 6.27 and 6.04, respectively, which are highly significant at a 1% level of significance, indicating that the model has an endogenous issue.In this case, using a conventional regression model may lead to biased results, so employing IV-Logit estimation is appropriate in the current study.
The study also examined the effectiveness of the instrumental variable "the nearest distance between the pig barn and the livestock department."Analogous to the study of Xu et al. (2018), the results show that the F values in the first stage of equations ( 3) and ( 6) are 12.05 and 13.90 (the critical value F = 10), respectively, concluding that the endogenous variable and instrumental variable are highly correlated, so there is no weak instrumental variable.In addition, we conducted a multicollinearity test, and the maximum and minimum VIF values were 2.75 and 0.67 (the critical value F = 10), respectively.So there were no multicollinearity issues among variables.

The impact of environmental literacy and social norms on farmers' low-carbon manure treatment technology adoption
The estimation results of Models 1 and 4 show that environmental literacy significantly influences the adoption of energy and resource technologies by farmers.H1 is confirmed.The marginal effect is 0.090 and 0.066, which means that if the level of environmental literacy increases by 1 unit, the probability of farmers' energy and resource technologies will increase by 9.0% and 6.6%, respectively.The results are consistent with the empirical findings of Adama et al. (2018) and Xue et al. (2021), who also showed that environmental literacy is beneficial in promoting the adoption of green technologies by farmers.
The improvement of farmers' environmental literacy indicates that farmers have a clear understanding of environmental issues, hold a stronger will to adopt green production technologies, and show a more responsible attitude towards environmental protection strategies (Li et al., 2020).In particular, environmental literacy can inspire farmers to adopt LMTT by strengthening environmental cognition.Farmers can recognize the importance of sustainable production by switching to low carbon production methods (Faisal et al., 2021;He et al., 2016).Furthermore, environmental literacy encourages farmers to actively adopt LMTT technology by broadening technology acquisition channels, improving technical operating standards and raising subsidy standards (Si, Wang, et al., 2020).Further, environmental responsibility is an essential driving factor for farmers to transform from economic rationality to ecological rationality (Bakker et al., 2021;Graddy-Lovelace, 2020).In practice, despite the long payback period of the LMTT investment, environmental responsibility can encourage farmers to increase the investment in the adoption of LMTT Notes: *, **, ***represent the significance level of 10, 5 and 1%, respectively.The marginal effect is reported in the table and the robust standard error is shown in parentheses IJCCSM R e t r a c t e d (Varma et al., 2021a).Finally, our research further promotes the development of traditional agricultural economic management theories such as government support and organization participation, and believes that environmental literacy can improve the endogenous motivation for farmers' technology adoption and has become a foundation for playing a role in the external environment, such as government and organizational support (N'souvi et al., 2021;Qi et al., 2021).
In social norms, the findings show a positive and significant influence on farmers' LMTT adoption.H2 is confirmed.The marginal effect is 0.061 and 0.073, which reveals that if the intensity of social norms increases by 1 unit, the probability of farmers' energy and resource technology adoption will increase by 6.1% and 7.3%, respectively.Our findings are consistent with the empirical results of He et al. (2017) and Zhong and Huang (2017).They showed that farmers in developing countries generally have a strong face concept, an inducing factor in the formation of social norms, which has been a critical determinant in the adoption of green technologies by farmers.Moreover, some scholars argued that the face concept is a manifestation of mutual benefit between individuals, the so-called "human relationship," and it also has the meaning of social evaluation, prestige and status (Dong et al., 2018).Furthermore, other scholars explored the influence of social norms on the adoption of green technologies by farmers from the perspective of peer effect and social pressure (Adnan et al., 2019;Bansal et al., 2021).Farmers usually follow the behavior of the same group to avoid collective condemnation and alleviate social pressure due to deviation of behavior, maintain social relations, maintain a sense of communal belonging and improve understanding of honor (Cai et al., 2019;Dan and Kong, 2015).Consequently, our research pioneered a new exploration of the promotion mechanism of social norms influencing farmers' technology adoption.
The estimation results of Models 2 and 4 show that environmental cognition has not shown significant results for the adoption of LMTT adoption; hence, H1a is falsified.Environmental skill and responsibility positively and significantly influence the adoption of agricultural energy and resource technologies, thus confirming H1b and H1c.Specifically, the marginal effect of environmental skill and environmental responsibility on farmers adopting energy technology is 0.095 and 0.032, which means that if the level of environmental skill and environmental responsibility increases by 1 unit, the probability of farmers adopting energy technology will increase by 9.5% and 3.2%.Similarly, the marginal effect of environmental skill and environmental responsibility on farmers adopting resource technology is 0.063 and 0.053, revealing that the probability of adopting resource technology will increase by 6.3% and 5.3%, respectively.These findings are supported by Li et al. (2020) and Musafiri et al. (2022).They also believed that low-carbon production skills and environmental responsibility for carbon emission reduction have beneficial effects on farmers' adoption of low-carbon agricultural technologies.Biogas fermentation requires excellent conditions to control temperature and moisture, and compost fermentation requires precise settings of fermentation tanks and storage time (Zahedi et al., 2022).Technology acquisition has become a bottleneck factor in promoting the adoption of lowcarbon agricultural technology by farmers (Damalas and Koutroubas, 2018).As for environmental responsibility, livestock manure carbon emission has a negative external effect, and economic rationality is prone to produce the "tragedy of the commons".A strong sense of environmental responsibility tends to drive farmers to make decisions about the adoption of LMTT from the perspective of ecological and social rationality (Tadaki et al., 2015;Zhang, Li, et al., 2020).
Moving toward the imperative social norms, the results show that it exerts a positive and significant influence on farmers' LMTT adoption.However, descriptive social norms have Environmental literacy and social norms R e t r a c t e d no significant impact on farmers' LMTT adoption.Therefore, H2b is confirmed and H2a falsified.Specifically, the marginal effect of imperative social norms on the adoption of LMTT by farmers is 0.028 and 0.086, which means that if the intensity of imperative social norms increases by 1 unit, the adoption of energy and resource technology by farmers will increase by 2.8% and 8.6%, respectively.When most farmers have established group norms such as "they should adopt LMTT," and if some farmers do not adopt the LMTT, they will be under tremendous social pressure.Therefore, farmers will consider the risk of penalties from public opinion when making decisions about the adoption of green technologies (Dong and Lian, 2021;Robinson et al., 2020).When environmental property rights are difficult to define, imperative social norms reduce the possibility of farmers' "free-rider effect," and the externality of social public opinion supervision to a certain extent can be internalized into individual behavior costs.However, descriptive social norms are not significant, indicating that when the rate of adoption of LMTT by farmers is low, they are less affected by the behavioral decisions of other farmers, finally eliminating the economic "peer effect" (Nguyen, et al., 2020).Compared to Models 2 and 5, the estimated results of Models 3 and 6 show that the marginal effect of environmental skill is reduced, indicating that if the simultaneity biases are not dealt with, the impact of environmental skill on the adoption of LMTT by farmers can be overestimated.Specifically, the effects of environmental skill on the adoption of LMTT (energy and resource technologies) by farmers are stronger than environmental responsibility.The marginal effect of environmental skill on the adoption of energy technology by farmers is 0.033 and 0.030, which means that if the level of environmental skill and responsibility increases by 1 unit, the probability of adoption of energy technology by farmers will increase by 3.3% and 3.0%, respectively.The marginal effect of environmental skill on the adoption of farm resource technology is 0.032 and 0.051, revealing that the probability of the adoption of farm resource technology will increase by 3.2% and 5.1%, respectively.Additionally, imperative social norms have a significant influence on farmers' LMTT adoption.If the intensity of imperative social norms increases by 1 unit, the rate of adoption of farmers' energy and resource technologies will increase by 2.9% and 8.3%, respectively.
For control variables, the results are significant for some variables.For example, the effect of education level on the adoption of LMTT by farmers is 0.082 and 0.058, which shows that if education level increases one year, the probability of the adoption of energy and resource technologies by farmers will increase by 8.2% and 5.8%.Good education is an essential incentive for farmers to opt for green production behavior (Adama et al., 2018).The effect of the breeding scale on the adoption of LMTT by farmers is 0.093 and 0.101, which means that if the breeding scale increases by one head, the probability of farmers adopting energy and resource technologies will increase by 9.3% and 10.1%, respectively, which is consistent with the empirical research of Hou and Hou (2019), who showed that the larger the breeding scale, the more farmers are willing to adopt clean production technologies.Farmers with higher family net income and government subsidies are more inclined to adopt LMTT.Consistent with the research results of Yuan and Zhang (2020), the cost reduction and incentives effects of economic capital positively affect the farmer's technology investment and adoption.Furthermore, the effect of organizational participation on the adoption of LMTT by farmers is 0.071 and 0.069, which reveals that if organizational participation increases by 1 unit, the probability of adoption of energy and resource technologies by farmers will increase by 7.1% and 6.9%, respectively.These findings confirm the view of Huang et al. (2020), who argued that organizational participation can IJCCSM R e t r a c t e d provide technical assistance, product sales and standard guidance to promote the adoption of green production technology by farmers.

4.3
The moderating effect of social norms in environmental literacy affecting farmers' lowcarbon manure treatment technology adoption This article introduces the interaction terms of environmental literacy and social norms in Models 1 and 4 to further analyze the moderating effect of social norms in environmental literacy that influences the adoption of LMTT by farmers and finally obtain the estimated results of Models 7 and 9.For Models 3 and 5, the interactive terms of environmental cognition and social norms, environmental skill and social norms and environmental responsibility and social norms are introduced, and Models 8 and 10 are obtained, respectively, as shown in Table 6.
From Models 7 and 9, the interaction terms of environmental literacy and social norms are significant at the significance level of 10% and 5%, respectively, indicating that social norms have a crucial determinant of environmental literacy that affects the adoption of energy and resource technologies by farmers.Furthermore, the moderating effect of social norms on environmental literacy that influences farmers' resource technology is more substantial, and therefore H3 is confirmed.The results are consistent with Daxini et al. (2019), who believed that social norms as an informal system could improve the impact of environmental skill and environmental responsibility on the adoption of green technologies by farmers.Just as Wang et al. (2017) also confirmed the moderating role of social norms in environmental literacy affecting farmers' fertilizers and pesticide reduction behavior.The higher the intensity of social norms, the higher the environmental literacy of farmers and ultimately leads to a better farmer enthusiasm for adopting green technologies.
In Models 8 and 10, the interaction terms of environmental skill and social norms and the interaction terms of environmental responsibility and social norms are significant, indicating that social norms also play an important moderating role.Meanwhile, compared to energy technology, social norms exert a more robust regulatory effect on environmental literacy, affecting the adoption of resource technologies by farmers.Therefore, H3a is falsified, H3b and H3c are confirmed.It reflects that social norms can effectively convey LMTT information to farmers through social networks, improving farmers' sense of  7 report that compared with the Logit and IV-Logit models estimation results in Table 5, the effects of the core explanatory variables did not change significantly.Hence, the benchmark model regression results show good robustness.

Conclusions and policy implications
Climate change has exerted a huge influence on humans and ecosystems, and the climate affects that it causes are increasing sharply.The livestock sector is an essential contributor to global carbon emissions.As a significant livestock breeding country globally, increasing the intensity of farmers' LMTT adoption in China is conducive to cope with the rising world's carbon emission and global climate damage.Unfortunately, the acceptance and adoption rate of farmers' LMTT is still relatively low.Furthermore, existing research has not theoretically analyzed the adoption of LMTT by farmers from the perspectives of behavioral economics and management.Consequently, we introduced environmental literacy and social norms into the analysis framework of farmers' LMTT adoption to identify critical driving factors in improving the probability of farmers' LMTT adoption and ultimately achieving carbon peak and carbon neutrality.Besides, the main theoretical contributions of this study are two aspects: On the one hand, as the adoption of farmers' LMTT has typical public goods attributes, this research breaks through the original theory of planning behavior and introduces environmental literacy and social norms into the analysis framework of the adoption of farmers' LMTT, thus enriching the theoretical connotations of traditional agricultural economics.On the other hand, this paper abandons the judgment of the rational assumption of the individual and analyzes the bounded rationality, such as ecological and social rationalities concerning the effects of environmental literacy and social norms, which will enrich the research limits of agricultural technology theory.
The overall findings show that environmental literacy and social norms positively and significantly influence farmers' LMTT adoption.Specifically, the effects of environmental literacy on the adoption of LMTT by farmers are mainly contributed by environmental skills and environmental responsibility.And the improvement of social norms in the adoption of LMTT by farmers is primarily due to the leading role of imperative social norms.Additionally, after addressing the simultaneity biases caused by the reverse effect between environmental skill and the adoption of LMTT by farmers, the results reveal that the effect of environmental skill is weakened.Furthermore, social norms are found to have moderating effects on environmental literacy that influence the adoption of farmers' LMTT.Compared with energy technology, social norms have a more robust regulatory effect in environmental literacy that affects the adoption of agricultural resource technology.
In essence, based on the empirical findings, the current study proposes some policy implications.First, the government should increase farmers' cognition and their confidence in LMTT adoption.In this regard, the government should use electronic and print media to publicize the application, operation and implementation of the LMTT.Second, the government should establish social organizations to provide technical assistance and support farmers in improving their knowledge about the LMTT.Third, the government should strengthen environmental education for farmers, especially in response to environmental issues caused by livestock manure, so that farmers can forecast the consequences of environmental hazards and enhance their sense of environmental responsibility.Fourth, the government should increase the subsidy standard according to the adopted area of the LMTT and give full play to the leveraging effect of fiscal policy.Finally, the government should encourage the construction of village regulations and folk agreements, including environmental protection initiatives and environmental damage penalties, which can compensate for the restraining and leading role of formal rules such as laws and regulations.
These research conclusions can provide an empirical and useful experience for other countries, especially developing countries.Our research also has some shortcomings: first, the LMTT in the current study includes energy and resource technologies, while the cost and benefit of these technologies are different.Therefore, it is necessary to further study how the cost-benefit relationship affects the adoption of LMTT by farmers.Second, there are apparent differences in the economic and social structures embedded in farmers of different regions and scales, which determines that the effects and transmission mechanism of environmental literacy and social norms on the adoption of LMTT by farmers can be heterogeneous.Thus, future research should consider the heterogeneity of breeding scale, and differentiated incentive or restrictive policies for the government should be proposed.Third, due to the lack of related data, the current study did not consider the impact of natural factors such as topography, temperature and humidity on the adoption of LMTT by Figure 1.The theoretical framework of this research

Figure 2 .
Figure 2. Sample study area Figure 3.Statistics of farmers'LMTT adoption in different provinces variables may cause endogenous issues.Of course, these shortcomings also provide exciting avenues for future research.
1.5 The innovation of this researchIn summary, it can be found that, first, previous studies only measured environmental literacy and social norms from a certain level, and a complete indicator system has not yet been constructed.Second, the previous literature has not incorporated environmental literacy and social norms into a unified analysis framework of farmers' environmentally friendly behaviors, especially farmers' LMTT adoption.Finally, previous research has rarely explored the reverse causality and endogeneity between environmental literacy or social norms and farmers' behaviors.Consequently, the main innovations of this study are as follows: we innovatively measure environmental literacy from environmental cognition, skill and responsibility and describe social norms from descriptive and imperative norms.In addition, environmental literacy and social norms are incorporated in the unified analysis framework of farmers' LMTT adoption.Meanwhile, the IV-Logit model is used to deal with

Table 2 .
Si, Lu et al. (2020), we selected gender, age, education level, net family income, breeding scale, family labor, farmland area, government supervision, government subsidies, technical training and organization participation as control variables.The descriptive statistics of control variables are shown in Table

Table 4 .
farmers, including literacy, norm and X. b i is the regression coefficient of the i-th influencing factor.m is random interference.In formula (2) the decision represents the decision of farmers to adopt energy or resource technologies.If energy or resource technologies are adopted, then the assigned value is 1.Otherwise, it shows that farmers have not adopted energy or resource technologies.Literacy means environmental literacy (environmental cognition, skill and responsibility).The norm signifies social norms (descriptive and imperative social norms).The X are the control variables.b 1 , b 2 and u are the estimated vectors of the environmental literacy, social norms and control variables estimated by the regression coefficients, respectively.« represents the independent and identically distributed random error term and w (•) is the probability function of the logistic distribution.

Table 6 .
(Sarma, 2022;Wens et al., 2021) Mojzisch, 2019) in adopting the LMTT(Ardoin et al., 2020;Germar and Mojzisch, 2019).Our research has further enriched traditional farmer behavior theory, just asGovindharaj et al. (2021)hold that farmers' behaviors are the result of the combined effect of internal and external factors, while our research further explores how exogenous variables, such as social norms, exert behavioral restraints or guidance effects on environmental literacy, e.g.endogenous variables.4.4 Robustness testPrevious studies on robustness test methods consisted mainly of model replacement, variable replacement and sample re-extraction(Sarma, 2022;Wens et al., 2021).Thus, we also employ the Probit and IV-probit models to test the robustness of the model estimation results by substituting the core explanatory variable 'environmental literacy'.Specifically, compared with the Logit model, the Probit model assumes that the original data follow a normal distribution rather than a logistic distribution, and the model application conditions are stricter.Additionally, we use "do you think the environmental pollution of manure is serious?" to represent environmental cognition, "how many times do you receive LMTT training per year?" to characterize environmental skill, and "are you willing to improve the ecological environment in the village?" to describe environmental responsibility.The model estimation results in Table Notes: *, **, ***represent the significance level of 10, 5 and 1%, respectively.The marginal effect is reported in the table, and the robust standard error is shown in parentheses Notes: *, **, ***represent the significance level of 10, 5 and 1%, respectively.The marginal effect is reported in the table, and the robust standard error is shown in parentheses IJCCSMR e t r a c t e d