Introduction

Global crop production reached 9.5 billion tons in 2021, with maize, wheat, and rice constituting 90% of total output1. Annually, 1.31 billion tons of crop residues from these staples are generated worldwide, concentrated in Asia (53.7%) and Europe (39.1%)2, the remainder is distributed across other regions including the Americas, Africa, and Oceania. Beyond these major grains, significant quantities of residues are also produced by vegetable and oil-bearing crops, which constitute an important yet often overlooked component of agricultural biomass3,4. Asian regions produce 764 million tons of rice straw, 74 million tons of wheat straw, and 9 million tons of maize straw5, while Europe yields 258 million dry tons of crop residue annually, primarily from wheat (42.2%), maize/barley (18.8%), and oilseed crops6. China’s crop residue output rose from 480 to 510 million tons (2013–2019), dominated by rice, wheat, and maize7, with vegetable and oil crop residues contributing substantially to the total biomass stream8, amplifying the urgency for sustainable management. Despite its resource potential, post-harvest crop residue is frequently burned or abandoned, particularly in developing regions, exacerbates air pollution, wildfire risks9, and correlates with 1.71–1.91% mortality rate increases per 10% burning increment10. Open-field burning affects 42.5% of residues in Karnataka, India11, 95% in the Philippines12, and 20% (140 Mt) of Chinese wheat straw13, releasing 3.4 Mt CO, 91 Mt CO2, and 1.2 Mt PM per 63 Mt combusted14.

Promoting the resourceful utilization of crop residue is therefore imperative. Adopting eco-friendly practices can reduce the global warming potential by 30%–57%15 and concurrently improve agricultural productivity and soil carbon storage16. Currently, the main ways for crop residue recycling utilization of crop residue worldwide include crop residue fertilizer (crop residue return), livestock feed, fuel (commercial and household energy), raw materials (for industrial and handmade products), and base materials (for bio-based materials)17 (Supplementary Materials 1). Utilization patterns, however, vary not only by region but also by residue type. Grain straws (e.g., from rice, wheat, maize) are commonly used for feed, bedding, and industrial purposes due to their favorable physical properties. In contrast, vegetable and oil crop residues, often characterized by higher moisture content, bulkiness, and seasonal availability, present distinct challenges for collection, transport, and storage3,8. This makes them more prone to being discarded or, where infrastructure and knowledge allow, preferentially used for on-site soil incorporation, composting, or as a component of animal feed in mixed farming systems4,8,18.

Crop residue recycling practices vary globally: in Japan, 61.5% of rice straw is incorporated as fertilizer19, while the U.S. retains over 90% of maize straw to prevent soil erosion20. As livestock feed, crop residue constitutes 81.4% of drought-period fodder in Ethiopia21 and 60% of animal feed in Nigeria22. Developed nations like Denmark and the U.S. utilize crop residue for bioenergy23,24, whereas developing countries such as India and Ghana predominantly use it for household fuel25,26.

Enhancing crop residue utilization faces multifaceted challenges, shaped by policy, market dynamics, and farmer behavior. China’s 1999 straw-burning ban reduced PM2.5 emissions by 46.9% (2013–2018)27, while subsidies and training increased recycling rates28,29. Similarly, Pakistan’s extension services improved adoption rates30. Market conditions also play a pivotal role: European farmers balance straw sales against fertilizer costs31, while Canadian farmers respond to higher crop residues recycling prices32. In India, straw-based industries create 5–10 jobs per unit33.

Various individual and household characteristics influence farmers’ decision-making of crop residue recycling utilization, including farm size34,35, age36,37, household income38, cultivated land size39, and education level40. Smallholders in Vietnam face resource and knowledge gaps41, while Indian farmers cite labor shortages and short sowing windows as barriers42. Conversely, awareness campaigns in Nepal reduced open burning by 34%43, and Chinese farmers adopting crop residue as a valuable resource improved harvesting efficiency44. Positive perceptions of crop residues’ economic and environmental benefits significantly enhance utilization willingness30,45.

China produces nearly one-third of the world’s crop residue46 has implemented stringent policies like the 1999 crop residue-burning ban and the 2016 Air Pollution Law, significantly reducing open-field burning47,48. These efforts boosted crop residue utilization rates from 82% (2017) to 88.1% (2021), with utilization rates of fertilizer, feed, fuel, bases, and raw materials at 60%, 18%, 8.5%, 0.7%, and 0.9%, respectively49. However, these aggregate statistics often mask significant variation across different residue streams. For instance, vegetable and oil crop residues may have distinct utilization pathways and lower overall recovery rates due to their unique physical and chemical properties, as observed in other agricultural systems4,8. At the same time, the significant differences in crop residue management between topographic landscapes were ignored. Terrain is a key constraint that determines the feasibility of agricultural mechanization and the cost of crop residue treatment50. In the plain area, the flat and contiguous terrain is conducive to the operation of large-scale machinery, which greatly reduces the marginal cost of crop stubble collection and transportation, which is a prerequisite for the realization of industrial scale utilization35. On the contrary, the agricultural system in hilly areas is characterized by land fragmentation, steep slopes and poor infrastructure, which severely limits the substitution of capital (machinery) for labor51. Farmers in plain areas may benefit from economies of scale, while small farmers in hilly areas often face excessive labor input and transportation difficulties. While existing research mainly focuses on plain areas or generalized analysis52, often ignoring the unique bottleneck mechanism of hilly terrain, which may lead to unfair burdens and low compliance in mountainous areas.

Therefore, it is necessary to determine the motivations, barriers, economic inputs and outputs, and influencing factors of crop residue management for key crop types including food, vegetables, and oil crops by smallholders in grain production areas of different terrains. The objectives of the study are as follows: (1) evaluate the average crop residue resource utilization rate, crop residue yield, motivation and obstacles of small farmers in plain and hilly areas of China’s major grain producing areas; (2) assess the economic inputs and outputs of crop residue recycling utilization across different regions and use methods; (3) explore the factors influencing farmers’ choice of different crop residue recycling utilization methods. The results of the study will offer insights to promote the resourceful application of crop residues, improve agricultural environments, and enhance the livelihoods of smallholder farmers in China.

Materials and methods

Questionnaire design and data collection

The research plan was approved by the Institutional Review Board of Henan Polytechnic University. Before participating, all subjects and/or their legal guardian(s) were informed of the purpose and contents of the study, to ensure the anonymity of their answers, and to provide informed consent according with the Declaration of Helsinki. Participation is entirely voluntary and respondents can opt out at any time.

This study was conducted from west to east at five sites in the main grain producing areas of central China (latitude 34° 05′ 28″ N to 35° 13′ 44.1′′ N, longitude 111° 39′ 14′′ E to 115° 18′ 21″E). The study area has a temperate semi-humid climate, with an annual rainfall ranging from 500 to 800 mm. The average annual temperature is 14 °C–21 °C in the plain area and 8–18 °C in the hilly area. Two sites (Luoning County and Xiuwu County) are located in the hilly area, while three sites (Xinzheng County, Qi County, Zhecheng County) are located in the plains.

We selected eight villages in hilly areas (Luoning County and Xiuwu County) and 11 villages in plain areas (Xinzheng County, Qixian County, and Zhecheng County) (Fig. 1). The respondents were household heads who mainly operate small-scale farms (less than two hectares). We conducted a questionnaire survey from February 21 to June 15, 2021. A structured questionnaire was designed to collect information from respondents on the following: (1) socio-economic characteristics of respondents, including years of education, age, gender, and agricultural income; (2) household agricultural production, including crop species, plantation area, and yield; (3) crop residue disposal methods of respondents, including utilization practices, quantities, and economic inputs and outputs of crop residue recycling utilization; (4) barriers, motivations, and variables influencing farmers’ crop residue recycling utilization ways (including individual and household characteristics, policies, sanctions, crop residue selling market, and cognitive factors). In the direct path analysis, variables can be categorized into five theoretical dimensions: individual characteristics (education years, age, gender), household socio-economic status (agricultural income, plantation area, crop species, crop residue yield), policy and market factors (training, subsidies, punishment, market availability) and cognitive factors (environmental value of crop residue, available value of crop residue, capital value of crop residue).

Fig. 1
figure 1

Location of the study area: (a) Xiuwu county, (b) Luoning county, (c) Xinzheng county, (d) Qixian county, (e) Zhecheng county.

Full size image

Four college students participated as interviewers in this survey. To ensure data quality and consistency across all interviews, a full-day training program was conducted prior to the field survey. This training covered the objectives of the research, detailed explanations of each questionnaire item, and standardized interviewing techniques. A key component of the training was a role-playing session where interviewers practiced administering the questionnaire and received feedback. A systematic questioning method was developed for introducing research and key issues to minimize interviewer-induced bias. Throughout the data collection period, regular meetings were held to address any field issues collectively and reinforce the standardized protocols, thereby ensuring inter-rater reliability. Each interview took approximately 30 min to complete, with voluntary participation from the interviewees. The procedures followed in this study comply with the Declaration of Helsinki and have been approved by the Ethics Committee (HUCH-2025-015). A total of 402 questionnaires were sent out and 382 complete questionnaires were retrieved, with a response rate of 95.02%, comprising 267 families residing in plain areas and 115 families residing in hilly areas (Supplementary Materials 2).

Data analysis

Data from 382 completed questionnaires were coded and analyzed using SPSS 23.0 and R program (version 4.1.1). The socio-economic characteristics of the farmers and their choices of crop residue recycling utilization were explored. The adoption rate, family crop residue yield, and economic inputs and outputs of crop residue recycling utilization were evaluated using descriptive statistical techniques (frequency, percentage, mean, and standard deviation). Individual characteristics, crop residue production, crop residue recycling utilization practices, inputs, and outputs were compared using one-way analysis of variance (ANOVA) between farmers in the plains and those in the hills. Correlation analysis and path analysis were employed to evaluate the direct and indirect impact of personal characteristics, planting conditions, policies, sales markets, and awareness on different crop residue utilization models in plain and hilly areas, respectively. To avoid overfitting, the models were not fully saturated.

Analysis of crop residue production, crop residue utilization inputs and outputs

Crop residue yield calculation

Crop residue yield was calculated using by crop yield, crop residue-to-grain ratio (SG) and the collectable coefficient (CC), following the methodology outlined by Ai et al.36 and Yang et al.37 The SG and CC used in this study are derived from previous research (Supplementary Materials 3). Crop residue yield (\(W_{S}\)) is calculated using Eq. 1 proposed by Seglah et al.26.

$${\text{W}}_{S}={W}_{P}\times SG\times {I}_{G}$$
(1)

where \(W_{P}\) = Crop yield; \(SG\) = Crop residue-to-grain ratio; \(I_{G}\) = Collectable coefficient.

Economic output of crop residue utilization

The economic output of crop residue utilization (\(UP_{i}\)) is calculated using Eq. (2).

$$f(UP_{i} ) = \sum\limits_{i = 1}^{n} {X_{i} U_{i} }$$
(2)

where \(UP_{i}\) = Output value of the i th type of crop residue utilization; \(X_{i}\) = the capital income of the i th type of crop residue utilization (the sum of cash income/substitution price and yield increase profit), \(U_{i}\) = Proportion of farmers’ utilization of the i th type of crop residue utilization.

The specific calculation for the substitution price of non-market uses (fertilizer, feed, fuel) and the yield increase profit are detailed as follows:

Crop residue as fertilizer: The substitution price was calculated based on the nutrient replacement value of crop residue. We first quantified farmers’ annual chemical fertilizer input (N, P₂O₅, K₂O). The amount of fertilizer replaced by crop residue was then determined using the substitution ratios established by Song et al.53: 38.4% for N, 18.9% for P₂O₅, and 85.5% for K₂O. The monetary value was derived using local market prices for chemical fertilizers. The yield increase profit was estimated by farmers based on their experiential knowledge and converted to monetary value using local grain purchase prices.

Crop residue as livestock feed: For simply processed crop residue, the substitution price was the local market price of crop residue fodder. For intensively processed crop residue (e.g., fermentation), the value was calculated as a proportion (78.76%) of the local commercial feed price, plus the monetary value of livestock weight gain (46.64%), based on the empirical model of Ma54 and local meat prices.

Crop residue as domestic fuel: The substitution price was based on the equivalent coal cost. Following Mei55 and Wang56, we assumed that one-third of household fuel came from crop residue and that 2 tons of crop residue replaces 1 ton of coal in calorific value. This was converted using local coal purchase prices.

Yield increase profit attribution: The attribution of yield increase to crop residue utilization (for both crop and livestock) is based on farmers’ self-reported estimates. This method, while containing an element of subjectivity, is widely employed in farm-level economic analyses57 as farmers possess localized knowledge of input–output relationships.

Economic input of crop residue utilization

Cash inputs were calculated as the sum of material inputs, including crop residue mulching machinery, equipment, pesticide, and herbicide inputs, crop residue crushing inputs, crop residue storage inputs, and transportation inputs, measured in USD/ha. Labor inputs were calculated based on labor employment costs (person /day/USD). According to Liu44, the local average labor employment cost is 0.366 USD/hour.

Results

Individual and household characteristics of the respondents

Males are more common in the plains (60.00%) and younger (52.17 ± 9.87 years) compared to those in the hilly areas. The average years of education for farmers in the plain areas are 7.56 ± 3.31 years, whereas farmers in hilly areas have an average of 5.90 ± 2.76 years of education. In the plain areas, farmers cultivate a significantly larger average of arable land (0.34 ± 0.19 ha) and more crop species (2.38 ± 0.68) relative to those in the hilly areas, where the average arable land is 0.24 ± 0.15 ha and the average crop species is 1.96 ± 0.55. The annual household income in the plain areas is 2339.82 ± 1257.37 USD, which is significantly higher than that in the hilly areas’ counterparts (1273.52 ± 668.16 USD) (Table 1).

Table 1 Socio-economic characteristics of the respondents and perceptions on recycling context.
Full size table

Policy, crop residue selling market and farmers’ awareness to crop residue recycling utilization

Farmers in plain areas received significantly higher subsidies (plain = 5.17 ± 6.75 USD, hilly = 3.55 ± 5.62 USD) and training times (plain = 5.17 ± 6.75, hilly = 3.55 ± 5.62 USD) than those in hilly areas. Additionally, farmers in the plain areas had more opportunities to connected with the crop residue selling market (plain = 0.58 ± 0.49, hilly = 0.35 ± 0.48), and experienced stricter punishments for burning bans (plain = 3.98 ± 0.95, hilly = 3.71 ± 1.03).

Farmers in hilly areas exhibited a higher recognition of the environmental value of crop residue recycling utilization (3.71 ± 0.87) than those in the plain areas (3.65 ± 1.14). Farmers in both hilly (3.57 ± 0.92) and plain (3.59 ± 1.11) areas perceived a relatively high availability of crop residue resources. The recognition of economic benefits of crop residue resource utilization by farmers in hilly areas (3.57 ± 1.06) was lower than that of farmers in plain areas (3.51 ± 0.92) (Table 1).

Crop and crop residue yield in the study area

The yields of maize (9974.99 ± 815.98 kg/ha), wheat (7615.35 ± 694.10 kg/ha), vegetables (17,625.00 ± 110.65 kg/ha) and oil-bearing crops (2625.00 ± 193.68 kg/ha) in the plain area were significantly higher than those in the hilly areas, which recorded yields of 9093.75 ± 682.47 kg/ha for maize, 6158.41 ± 90.42 kg/ha for wheat, 15,375.00 ± 629.72 kg/ha for vegetables, and 2287.50 ± 131.88 kg/ha for oil-bearing crops.

The yield of crop residue from maize (8610.41 kg/ha), wheat (8090.55 ± 737.41 kg/ha), vegetables (31,143.38 ± 195.52 kg/ha) and oil-bearing crops (3099.40 ± 228.69 kg/ha) in the plain areas was significantly higher compared to that in the hilly area, where yields were 7901.49 ± 675.77 kg/ha for maize, 6542.70 ± 96.07 kg/ha for wheat, 27,167.63 ± 1112.72 kg/ha for vegetables, and 2700.91 ± 155.72 kg/ha for oil-bearing crops (Fig. 2).

Fig. 2
figure 2

Yield of crop and crop residue between plains and hilly area. Abbreviation: MP, maize in plains; MM, maize in hilly areas; WP, wheat in plains; WM, wheat in hilly areas; VP, vegetables in plains; VM, vegetables in hilly areas; OP, oil-bearing crops in plains; OM, oil-bearing crops in hilly areas; * Significant at p < 0.05; **Significant at p < 0.01.

Full size image

In the plain area, each household produce average 7759.24 kg crop residues per year, while in the hilly area, each household produce average 3812.04 kg crop residues per year, which was significantly less than that in the plain area (p < 0.01) (Supplementary Materials 4).

Differences in crop residue recycling utilization between regions and crops

Differences in crop residue recycling utilization between regions

The overall crop residue recycling utilization by farmers was 82.59% across four major crops of wheat, maize, vegetable and oil-bearing crops. The uses rates ranked from highest to lowest were as follows: crop residue used for fertilizer (49.76%), livestock feed (19.18%), discarded or burned (17.41%), sold (9.29%), domestic fuel (2.94%), and raw material (1.41%) (Fig. 3a).

Fig. 3
figure 3

Categories of disposal (a) Crop residue resourceful utilization by total farmers, (b) Crop residue resourceful utilization by farmers in plain and hilly area, (c) Rates of crop residue recycling utilization across crops.

Full size image

The utilization rate in the plain area was 87.17%, while in the hilly area was 71.07%. The most common crop residue recycling utilization ways in the plain and hilly areas were crop residue used for livestock feed (plain = 18.75%, hilly = 20.25%) and crop residue fertilizer (plain = 52.80%, hilly = 42.15%). Other utilization ways were different, in the plain area, 12.83% of farmers chose to discard crop residue, 12.34% of farmers chose to sell crop residue, and the remaining crop residue was used for raw material (1.64%) and domestic fuel (1.64%). In the hilly area, in addition to crop residue feed and fertilizer, a larger proportion of farmers chose to discard crop residue (28.93%), while a smaller percentage opted to use crop residue for domestic fuel (6.20%), selling (1.65%) and raw material (0.83%) (Fig. 3b).

Differences in crop residue recycling utilization between crop species

The main crop residue recycling utilization ways of maize in the plain area are crop residue used for livestock feed (45.30%) and fertilizer (43.65%), while farmers in hilly area used maize crop residue as livestock feed (39.78%) or unutilized (45.16%). Wheat crop residue is predominantly returned as fertilizer in both plain area (81.97%) and hilly areas (93.68%). A large proportion of vegetable crop residue is unutilized (plain = 63.27%, hilly = 71.43%). The utilization of oil-bearing crop residue was diversified, in the plain areas, the oil-bearing crop residue was used for livestock feed (23.53%), fertilizer (41.18%), sold (15.29%) and domestic fuel (11.76%). Conversely, in hilly areas, oil-bearing crops were used for livestock feed (15.38%), fertilizer (23.08%) and domestic fuel (30.77%), with the remaining 30.77% of the crop residue being discarded or burned (Fig. 3c).

Economic input–output of crop residue recycling utilization based on regional differences

In the plain area, crop residue disposal can average save or earn 140.34 USD per hectare, and can lead to a yield increase of 24.50 USD per hectare. Alternatively, in hilly area, the crop residue disposal can average save or earn 125.77 USD per hectare, with lead to a yield increase of 18.78 USD per hectare, increased profits from crop/livestock yield are significant higher in the plains than in the hilly areas (Table 2).

Table 2 Costs, benefits and return on investment of farmers’ crop residue disposal for methods.
Full size table

The financial and labor inputs for crop residue disposal in the plain area are significantly higher than those in the hilly area. On average, farmers in the plain area invest 49.65 dollars and 30.05 dollars per hectare for financial and labor resources, respectively, while farmers in hilly areas invest an average of 37.82 dollars and 28.97 dollars per hectare for financial and labor resources. The financial input in the plain area is significantly higher than that in the hilly area, with no significant difference in labor input. Among the different disposal methods, discarding requires the least input, followed by crop residue fuel, crop residue raw material, crop residue sale, crop residue fertilizer, and crop residue feed (Table 2).

Return on investment (ROI) for the five treatments ranks highest with crop residue feed, followed by crop residue fuel, crop residue fertilizer, discard, crop residue raw material, and crop residue sale. The average ROI of crop residue recycling utilization in the plain area (113.92%) exceeds that in the hilly area (98.54%) (Table 2).

Motivation and obstacles of straw recycling utilization

The main motivation for farmers’ crop residue recycling utilization is environmental protection (plain = 3.49 ± 0.66, hilly = 3.55 ± 0.82), followed by crop residue recycling utilization (plain = 3.50 ± 0.80, hilly = 3.28 ± 0.74). The driving force for the profitability of crop residue recycling utilization is higher in the plain area (3.44 ± 0.90) than in the hilly area (3.04 ± 0.77). Additionally, farmers in the hilly area (3.75 ± 0.99) are more driven by the burning ban than farmers in the plain area (3.07 ± 0.98) (Fig. 4a).

Fig. 4
figure 4

Motivation and obstacles of farmers toward crop residue recycling utilization: (a) motivations; (b) obstacles. Notes: * Significant at p < 0.05; **Significant at p < 0.01.

Full size image

The main obstacles to farmers’ crop residue recycling utilization is the additional input required for crop residue recycling utilization (plain = 3.95 ± 0.77, hilly = 3.98 ± 0.85) and a lack of labor (plain = 3.65 ± 0.84, hilly = 3.83 ± 0.78). Farmers in hilly areas perceive more obstacles; their recognition of a lack of knowledge (3.79 ± 0.63), equipment (3.92 ± 0.69), and crop residue market availability (3.68 ± 0.83) is higher than that of farmers in plain areas (3.21 ± 0.84, 3.30 ± 0.87, and 2.79 ± 0.69, respectively) (Fig. 4b).

Influencing factors of farmers’ willingness to crop residues utilization

Influencing factors of farmers’ straw recycling utilization in plain area

The factors significantly associated with farmers in plain areas to use crop residue for livestock feed (6), fertilizer (4), sale (8), fuel (4), and raw materials (5) were measured by correlation analysis (Supplementary Materials 5). A path analysis was performed to examine the determinants of farmers’ crop residue recycling in the plain area. The model demonstrated good fit, with all indices (χ2/df, RMSEA, GFI, CFI, NFI) meeting standard criteria. Furthermore, the model accounted for a moderate to substantial share of the variance, with R2 values between 0.183 and 0.419 (i.e., 18.3–41.9%) across the outcomes. The significant paths are visualized in Fig. 5. Farmers’ use of crop residue as fertilizer was significantly affected by crop species (standardized path coefficients [SPC] = 0.26), subsidies (SPC = 0.12) and penalties (SPC = 0.12). Farmers’ use of crop residue as livestock feed was significantly affected by planting area (SPC = − 0.11), market availability (SPC = 0.09) and punishment (SPC = -0.12). Farmers’ crop residue sales were significantly affected by planting area (SPC = 0.32), subsidies (SPC = 0.21) and available value (SPC = 0.18). Farmers’ use of crop residue as domestic fuel was significantly affected by gender (SPC = − 0.14), crop residue yield (SPC = 0.12) and environmental value (SPC = − 0.13). Farmers’ use of crop residue as raw materials was significantly affected by gender (SPC = − 0.09), subsidies (SPC = − 0.09) and available value (SPC = − 0.10).

Fig. 5
figure 5

Factors affecting farmers in plain areas to use crop residue for fertilizer (a), livestock feed (b), sale (c), fuel (d), raw materials (f) and discarded (e). Notes: * Significant at p < 0.05; **Significant at p < 0.01.

Full size image

Influencing factors of farmers’ crop residue recycling utilization in hilly area

The factors significantly associated with farmers in hilly areas to use crop residue for livestock feed (10), fertilizer (10), sale (11), fuel (10), and raw materials (12) were measured by correlation analysis (Supplementary Materials 6). We employed path analysis to assess the factors influencing farmers’ crop residue recycling utilization. The model achieved acceptable model fit, as all key indices (χ2/df, RMSEA, GFI, CFI, NFI) adhered to recommended thresholds, confirming its robustness. It explained a moderate to substantial portion of the variance (R2 ranging from 0.310 to 0.440). The resulting significant paths are displayed in Fig. 6. Farmers’ use of crop residue as fertilizer was significantly affected by age (SPC = 0.40), crop residue yield (0.09), crop species (SPC = 0.26), environmental value (SPC = 0.17) and capital value (SPC = 0.13). Farmers’ use of crop residue as livestock feed was significantly affected by gender (SPC = 0.21), agricultural income (SPC = 0.22), crop residue yield (SPC = 0.13), punishment (SPC = − 0.32), environmental value (SPC = 0.08) and capital value (SPC = 0.23). Farmers’ crop residue sales behavior was significantly affected by age (SPC = 0.12), education years (SPC = 0.25), plantation areas (SPC = 0.27), training (SPC = − 0.13) and market available (SPC = 0.21). Farmers’ use of crop residue as domestic fuel was significantly affected by age (SPC = 0.46), plantation area (SPC = − 0.15), crop species (SPC = − 0.13), subsidies (SPC = − 0.12), available value (SPC = 0.25) and capital value (SPC = − 0.16). Farmers’ use of crop residue as raw materials was significantly affected by age (SPC = 0.27), education years (SPC = − 0.24), agricultural income (SPC = − 0.17), crop species (SPC = − 0.09), subsidies (SPC = 0.20), training (SPC = 0.19) and available value (SPC = − 0.11).

Fig. 6
figure 6

Factors affecting farmers in hilly areas to use crop residue for fertilizer (a), livestock feed (b), sale (c), fuel (d), raw materials (f) and discarded (e). Notes: * Significant at p < 0.05; **Significant at p < 0.01.

Full size image

Discussion

Smallholder-produced crop residue primarily originates from corn, wheat, vegetables, and oil crops, with fertilizer and livestock feed being the dominant uses in rural China27. Globally, 67–72% of farmers in Canada and Ghana prefer crop residue as fertilizer due to its simplicity48,58. However, complex methods like biogas production face low adoption rates, with crop residue-based renewable energy contributing less than 10% of electricity in Ukraine59. Limited knowledge and skills are key barriers, as seen in India33 and confirmed in our study. In China, rural labor migration exacerbates the issue, leaving women and the elderly in hilly areas with limited education and income60. Farmer field schools (FFS), proven effective in Bangladesh, Ethiopia, and Jamaica61,62,63, could address these gaps by enhancing knowledge and promoting diversified crop residue utilization. Given that our demographic analysis characterizes hilly areas by older, less educated, and female populations facing severe knowledge barriers, training methods are insufficient. We recommend tailored Farmer Field Schools where curricula for plain areas focus on machinery maintenance and deep-ploughing standards to optimize cost-efficiency. Conversely, training in hilly regions must target the "left-behind" demographic by promoting low-labor biological techniques, such as decomposing agents, and establishing "Women’s Mutual Aid Groups" to overcome information gaps.

Our study reveals a divergence in the primary obstacles perceived by farmers, a distinction crucial for effective policymaking. In the plains, the main barrier was the additional financial and labor input required for recycling. This reflects a more advanced but cost-sensitive stage of agricultural management in plain area, where farmers weigh the benefits of crop residue recycling against significant out-of-pocket expenses. Huang et al.27 emphasize that in China, recycling demands additional equipment and labor inputs, disproportionately burdening farmers who bear full financial responsibility while capturing minimal economic or environmental benefits. In particular, the high discarded rate for vegetables contrasts sharply with the much lower waste rate of grain stalks. A study from northern China has confirmed that crop residue waste such as roots, stems, leaves and rotten fruits is more likely to be discarded at will because composting with vegetable residues may lead to profit losses, and farmers are reluctant to adopt63. This challenge is not unique to China, in Spain and Sierra de Enguera, recycling costs can reach 22.5% of farmers’ income or be 1.93 times higher than traditional planting costs64,65. However, Subsidies for crop residue sales in our study area are scarce and insufficient to offset farmer costs, failing to incentivize effective crop residue recycling. This aligns with Sun et al.48, who note that China’s crop residue recycling subsidies remain negligible due to their low value and indirect allocation to farmers. As Qin and Lin66 and Yu et al.67 emphasize, targeted financial incentives could enhance farmer engagement and biomass energy efficiency. Thus, in plains and high-cost hilly areas, calibrated financial incentives must be implemented to effectively address cost barriers and enhance farmers’ participation. According to Supplementary Materials 4, the cash input for the sale of waste in the plain area is about 4.91 USD/ton, and that in the hilly area is 7.46 USD /ton. Comparing the willingness to pay for crop residues in the same research area68 (1.16–3.09 USD/ton in the plain area, 1.67–4.44 USD/ton in the mountainous area), this difference indicates that the existing subsidy level is not enough to offset the actual treatment costs, especially in the hilly area. Based on these cost constraints, we recommend that in plain areas, about 6-8USD/ton of compensation is sufficient to cover cash inputs and provide an effective incentive for surplus sales, while in hilly areas, higher compensation of about 9-12USD/ ton is required due to higher unit costs due to terrain-related scale constraints. These results highlight the limitations of uniform subsidy standards and support the necessity of establishing a region-specific compensation mechanism.

In contrast, hilly area farmers faced more foundational barriers, a pronounced lack of knowledge and equipment, coupled with limited market access. This points to a scenario where basic enabling conditions for crop residue recycling are absent. This situation is reminiscent of challenges faced by smallholders in Vietnam and India, where resource and knowledge gaps severely limit adoption41,58. China’s current policy, heavily reliant on burning bans, has been effective in reducing open-field burning27 but is a blunt instrument that fails to address these nuanced barriers. As observed in Punjab, effective extension services can increase crop residue recycling by 22%30, suggesting that in hilly regions of China, a shift from punitive to supportive measures—including knowledge transfer and equipment subsidies—is urgently needed. Agricultural extension systems play a vital role in promoting farmers’ adoption of eco-friendly practices. Seglah et al.26 linked insufficient extension training to farmers’ knowledge gaps and low crop residue recycling adoption. Despite deploying 550,000 personnel in 2019, China’s agricultural extension services (AES) showed limited impact, with over 50% of farmers in our study area reporting no relevant training69. In Shaanxi Plain, AES interventions had no measurable effect on crop residue return practices69. To address these gaps, China should prioritize context-specific extension reforms, particularly in hilly regions. Strategies include leveraging social media (e.g., WeChat, TikTok) and model farmers to amplify outreach, coupled with training programs tailored to local agroecological conditions.

A central finding of this study is the significant disparity in the Return on Investment (ROI) of crop residue utilization between plain (113.92%) and hilly (98.54%) areas. This efficiency gap can be largely attributed to differences in plantation scale and the feasibility of mechanization. Farmers in the plains, with larger average landholdings (0.34 ha) and higher collectable crop residue yields (7759.24 kg/household), are better positioned to leverage economies of scale. This aligns with findings from India and Thailand70,71, where higher per-unit crop residue yields reduce transportation and processing costs, making recycling more economically viable. This study demonstrates that plantation size and crop residue quantity critically influence farmers’ recycling practices. High-income smallholders with larger farms and greater crop residue output are more likely to utilize crop residue as livestock feed, fertilizer, or salable goods. Sun et al.48 attribute this to expanded farmland enabling mechanization, thereby lowering long-term crop residue incorporation costs. Similar patterns emerge in India and Thailand, where higher per-unit crop residue yields reduce transport expenses70,71. In hilly regions, limited scale effects restrict market access and mechanization for smallholders, women, and elderly farmers72. Here, manual harvesting of maize, vegetables, and oil crops leads to low yields, favoring labor-intensive disposal methods like domestic fuel use or crop residue crafts over mechanized recycling. This aligns with rural Turkey, where women predominantly use crop residue biomass for household energy73. Wang et al.74 corroborate the labor-mechanization trade off, noting plains areas require 72% less mechanical input per hectare for crop residue recycling than hilly zones. In China, constrained farmland scale hinders technical efficiency gains, with 200 million smallholders operating sub-2-hectare plots67. Land fragmentation exacerbates these challenges75. To address this, policymakers should streamline rural land transfers, consolidate farmland, and scale mechanized crop residue management to boost agricultural sustainability.

In addition to large-scale agricultural production, a flourishing crop residue recycling market is crucial44. Our study found over half of hilly-area farmers lacked market access, a key barrier despite their strong motivation to sell crop residue for income. Many resorted to roadside or field discards due to no buyer contact, allowing companies to collect crop residue without compensation and diminishing farmers’ economic gains. Wu et al.76 link such market deficiencies to higher marketing costs and reduced transaction ease. High collection/transportation expenses further hinder market activity, obstructing energy plants requiring large-scale, stable supplies. Munthali and Xuelian77 note China’s uneven crop residue market distribution and weak policy support, lowering buyer profitability and motivation. Globally, nations with underdeveloped markets are advancing crop residue recycling to meet environmental goals. To resolve the constrains where market access in hilly areas is significantly lower than in plains, a government-supported "Hub-and-Spoke" logistics system should be considered. Subsidies should prioritize village-level transfer stations and small-scale vehicles to bridge the “First Mile” gap in fragmented terrain, thereby aggregating biomass to an economic scale for commercial transport. Furthermore, for vegetable residues with high moisture content and discard rates exceeding 60%, we advise against transport. Policies should instead support on-site processing facilities like field-side fermentation tanks to achieve localized circularity without incurring prohibitive logistical costs.

Another notable finding of our research is the high discarded rate of vegetable crop residue, which stands in sharp contrast to the much lower abandonment rates for cereal residues. This pattern reflects a combination of structural, agronomic, and market-related constraints. Unlike rice, wheat, or maize, vegetable production cycles are shorter and generate residues with higher moisture content, lower storage stability, and limited mechanization options, making collection and reuse more labor-intensive78. Studies from China, Vietnam79, and India80 similarly note that vegetable residues are often left in fields because their high-water content restricts transportability and accelerates decomposition. Furthermore, vegetable crop residue has weaker market demand, which is less suitable for livestock feed, rarely used as industrial raw material, and is often excluded from crop residue-purchase programs that prioritize cereal crops. Socioeconomic factors also contribute to the high discard rate. Vegetable farmers, who often operate small-scale, multi-cycle production systems, face limited time windows between harvests and planting, leaving insufficient labor for residue processing. The combination of low economic return, high labor inputs, and insufficient technical support helps explain why vegetable crop residue remains the least utilized residue type despite its substantial biomass potential.

Limitations and future prospects

This study has several limitations. First, key data on economic inputs, outputs, and utilization practices were self-reported by farmers and were not independently verified, which may introduce self-reporting bias. Second, the cross-sectional nature of our survey design, while employing path analysis to identify associations, inherently limits our ability to establish causal relationships between the factors studied and farmers’ crop residue utilization behaviors. Third, regarding sample representativeness, the 19 villages across five counties were purposively selected to capture the typical agricultural systems and topographical features (plains vs. hills) of the central grain-producing region in Henan Province. While this approach ensures relevance to the research context, the findings may not be fully generalizable to all plain or hilly areas across China. Finally, given the well-known government bans on crop residue burning, farmers might have been susceptible to social desirability bias, potentially leading to an over-reporting of recycling rates and an under-reporting of burning or discarding practices, despite our efforts to ensure anonymity and minimize this risk during the interviews.

To address these limitations and advance this research field, future studies should: (1) incorporate objective data sources (for example, remote sensing using patterns, precise yield measurements) to verify and supplement farmers’ reports; (2) Adopt longitudinal design to track behavioral changes and establish a causal mechanism; (3) A hybrid approach is adopted, combining quantitative investigations with in-depth qualitative interviews to better reveal underlying motives and alleviate social desirability biases.

Furthermore, we emphasize that future research should increasingly explore the potential integration of artificial intelligence and machine learning (AI/ML) methods. These technologies can significantly enhance the predictive modeling of crop residue utilization trends, optimize the resource allocation of the crop residue recycling market, and identify the complex nonlinear relationships among socio-economic, policy and environmental factors that may be overlooked by traditional statistical methods.

Conclusion

This study contributes to existing literature by analyzing the application rate, efficiency, input–output dynamics, willing and influencing factors of farmers in plain and hilly areas regarding crop residue resources. Farmers mainly use corn and wheat crop residue for feed and fertilizer, with limited exploration of other application methods. The adoption rate and efficiency of crop residue are higher in plain areas in contrast to hilly areas. Farmers are motivated to utilize crop residue resourcefully to protect the environment and promote recycling, despite facing challenges such as high investment requirements for additional use methods and limited crop residue sales markets which dampen their enthusiasm. The regional differences in economic efficiency are significant. The return on investment in plain areas is higher than that in hilly areas (113.92% vs. 98.54%). The behavior of farmers in crop residue recycling utilization is primarily affected by the size of planting area, agricultural income, and crop residue yield. Farmers with larger arable land and greater crop residue resources tend to engage in crop residue recycling utilization for feed, fertilizer, and sales, benefiting from economies of scale in crop residue processing. We also identify contrasting behavioral barriers and a previously overlooked gender effect in plains that is absent in hilly regions. In hilly areas, women and elderly farmers often resort to labor-intensive crop residue recycling utilization methods like household fuel and raw materials, resulting in lower efficiency compared to plain areas. Based on these findings, we recommend targeted interventions to enhance crop residue utilization: (1) differentiated Farmer Field Schools (FFS), focusing on machinery operation and deep-ploughing in plains, and low-labor biological methods and women’s mutual aid groups in hilly areas; (2) tailored financial incentives to offset investment costs, approximately 6–8 USD/ton in plains and 9–12 USD/ton in hilly areas; (3) land consolidation and contiguous management to enable mechanized, large-scale residue collection; and (4) development of regional residue markets, including centralized collection networks and industrial linkages, with hub-and-spoke logistics and on-site processing for high-moisture residues in hilly areas.