Research on the Low-Carbon Effects of Farmers’ Livelihood Transition in the Northeast China Tiger and Leopard National Park: A Case Study of Dongning Area

Abstract

Identifying the low-carbon effects of farmers’ livelihood transformation in the Northeast Tiger and Leopard National Park is crucial for rural communities’ green development. Based on survey data from farmers in the Dongning area of the park, this study uses the carbon emission inventory method to analyze farmers’ livelihood transformation paths and low-carbon effects from 2016 to 2023. The results show that (1) farmers’ livelihood strategies in the park are categorized into traditional agriculture-dominated, traditional non-agriculture-dominated, specialty farming/breeding, diversified, and non-agriculture-dominated. From 2016 to 2023, 18.22% of farmers changed their livelihood strategies. (2) Per capita household carbon emissions of farmers in the park generally decreased. Specifically, per capita carbon emissions from living consumption declined, while those from production slightly increased. (3) Among transformation paths, shifts to traditional agriculture-dominated or specialty farming/breeding from traditional non-agriculture-dominated or non-agriculture-dominated strategies led to increases in per capita household carbon emissions. In contrast, other transformed groups and those maintaining their original strategy saw decreases in per capita household carbon emissions.

1. Introduction

The most fundamental measure to protect Amur tigers and Amur leopards lies in preserving their habitats. During farmers’ livelihood transition processes, reducing fertilizer usage and deforestation directly contributes to lowering local carbon emissions, while indirectly manifests in habitat restoration that promotes population growth of these big cats. It is imperative to minimize anthropogenic disturbances to natural ecosystems, prioritizing natural restoration complemented by auxiliary measures. The application of advanced yet practical restoration and governance technologies can enhance biodiversity and improve ecosystem carrying capacity, thereby revitalizing the ecological environment of tiger and leopard habitats. Current research confirms a reciprocal feedback mechanism between rural livelihoods and ecological systems. Ecologically, these environments provide essential natural resources and spatial foundations for household survival. Socioeconomically, as the primary decision-making entities and dominant economic units within the national park framework, farmers’ livelihood strategies fundamentally determine land-use patterns, resource utilization efficiency, and exert profound impacts on the ecological rehabilitation of felid habitats [1,2].

Scientific evaluation of the environmental impacts from farmers’ livelihood strategy transitions constitutes the strategic pivot and operational foundation for regulating sustainable development pathways. Under the global warming trend, carbon emissions have emerged as critical indicators reflecting ecological conditions, with agricultural production and household consumption accounting for substantial components. To achieve China’s dual carbon goals, the government has instituted the “15 Household Low-Carbon Initiatives” under current socioeconomic conditions, encompassing practical measures like energy-efficient lighting adoption, reduced bottled water consumption, and minimized air conditioning usage. Complementing this, the Intergovernmental Panel on Climate Change (IPCC) identifies household consumption pattern modification as a viable approach for carbon mitigation and climate change adaptation [3]. Carbon emission dynamics serve as a mediating factor in the environmental repercussions of livelihood transitions. By employing household carbon emission characteristics as an analytical medium, researchers can quantitatively assess the ecological footprint of residential activities, thereby establishing a methodological framework for evaluating low-carbon outcomes in agrarian livelihood transformations [4].

However, most previous studies have relied on household consumption statistics and focused on carbon emissions at the national or regional level. Few have examined how changes in household livelihood strategies influence household-level carbon emissions from the perspective of individual household livelihood activities. Emission inventory methods are typically used to calculate direct carbon dioxide emissions from household consumers’ daily activities, such as lighting and cooking, which result from the use of energy and resources [5]. This method was explicitly defined by the IPCC in the 2006 IPCC Guidelines for National Greenhouse Gas Inventories, and it has since been widely adopted in academic research for estimating direct carbon emissions from household consumption. According to the definitions and computational protocols established in emission inventory methodology, direct CO₂ emissions originate from the utilization of direct energy sources. The quantification of CO₂ emissions derived from specific direct energy consumption is governed by two primary determinants: the consumption volume of the energy source and the carbon intensity coefficient inherent to its specific fuel type. The specific calculation method involves multiplying the energy consumption of households in different activities by the corresponding carbon emission coefficients and then summing up these results to obtain the total direct carbon dioxide emissions from these activities.

Since the establishment of the Northeast Tiger and Leopard National Park, the livelihood capital of farmers in and around the park has continuously improved, their livelihood strategies have been optimized, and their household consumption structures have undergone changes, all of which impact family carbon emissions. Research on the ecological effects of farmers’ livelihood transformation in the Northeast Tiger and Leopard National Park helps understand the developmental dynamics of local farmers’ livelihoods and their carbon emission characteristics, providing insights for promoting sustainable development in the region. To this end, this study employs the carbon emission inventory method. By identifying pathways of farmers’ livelihood transformation, it analyzes how such transformation affects household carbon emissions, ultimately drawing conclusions and proposing policy recommendations.

2. Research Methods

2.1. Study Area

This study selects the Northeast Tiger and Leopard National Park as its empirical research subject, being among the first five officially established national parks in China. Spanning Jilin and Heilongjiang provinces while bordering Russia and North Korea, the park encompasses 14,065 km², consolidating 19 protected areas within its boundaries. It constitutes the core habitat hosting the largest, most active, and most critical breeding populations of Amur tigers and leopards in China, simultaneously representing a vital wildlife distribution zone and one of the most biodiverse temperate regions in the Northern Hemisphere. The Heilongjiang section covers 450,800 hectares (32% of the total park area), with the Dongning Branch Administration governing 53,400 hectares in southern Dongning City. This jurisdiction extends from Liangzichuan Village (northeast) to Sanjianlazi Village (south), Xiaolanni Ditch (west), and Huangsongdianzi Village (north), situated between 130°21′46″ and 131°13′59″ E longitude and 43°33′50″–44°0′18″ N latitude.

2.2. Data Sources

The research data primarily stems from farmer questionnaires and in-depth interviews. The study area encompasses communities within the Dongning area of the Northeast China Tiger and Leopard National Park, specifically including 13 villages (natural hamlets) in Laoheishan Town, 10 villages in Daduchuan Town, and 5 villages in Sanchakou Korean Ethnic Town, totaling 28 villages and hamlets along with two state-owned forest farms. A total of 539 questionnaires were randomly distributed within selected administrative villages. After eliminating samples with substantial missing information and obvious response errors, 527 valid household samples were ultimately obtained, yielding a questionnaire validity rate of 97.77%. The spatial geographic distribution of surveyed townships is illustrated in Figure 1.

The content of the survey questionnaire includes (1) basic information about rural households in 2016 and 2023, including basic information about family members and household characteristics (gender, age, education level, etc.); (2) the production and income status of rural households in 2016 and 2023; and (3) major consumption data of rural households in 2016 and 2023, including direct energy use (straw, coal, gasoline, electricity, etc.) and living expenses (food, clothing, housing, medical care, education, tourism, etc.).

The respondent households exhibited the following characteristics (

Table 1. The characteristics of the 527 households interviewed.

Survey Content		Year
		2016	2023
Household size (persons per household)		3	-
Number of labor force members/(persons per household)		2	-
Annual household income (yuan)		75,149.67	88,846.20
Total annual household living expenses (yuan)		42,174.02	38,543.26
Educational attainment of labor force (%)	Illiterate	1.71	-
	Primary school	19.92	-
	Junior high school	65.65	-
	High school	9.11	-
	College and above	3.61	-

“-” indicates that there has been no change in 2023 compared with 2016.

): (1) Smaller household size, with an average of 3 members per household. (2) Fewer labor force members, with an average of 2 laborers per household, accounting for two-thirds of the total household population. (3) Moderate educational attainment among labor force members, with most having received education above the junior high school level. (4) An upward trend in household annual income. The average annual income of rural households increased from 75,149.67 yuan in 2016 to 88,846.20 yuan in 2023, representing an increase of 18.23%. (5) A decline in annual household living expenses. The average annual expenditure of rural households decreased from 42,174.02 yuan in 2016 to 38,543.26 yuan in 2023, a reduction of 8.61%.

2.3. Methodology

2.3.1. Classification of Farmer Livelihood Strategies

The classification of livelihood strategies is a prerequisite for characterizing livelihood transformation pathways. Field surveys show that there are five types of livelihood activities among rural households within and around the Northeast Tiger and Leopard National Park: traditional agricultural activities, such as growing grain crops (rice, corn), legume crops (soybeans, etc.), and cash crops like melon seeds and Chinese medicinal materials (hereinafter referred to as “grain and cash crops”); specialty farming/breeding, mainly involving specialty crops and livestock, such as black fungus and natural honey; and non-agricultural livelihood activities, including long-term migrant work, irregular employment (casual labor), ecological protection, self-employment, and livestock breeding (cattle, sheep, pigs, chickens, etc.). Most existing studies have defined household livelihood types from a diversified perspective [6,7], based on the primary direction of labor input and the proportion of various sources of income. In light of this, this paper, based on the identified livelihood activities, quantitatively considers the household income structure and qualitatively analyzes the main direction of labor input. It divides the livelihood strategies of rural households in the NTLNP into five categories: traditional agriculture-dominated, traditional non-agriculture-dominated, specialty farming/breeding, diversified, and non-agriculture-dominated (

Table 2. Definition of farmer’s livelihood types in the Northeast Tiger and Leopard National Park.

Types of Livelihood Strategies	Proportion of Income			Primary Direction of Labor Input
	Traditional Agricultural Production Income	Specialty Agricultural Production Income	Non-Agricultural Sector Earnings
Traditional agriculture-dominated type	70% and above	0~30%	0~30%	Staple and Cash Crop Cultivation, Livestock and Aquaculture Farming
Traditional non-agriculture-dominated type	50% and above	0~50%	0~50%	Staple and Cash Crop Cultivation, Livestock and Aquaculture Farming
Specialty farming/breeding type	0~40%	60% and above	0~40%	Specialty Agricultural Production
Diversified type	0~50%	0~60%	0~50%	Conventional Agriculture, Specialty Production, and Non-Agricultural Livelihood Diversification
Non-agriculture-dominated type	0~50%	0~50%	50% and above	Non-Agricultural Activities and Alternative Livelihood Strategies

).

2.3.2. Carbon Emission Accounting for Farmers

Based on survey data on farmers’ livelihood capital and primary livelihood activities in the NTLNP from 2016 to 2023, this study refers to the carbon emission inventory method proposed by the IPCC to calculate household carbon emissions. The formula is as follows:

$$C=\sum C_i+\sum C_j$$

(1)

In the formula, $C$ represents the household carbon emissions of rural families; $C_i$ represents the production-related carbon emissions of rural households; and $C_j$ represents the consumption-related carbon emissions of rural households.

(1)

Production Carbon Emissions

The production carbon emissions of rural households are mainly considered in the following three aspects: (1) greenhouse gas emissions caused by agricultural material inputs; (2) greenhouse gas emissions generated by crop planting; and (3) greenhouse gas emissions caused by livestock and poultry farming.

Combining previous research results and consulting relevant experts, it is believed that the carbon emissions generated by agricultural material inputs mainly come from two aspects: First, the direct or indirect carbon emissions caused by the input of agricultural materials such as fertilizers, pesticides, and agricultural films (ground films); second, the carbon emissions caused by the electricity consumption of agricultural plowing and irrigation activities. The carbon emissions generated by agricultural material inputs mainly come from fertilizers, pesticides, agricultural films, agricultural plowing, agricultural irrigation, agricultural diesel, and agricultural machinery. The calculation formula is as follows:

$$C_z=\sum Q_{zi}\times {\alpha }_{zi}$$

(2)

where $C_z$ is the carbon emissions from agricultural production material inputs; $Q_{zi}$ is the input amount of each agricultural production material; and ${\alpha }_{zi}$ is the carbon emission coefficient for each agricultural production material, with values provided in

Table 3. Carbon emission coefficient of agricultural production material input.

Agricultural Production Materials	Emission Coefficient (Calculated as Carbon)	Reference
Fertilizers	0.8956 kg·kg−1	[8,9,10,11,12]
Pesticides	4.9341 kg·kg−1
Plastic film	5.18 kg·kg−1
Agricultural diesel oil	0.5927 kg·kg−1
Agricultural irrigation	20.4760 kg·hm−2
Agricultural machinery equipment	0.1800 kg∙kW−1
Agricultural ploughing	3.1260 kg·hm−2

.

The main crops in the farmland of the Northeast Tiger and Leopard National Park and its surrounding areas include upland crops such as corn, soybeans, rice (medium-season rice), and potatoes, as well as economic crops such as wood ear mushrooms. Factors affecting carbon emissions include crop planting area, crop type, and their carbon emission coefficients. As the carbon emission coefficient for wood ear mushrooms couldn’t be found in the literature, this paper roughly estimated it based on the main energy-consuming links involved in wood ear mushroom planting. The carbon emission coefficient for wood ear mushroom planting can be considered in three aspects: mushroom spawn making, planting, and drying. Since plastic film and electricity/coal used in drying are calculated in other links, the carbon emission coefficient for wood ear mushrooms mainly comes from spawn making. Spawn is made from raw materials such as sawdust and bran; considering the energy consumption of raw material production, transportation, and processing, the carbon emission for producing 1 kg of spawn is estimated at about 0.5 kg of CO₂ equivalent. Generally, each 1 kg of spawn produces about 0.15 kg of wood ear mushrooms. Thus, the carbon emission coefficient for the spawn link per kilogram of wood ear mushrooms is roughly 3.3 kg of CO₂ equivalent. The carbon emission calculation formula for major crops is as follows:

$$C_{pi}=\sum Q_{pi}\times {\alpha }_{pi}$$

(3)

where $C_{pi}$ is the carbon emissions from crop cultivation; $Q_{pi}$ is the planting area of each type of crop; and ${\alpha }_{pi}$ is the carbon emission coefficient for each crop, with values provided in

Table 4. Carbon emission coefficient of main crops.

Crop Type	Emission Coefficients	Reference
Corn	2.532 (calculated as N2O)/(kg·hm−2)	[8,13,14,15,16]
Rice	0.24 (calculated as N2O)/(kg·hm−2)
Soybean	0.77 (calculated as N2O)/(kg·hm−2
Wood ear mushroom	3.3 (calculated as C2O)/(kg·kg−1)
Flue-cured tobacco	5167.80 (calculated as C)/(kg·hm−2)
Vegetable	4.21 (calculated as N2O)/(kg·hm−2)
Other dryland crops	0.95 (calculated as N2O)/(kg·hm−2)

Note: According to the greenhouse effect caused by 1 kg of N₂O, it is equivalent to 273 kg of CO₂, and 1 kg of CO₂ equals 12/44 kg of carbon. N₂O emissions can be converted to carbon emissions.

.

In livestock farming, enteric fermentation in animals produces large amounts of CH₄, and manure management also generates significant amounts of CH₄ and N₂O [17]. Currently, the main types of livestock raised by rural households in the Northeast Tiger and Leopard National Park are cattle, mules, donkeys, sheep, pigs, and chickens. The emission calculation formulas are as follows:

$$C_{ci}=\sum Q_{ci}\times {\alpha }_{ci}$$

(4)

$$C_{fi}=\sum Q_{fi}\times {\alpha }_{fi}$$

(5)

In the formula, the symbols represent the carbon emissions from enteric fermentation and manure management of livestock, respectively; $Q_{ci}$ and $Q_{fi}$ represent the number of each type of livestock; and ${\alpha }_{ci}$ and ${\alpha }_{fi}$ represent the carbon emission coefficients for enteric fermentation and manure management of each livestock type, with values provided in

Table 5. Greenhouse gas emission coefficient of livestock and poultry breeding.

Animal Type	CH4 Emission Coefficient/(kg∙head−1∙a−1)		N2O Emission Coefficient/(kg∙head−1∙a−1)	Reference
	Enteric Fermentation	Fecal Management	Fecal Management
Cattle	47.8	1	1.39	[18,19,20]
Pigs	1	3.5	0.53
Sheep	5	0.16	0.33
Donkeys/Mules	10	0.9	1.39
Poultry	—	0.02	0.02
Bees	1 (calculated as CO2)/(t·t−1)

Note: According to the greenhouse effect caused by 1 kg of N₂O, it is equivalent to 273 kg of CO₂, and 1 kg of CO₂ equals 12/44 kg of carbon. N₂O emissions can be converted to carbon emissions; the carbon emission calculation for honey includes the entire beekeeping production process, assuming one beehive produces 25 kg of honey per year.

.

(2)

Consumption Carbon Emissions

Household energy use and structure significantly impact the ecological environment. In this study, carbon emissions from rural household energy use are divided into direct energy-related emissions and consumption-related emissions. Rural households mainly use coal, liquefied gas, gasoline, and electricity, along with large amounts of firewood and straw. Carbon emissions from household energy use depend on the consumption of these energy types and their corresponding emission factors. The calculation formula is as follows:

$$C_{di}=\sum Q_{di}\times {\alpha }_{di}$$

(6)

where $C_{di}$ represents the carbon emissions from direct household energy consumption; $Q_{di}$ represents the usage of each direct energy type (e.g., coal, liquefied gas, gasoline, electricity, firewood, straw); and ${\alpha }_{di}$ represents the carbon emission coefficient from combustion for each energy type, with values provided in

Table 6. Carbon emission coefficient of direct energy consumption.

Energy Type	Emission Coefficients	Reference
Coal	0.5601 t∙t−1, (calculated as C)	[21,22,23]
Gasoline	0.8146 t∙t−1, (calculated as C)
Liquefied Gas	0.8634 t∙t−1, (calculated as C)
Electricity	0.6342 kg·(kWh)−1, (calculated as CO2)
Firewood	1.5000 × 104 t∙(104 t)−1, (calculated as CO2)
Straw	1.2038 × 104 t∙(104 t)−1, (calculated as CO2)
Mushroom Spawn Bag	1.5000 t∙t−1, (calculated as C)	Estimated

Note: 1 kg of CO₂ equals 12/44 kg of carbon; CO₂ emissions can be converted to carbon emissions.

.

Consumption by farmers is divided into food, tobacco and alcohol, clothing, housing, household goods and services, transport and communications, education, culture and entertainment, medical care, and other services. Therefore, carbon emissions from consumption are determined by the type of household consumption, expenditure, and related emission factors, and the calculation formula is as follows:

$$C_{inj}=\sum Q_{inj}\times {\alpha }_{inj}$$

(7)

where $C_{inj}$ represents the carbon emissions from living consumption; $Q_{inj}$ represents the expenditure on various types of living consumption; and ${\alpha }_{inj}$ represents the carbon emission coefficient for each type of living consumption expenditure, with values provided in

Table 7. Carbon emission coefficient of living consumption and reference sources.

Category	Emission Coefficient (Calculated as CO2)/(kg∙yuan−1)	Reference
Food	0.095	[20]
Clothing	0.126
Residence	0.192
Household goods and services	0.158
Transportation and communication	0.159
Education, culture, and recreation	0.160
Medical care	0.177
Other services	0.064

Note: 1 kg of CO₂ equals 12/44 kg of carbon; CO₂ emissions can be converted to carbon emissions.

.

To calculate carbon emissions from household living consumption, the following methods were used: food and beverage purchases were used for food-related emissions, clothing purchases for clothing emissions, housing construction or purchases for housing emissions, durable goods purchases for household goods and services emissions, electricity, communication, and fuel expenses for transport and communication emissions, education and tourism expenses for education, culture, and recreation emissions, and medical expenses for healthcare emissions. Other service expenditures were negligible and excluded from the carbon emission calculations.

3. Results

3.1. Trajectories of Farmers’ Livelihood Strategy Transformation

From 2016 to 2023, the livelihood strategies of 18.22% of farmers changed, mainly shifting from traditional agriculture-dominated to traditional non-agricultural and from traditional non-agricultural to non-agricultural-dominated. The remaining 81.78% of farmers retained their original livelihood strategies, with the most common being specialty farming/breeding and non-agricultural dominated strategies (Figure 2).

3.2. Farmers’ per Capita Household Carbon Emissions

3.2.1. Changes in Farmers’ per Capita Household Carbon Emissions

Based on the identified livelihood strategies of farmers in the Northeast Tiger and Leopard National Park in Section 2.3, in 2016, the number of households in the study area categorized as traditional agriculture-dominated, traditional non-agriculture-dominated, specialty farming/breeding, diversified, and non-agriculture-dominated were 141, 112, 114, 26, and 134, respectively. Among these, traditional agriculture-dominated households had the highest proportion at 26.76%, while diversified households had the lowest at 4.93%. By 2023, the number of households in these categories changed to 108, 103, 112, 37, and 167, respectively, with non-agriculture-dominated households increasing to 31.69% of the total. This indicates a continuous rise in the non-agriculturalization of livelihoods in the park’s communities.

From 2016 to 2023, the per capita household carbon emissions in the communities of the Northeast Tiger and Leopard National Park showed an overall downward trend, decreasing by 11.15%. Overall, the per capita carbon emissions from living consumption were significantly higher than those from production activities. In terms of living carbon emissions, the per capita carbon emissions from living consumption decreased by 9.53%, while those from direct energy consumption decreased by 20.46%. This suggests that the reduction in carbon emissions from direct energy use, such as straw (corn stalks), coal, and firewood, was the key factor driving the decrease in per capita household carbon emissions among farmers in the park. In contrast, per capita production of carbon emissions showed an increasing trend, rising by 8.18%. Within this category, the per capita carbon emissions from livestock farming decreased the most, by 11.15%, while those from crop cultivation increased the most, by 22.70%. This indicates that livestock farming has relatively strong potential for emission reduction in the production processes of farmers (see Figure 3).

3.2.2. Changes in per Capita Household Carbon Emissions Among Farmers of Different Livelihood Types

From 2016 to 2023, carbon emissions from crop cultivation, agricultural production material inputs, and living consumption all showed upward trends across the traditional agriculture-dominated, traditional non-agriculture-dominated, specialty farming/breeding, diversified, and non-agriculture-dominated types. Among these, the traditional non-agriculture-dominated type saw the largest increase in crop cultivation carbon emissions at 30.05%, while the specialty farming/breeding type had the smallest rise in agricultural production material inputs at 3.64% (see

Table 8. Per capita household carbon emission structure of farmers with different livelihood types. Unit: kg (calculated as carbon).

Type	Traditional Agriculture-Dominated Type		Traditional Non-Agriculture-Dominated Type		Specialty Farming/Breeding Type		Diversified Type		Non-Agriculture-Dominated Type
	2016	2023	2016	2023	2016	2023	2016	2023	2016	2023
Crop Cultivation	218.40	277.25	151.05	196.45	952.35	1136.55	144.30	187.50	94.20	116.75
Livestock Farming	213.20	196.30	140.85	129.65	75.70	54.45	115.90	106.20	79.50	68.85
Agricultural Production Material Inputs	1052.30	1104.60	762.25	805.95	1074.15	1113.25	702.90	743.35	460.60	220.00
Direct Energy Consumption	2308.75	1739.25	2234.45	1659.70	2503.30	1949.55	2228.70	1665.30	2264.50	1702.95
Living Consumption	332.89	364.63	307.29	336.59	348.90	382.16	272.33	298.30	345.70	378.66
	4125.54	3682.03	3595.88	3128.33	4954.39	4635.96	3464.13	3000.65	3244.5	2487.20

).

Looking at per capita production carbon emissions, from 2016 to 2023, among the different livelihood types, the diversified type had the largest decrease in direct energy consumption at 28.07%; the traditional non-agriculture-dominated type followed with a 25.72% reduction; and the traditional agriculture-dominated type had the smallest decline in livestock farming at 7.92%. Notably, the traditional agriculture-dominated type saw a rise in per capita production of carbon emissions from crop cultivation, increasing by 26.94%.

In terms of per capita living carbon emissions, all livelihood types experienced increases in living consumption-related emissions, with relatively consistent growth rates. The diversified type had the highest increase at 9.536%, indicating it had the weakest potential for emission reduction.

During the livelihood transformation process, household carbon emissions of farmers also changed (see

Table 9. State transition matrix of carbon emissions of farmers with different livelihood transformation paths. Unit: kg (calculated as carbon).

		2023
		Traditional Agriculture-Dominated Type	Traditional Non-Agriculture-Dominated Type	Specialty Farming/Breeding Type	Diversified Type	Non-Agriculture-Dominated Type
2016	Traditional agriculture-dominated type	−443.51	−997.20	510.42	−1124.89	−1638.33
	Traditional non-agriculture-dominated type	86.15	−467.55	1040.08	/	−1108.68
	Specialty farming/breeding type	−1272.36	/	−318.43	−1953.75	−2467.19
	Diversified type	/	/	/	−463.48	−976.92
	Non-agriculture-dominated type	437.53	−116.16	1391.46	−243.85	−757.29

). Specifically, farmers shifting from traditional non-agriculture-dominated or non-agriculture-dominated to traditional agriculture-dominated, or from traditional agriculture-dominated or traditional non-agriculture-dominated to specialty farming/breeding, saw an increase in per capita household carbon emissions. The largest increase was observed among those moving from non-agriculture-dominated to specialty farming/breeding (28.09%), while the smallest was among those shifting from traditional non-agriculture-dominated to traditional agriculture-dominated (2.09%). In contrast, farmers transitioning into other livelihood types generally saw a decrease in per capita household carbon emissions. The most significant reduction was among those moving from specialty farming/breeding to non-agriculture-dominated (−76.04%), while the smallest decrease was among those shifting from traditional agriculture-dominated to diversified (−7.04%).

Among farmers maintaining their original livelihood type, per capita household carbon emissions decreased across all categories. The most notable reduction was among non-agriculture-dominated households (−23.34%), while the smallest decrease was among labor-dominated households (−6.43%).

4. Discussion

Research shows that the per capita household carbon emissions of farmers in the Dongning area of the Northeast Tiger and Leopard National Park show a downward trend, but there are differentiated carbon emission effects across different livelihood transformation paths. Specifically, farmers shifting from specialty farming/breeding to non-agriculture-dominated show the strongest emission reduction effect, with a 23.7% decrease in per capita carbon emissions. In contrast, farmers moving from non-agriculture-dominated to specialty farming/breeding see an 18.2% increase in per capita carbon emissions, making them the group with the most significant emission increase after transformation. This can be explained by the differences in production factor reliance across livelihood models: compared to black fungus cultivation and beekeeping, traditional agriculture-dominated farmers are less constrained by economic returns, market pressures, and risk avoidance, leading to lower usage rates of agricultural machinery, pesticide application intensity, and fertilizer input, thereby reducing carbon emissions. Traditional agriculture-dominated farmers usually have large-scale planting and breeding activities, requiring daily field management, which keeps their per capita production carbon emissions at a high level. In comparison, non-agriculture-dominated livelihood strategies lead farmers to integrate more into non-agricultural employment and market activities, potentially leading to the choice of high-carbon lifestyles. Meanwhile, livelihood strategy transformation can alter energy consumption structures, such as increased demand for gasoline and electricity due to the growth in household appliances and vehicles, directly intensifying farmers’ carbon emission pressures.

This study, through questionnaires and in-depth interviews with farmers in the Dongning area of the Northeast Tiger and Leopard National Park, deeply explores the low-carbon effects of farmers’ livelihood transformation. To accurately calculate the low-carbon effects of livelihood transformation among farmers in the Tiger and Leopard Park, this paper combines farmers’ recollections with account reviews to collect relevant data on living consumption and direct energy consumption for 2016 and 2023. Although using survey data from farmers helps ensure the continuity and comparability of research data, this method still has certain limitations and cannot completely avoid errors. For instance, during the survey period, the prices of living and direct energy consumption required by farmers are easily affected by weather, market supply and demand, exchange rate fluctuations, and government policies, leading to price volatility. Additionally, the surveyed farmers in the study area are mainly aged 55–65 and generally have low education levels, making them more prone to biases during the data recollection process, which may affect the accuracy of carbon emission calculations. Due to data biases and limitations, along with potential endogeneity among variables, this study cannot rigorously establish a causal relationship between livelihood transitions and carbon emissions. Future research will delve deeper into the key drivers influencing the carbon emission effects of farmers’ livelihood transitions to provide a solid scientific basis for targeted mitigation policies. Meanwhile, by clarifying the quantity and structure of farmers’ food consumption, we will develop a more comprehensive and refined carbon accounting framework for rural households.

5. Conclusions

Since the construction of the Northeast Tiger and Leopard National Park, the livelihood capital of farmers within and around the park has improved, with better livelihood strategies and changing consumption structures, all of which have impacted household carbon emissions. Using the IPCC’s emission inventory method and survey data from 2016 to 2023, this paper comprehensively assesses the carbon emissions from the livelihood transformation of farmers in the park, including both production- and consumption-related emissions. The following conclusions are drawn:

From 2016 to 2023, the overall per capita household carbon emissions of farmers in the park showed a downward trend. Specifically, their per capita carbon emissions from living consumption decreased, while production-related emissions slightly increased. However, all farmer groups, regardless of their livelihood strategy type, saw a decline in per capita household carbon emissions.

Among farmers who transformed their livelihood strategies, those who shifted to traditional agriculture-dominated or specialty farming/breeding from traditional non-agriculture-dominated or non-agriculture-dominated showed an upward trend in per capita household carbon emissions. Conversely, other transformed farmers and those who maintained their original livelihood strategies experienced a decline in per capita household carbon emissions.

The livelihood transformation of farming households in the Dongning section of the Northeast Tiger and Leopard National Park influences both their per-capita production-related and living-related carbon emissions; going forward, for production emissions, farmers should be encouraged to adjust agricultural production, operation, and management models, strengthen field management, and reduce unnecessary inputs of chemical fertilizers, pesticides, and plastic films, while guiding low-intensity activities such as under-forest planting and breeding in shallow mountainous areas and simultaneously advancing pilot programs for centralized cattle farming with ecological impact assessments; for living emissions, rural “green publicity” should be intensified to encourage participation in events such as “World Low-Carbon Day” to foster healthy, low-carbon lifestyles and consumption habits, while developing rural clean-energy infrastructure and exploiting locally available renewable green-energy resources to promote household adoption of solar, wind, and other clean-energy sources, thereby reducing dependence on fuelwood, crop residues, and fossil fuels and lowering household living-related carbon emissions.