Spatiotemporal Patterns and Zoning-Based Compensation Mechanisms for Land-Use-Driven Carbon Emissions Towards Sustainable Development: County-Level Evidence from Shaanxi Province, China

Abstract

Under the global climate governance framework, advancing China’s “Dual Carbon” goals within the context of sustainable development requires detailed, micro-level research. While existing studies predominantly focus on national or provincial macro scales, there remains a critical gap in county-level analyses that account for regional heterogeneity—particularly in geographically and economically transitional provinces like Shaanxi. This study focuses on 107 counties in Shaanxi Province, using land-use data from 2000 to 2022 to construct carbon emission and carbon compensation accounting models. We measure horizontal carbon compensation standards, examine spatiotemporal patterns of carbon emissions, delineate compensation zones, and propose regional low-carbon development strategies to inform sustainable development planning. The results show the following: (1) They reveal a steady increase in CO₂ emissions over the period (from 940 million tons in 2000 to 2.089 billion tons in 2022), highlighting an ongoing challenge for sustainability, with a spatial pattern of “high in the north, low in the south, and outward expansion from the center.” (2) In 2022, carbon payments across the province totaled CNY 1.068 billion, while compensation reached CNY 670 million, with significant spatial heterogeneity: 87 counties identified as payers (66 heavy) and 20 as receivers (17 heavy). (3) By integrating the Economic Contribution Coefficient, Ecological Support Coefficient, and Carbon Offset Rate with Major Function-oriented Zoning, we classify the counties into 12 carbon compensation subregions and recommend gradient-based development strategies. This refined zoning framework provides a clear operational framework for formulating differentiated low-carbon land-use optimization strategies and regional carbon compensation policies tailored to the characteristics of different functional zones. The research findings offer differentiated compensation standards and low-carbon land-use planning guidelines to support Shaanxi Province’s transition towards sustainable development, serving as a reference for carbon governance and sustainable development practices in China’s provinces with transitional geographical features and promoting the realization of China’s “Dual Carbon” targets as integral components of national sustainable development.

1. Introduction

Climate change, as a central challenge in global environment governance, has been paid close attention to by many scholars. The IPCC Sixth Assessment Report noted that global surface temperatures have risen by 1.09 °C compared to pre-industrial levels [1,2]. Among greenhouse gases, CO₂ stands out due to its substantial emissions and prolonged atmospheric retention, making carbon emission reduction a critical pathway to mitigate global warming [3,4]. China emits nearly 10 billion CO₂ every year, accounting for more than a quarter of global emissions [5]. At the same time, its terrestrial area constitutes 6.4% of the world’s total landmass, positioning it as a pivotal region in global and regional carbon cycles [6]. As a major carbon emitter, China put forward the “Dual Carbon” goal in 2020 [7] and issued a series of important policy documents emphasizing that carbon will become an important part of building ecological civilization in order to promote green development. Notably, in the field of climate change research and carbon emission dynamics, carbon emissions caused by land-use change occupy a prominent position, second only to carbon emissions caused by fossil fuel combustion and utilization [8,9,10]. Therefore, investigating the spatiotemporal heterogeneity in carbon emissions through the lens of land-use dynamics provides significant practical implications for informing regional mitigation strategies and advancing low-carbon development pathways.

Contemporary research has progressively unraveled the multidimensional complexity of land-use carbon dynamics. Initial efforts focused on spatiotemporal pattern identification through carbon coefficient methodologies enhanced by Gini coefficient and Moran’s I analyses [11,12,13], establishing foundational frameworks at national [14,15,16,17], provincial, urban agglomeration [18,19,20], and river basin [21,22,23,24] scales. Subsequent investigations employed decomposition models (LMDI) and spatial regression techniques (GWR) to disentangle the interacting effects of energy transitions, economic restructuring, and urban expansion [25,26,27]. Parallel advancements in predictive modeling adapted grey system theory for emission trajectory projections [28,29,30], while compensation mechanism studies integrated comparative advantage indices with spatial optimization protocols [31,32,33]. Methodological syntheses now converge on hybrid approaches combining IPCC standards with nighttime light remote sensing (DMSP-OLS/NPP-VIIRS) [34,35,36] and socio-economic decoupling indicators [37,38,39,40]. However, persistent gaps remain in subnational analyses, particularly at China’s county level, where data granularity constraints and spatial heterogeneity hinder the operationalization of macro-scale climate policies. The evolution of land-use carbon dynamics modeling has traversed three distinct theoretical paradigms. The first-generation static models (1980s–early 21st century), epitomized by the IPCC Tier 1 methodology, relied on static carbon stock assumptions with limited spatial differentiation, as evidenced in early national-scale inventory compilation practices. The second-generation dynamic coupled models (2010s) pioneered the integration of feedback mechanisms between land use and carbon cycles, with notable examples including the Global Change Assessment Model (GCAM) [41], coupling economic drivers with biophysical processes; and the CLUE model series [42], embedding land-use competition algorithms. The third-generation high-resolution models (2020s), exemplified by LUCCA [43] as the technological benchmark, emphasize multi-scale data integration through the fusion of remote sensing phenological characteristics with agent-based decision modeling. While these advancements have significantly enhanced spatiotemporal simulation accuracy, critical theoretical gaps persist in reconciling top-down policy constraints (such as main function divisions) with bottom-up socio-ecological feedback mechanisms—a contradiction particularly pronounced in China’s county-level governance contexts.

As the core node of the ecological barrier of the Silk Road Economic Belt and the Yellow River Basin, Shaanxi Province bears the synergistic pressure of the “Western Development Strategy” and the “Dual Carbon” goals. Concurrently, as China’s third-largest coal-producing province (accounting for 16.5% of the national raw coal output in 2022) and the largest oil/gas producer, Shaanxi Province constitutes a critical energy consumption and carbon emission hub. Its total carbon emissions carry significant implications for the national realization of the “Dual Carbon” goals (carbon peaking and carbon neutrality). From the perspective of Major Function-oriented Zoning (MFZ), the differentiated development path of “ecological conservation, agricultural leading, industrial agglomeration” creates gradient differentiation in energy structure, industrial structure, and carbon emission intensity across Shaanxi Province at the county level. Meanwhile, county-level units—serving as primary drivers of local economic growth, foundational units for policy implementation, and capillary networks for social governance—provide a critical lens to uncover intraregional heterogeneity and interregional interactions. Investigating these units enables the design of spatially tailored, scientifically grounded, and actionable strategies for low-carbon land-use optimization. Such efforts are vital for refining regional carbon compensation mechanisms and advancing both emission reduction and carbon sequestration.

This study addresses the micro–macro divide by focusing on Shaanxi Province’s 107 districts and counties through three methodological innovations: first, we constructed a county-level carbon budget accounting framework spanning 2000–2022, integrating IPCC emission coefficients with multi-source geospatial data (including land-use changes and NPP-VIIRS nighttime light imagery) to enable high-resolution spatiotemporal analysis. Second, we incorporated MFZ constraints into carbon compensation zoning criteria, effectively addressing spatial heterogeneities often oversimplified in provincial-level studies. Third, we propose spatially adaptive governance strategies aligned with functional zone characteristics—prioritizing ecological restoration in southern forest ecosystems, optimizing energy structures in northern coal-resource regions, and enhancing intensive land-use management in central urban agglomerations. Through these objectives, this research aims to bridge the gap between macro-level climate policies and micro-level implementation complexities at the county scale. These place-based solutions aim to balance development equity tensions in Shaanxi while offering scalable governance models for regional carbon mitigation under China’s “Dual Carbon” framework.

2. Materials and Methods

2.1. Study Area

Situated in China’s inland region (105°29′–111°15′ E, 31°42′–39°35′ N, Figure 1), Shaanxi Province exhibits a distinctive north–south elongation with east–west width contraction, encompassing 205,600 km² of geographically strategic territory. This province serves as a critical nexus between the New Eurasian Continental Bridge and multiple economic regions, including Northwest, Southwest, North, and Central China. The province demonstrates complex topography partitioned into three geomorphic zones: the Loess Plateau in the north, Guanzhong Plain in the central region, and Qinba Mountainous Area in the south. This longitudinal terrain gradient correlates with distinct climatic variations, transitioning from temperate continental conditions in the north (characterized by significant thermal–aridity contrasts) to temperate monsoon regimes in the central plains, and finally to subtropical humid climates in the south, with annual precipitation ranging from 340 to 1240 mm. As China’s third-largest coal producer and leading oil/gas extraction hub, Shaanxi simultaneously constitutes a major energy consumption and carbon emission center. This dual status as both an energy supplier and consumer creates unique developmental challenges. By 2022, Shaanxi Province had registered a population of 39.56 million and achieved a Gross Domestic Product (GDP) of CNY 3277.268 billion. Its raw coal output reached 746 million tons with a year-on-year growth of +5.4%, while crude oil production stood at 25.3662 million tons and natural gas output reached 30.711 billion cubic meters, reflecting year-on-year changes of −0.6% and +4.4%, respectively. Shaanxi Province has established a multi-level carbon emission management system to support China’s national “Dual Carbon” goals. At the provincial level, policies such as the “Implementation Plan for Carbon Peak in Shaanxi Province” have been introduced, setting targets for reducing the carbon emission intensity. These targets and requirements have been cascaded to the district and county levels, incorporated into local government performance evaluations, and directly influence industrial access, land use, and energy decision-making. This differentiated target decomposition combined with Major Function-oriented Zoning (MFZ)-based regulatory policies serves as a critical institutional foundation underpinning this research.

The province’s economic disparities, exacerbated by regional variations in terrain, climate, and resource endowments, are compounded by carbon emission intensities that pose significant barriers to sustainable development. Current environmental pressures, particularly elevated emission levels, necessitate urgent structural transformation in energy systems. Balancing economic growth imperatives with ecological conservation goals requires innovative pathways toward low-carbon transition. Strategic priorities include optimizing the energy mix, enhancing carbon sequestration capacities through ecological restoration, and implementing spatially differentiated mitigation policies that account for the province’s diverse geomorphic and climatic conditions.

2.2. Data Source

This study integrated three distinct data modalities to support analytical objectives: (1) terrestrial land-cover records. These data are sourced from the 30 m annual land-cover data released by the team led by Yang Jie at Wuhan University [44]. Utilizing ArcGIS geospatial processing modules, original land-cover classifications underwent aggregation into six operational categories: cultivated land, grass land, forest land, water, unutilized land, and construction land. (2) Nighttime light data: calibrated and integrated DMSP-OLS and NPP-VIIRS datasets by Wu et al. [45]. (3) Socioeconomic statistics: comprehensive statistical records were compiled from multiple official sources including the Shaanxi Provincial Statistical Yearbook, National Energy Statistical Yearbook, and China City Statistical Yearbooks. Data gaps were supplemented through systematic retrieval from municipal statistical bulletins and provincial government portals, with missing values imputed using provincial-level growth rate projections where authorized sources were unavailable.

2.3. Research Methods

2.3.1. Spatiotemporal Change Identification of Land Use

Land-use transition matrices provide a quantitative framework for analyzing temporal changes in land-cover compositions, capturing directional conversion patterns between discrete land-use categories across specified temporal intervals.

$$S_{\mathrm{ij}}=\left\{\begin{array}{cccc}{S}_{11}& S_{12}& \cdots & S_{1n}\\ S_{21}& S_{22}& \cdots & S_{2n}\\ \cdots & \cdots & \cdots & \cdots \\ S_{n1}& S_{n2}& \cdots & S_{nn}\end{array}\right\}$$

(1)

In the formula, the term S_ij quantifies the spatial extent of land-cover transition from class i to class j across the temporal domain of investigation. The parameter n, which equals six in this study, corresponds to the total number of land-use categories delineated according to the National Land Use Classification Standard.

Land-use transition dynamics serve as critical indicators for quantifying temporal variations in land-cover configurations, capturing both the magnitude and velocity of land-use category conversions within defined geographic boundaries. While single-category dynamic indices measure sector-specific transformation rates, the comprehensive land conversion index (LC) synthesizes these metrics to reflect the cumulative intensity of all land-use transitions occurring throughout the study area during the investigation period.

$$LC={\displaystyle \frac{{\sum }_{i=1}^n\left|U_i-U_j\right|}{{\sum }_{i=1}^n{U}_i}}\times {\displaystyle \frac{1}{T}}\times 100\%$$

(2)

In the mathematical formulation, U_i and U_j quantify the spatial extent of land-use category i at the initial temporal endpoint and category j at the terminal temporal endpoint of the investigation, respectively. The parameter n was fixed at six in this study. The variable t denotes the total temporal interval spanning the observational period.

2.3.2. Calculation of Land-Use Carbon Budget

Carbon budget quantification for Shaanxi Province’s land-use systems was conducted using a hybrid methodology combining established emission factor protocols with nocturnal luminosity calibration. This approach integrated two complementary measurement frameworks. For emission quantification, a tiered emission inventory was developed using IPCC-compliant carbon emission coefficients, accounting for anthropogenic fluxes from construction land uses, biogenic respiration processes (human/livestock metabolism), and agricultural production activities. The carbon sink was modeled through biophysical accounting of forest ecosystems, water, grass land communities, and unutilized lands, incorporating crop growth dynamics within agricultural matrices. In this study, during the calculation of the carbon budget, it is assumed that the land-use types and the carbon emission coefficients of various energy sources remain stable throughout the research period and do not change significantly with time or region.

(1)

Carbon emissions measurement

Construction land carbon emissions were quantified through an energy-consumption-based accounting framework. This methodology incorporated eight primary energy vectors: coal, coke, crude oil, gasoline, kerosene, diesel fuel, natural gas, and electricity (

Table 1. Conversion coefficient of energy standard coal and carbon emission coefficient.

Type of Energy Source	Standard Coal Conversion Coefficient	Carbon Emission Factor
Coal (t)	0.7143	0.7559
Coke (t)	0.9714	0.8550
Crude oil (t)	1.4286	0.5857
Gasoline(t)	1.4714	0.5538
Kerosene (t)	1.4714	0.5714
Fuel oil (t)	1.429	0.619
Diesel (t)	1.4571	0.5921
Natural gas (m3)	1.7143	0.5042
Electric (kWh)	0.1229	0.2132

). These energy sources were standardized to coal equivalents using national conversion benchmarks, enabling cross-energy type comparison. Emission quantification was performed through mass balance calculations integrating energy-specific carbon emission coefficients, with the general computational formula expressed as follows:

$$E={\displaystyle \frac{44}{12}}\times \sum _{i=1}^9{E}_i\times f_i\times {\theta }_i$$

(3)

In the formula, E_i is the consumption of the i-th energy, f_i is the carbon emission coefficient of the i-th energy, and θ_i is the standard coal conversion coefficient of the i-th energy.

Because there are some limitations in the acquisition of energy consumption data in districts and counties, nighttime light is closely related to regional economic activities, and economic activities are significantly correlated with carbon emissions [46,47,48,49]. Therefore, this study attempts to use nighttime light data to invert carbon emissions in districts and counties of Shaanxi Province.

Measurement of human and livestock respiratory carbon emissions was conducted using the following equation:

$$F_P=\sum \left(P_n\times T_n\right)$$

(4)

In the formula, F_P is the carbon emissions of human and livestock respiration, P_n is the number of humans or livestock, and T_n is the carbon emission coefficient (

Table 2. Carbon emission coefficient table.

Type		Carbon Emission Coefficient	References
Human and livestock respiration	Human breathing	0.079 t/(person·a)	Qi C., Ning Z., et al. [47]
	Pig	0.075 t/(head·a)
	Cattle	0.796 t/(head·a)
	Goat	0.005 t/(head·a)
Agriculture production process	Chemical fertilizers	0.8956 kg/t	Hu, et al. [50]
	Pesticides	4.9341 kg/kg
	Agricultural machinery use	0.18 kg/kW−1	Zeng, Li, Wu, et al. [51]
	Cropland	16.47 kg/hm2

).

Agricultural production process emissions were calculated using an activity-based accounting framework that integrates four primary emission sources: fertilizer application, pesticide utilization, agricultural machinery operation, and crop cultivation cycles. The quantification methodology follows the general emission equation:

$$F_t=C_{f}A+A_{a}B+P_{z}C+S_{i}D$$

(5)

In the formula, F_t is the total carbon emission of agricultural activities; A, B, C, and D are the carbon emission coefficients of chemical fertilizer, agricultural, pesticide, and crop planting area, respectively (Table 2); C_f is the amount of chemical fertilizer used; A_a is the agricultural area; P_z is the pesticide area; and S_i is the crop planting area.

(2)

Carbon absorption calculation

Based on existing research, carbon absorption mainly refers to the carbon absorbed by various carbon sink lands, including forest land, grass land, water bodies, unutilized land, and crops during the growth process. The formula is as follows:

$$H_j=\sum P_i\times Q_i$$

(6)

In the formula, P_i is the area of the i-th type of land, Q_i is the carbon sink coefficient of the second type of land (

Table 3. Carbon emission factors for each land-use type.

Land-Use Type	Carbon Emission Factor/(hm2·a)	Reference
Forest land	−5.81	Fang [52], Tianqi R., et al. [53]
Grass land	−0.022	Han W., et al. [54]
Water	−0.253	Gao L., et al. [55]
Unutilized land	−0.005	Lai [56], Ma S., et al. [57]

), and i indicates that the four types of land are forest land, grass land, water area, and unutilized land.

Crop carbon sink calculation was carried out using the following equation:

$$H_c=\sum _{i=1}^n{Z}_i\times \left(1-P_i\right)\times {\displaystyle \frac{Y_i}{H_i}}$$

(7)

In the formula, Z_i is the carbon absorption rate of the i-th crop, P_i is the moisture content of the i-th crop, Y_i is the yield of the i-th crop(t), and H_i is the economic coefficient of the i-th crop (

Table 4. The economic coefficients, moisture contents, and carbon absorption rates of crops.

Crop Type	Economic Coefficient	Moisture Content (%)	Carbon Absorption Rate	Crop Type	Economic Coefficient	Moisture Content (%)	Carbon Absorption Rate
Rice	0.4	12	0.45	Peanuts	0.43	10	0.45
Wheat	0.4	12	0.45	Rapeseed	0.25	10	0.45
Other cereals	0.4	12	0.45	Sesame	0.43	10	0.45
Legumes	0.34	13	0.45	Corn	0.4	13	0.471
Tubers	0.7	70	0.423	Vegetables	0.6	90	0.45
Cotton	0.1	8	0.45	Melons	0.7	90	0.45

).

2.3.3. Calculation of Economic Contribution Coefficient and Ecological Support Coefficient

The Economic Contribution Coefficient (ECC) serves as a normalized indicator to quantify the economic efficiency of regional carbon emissions. This dimensionless metric evaluates the marginal economic output generated per unit of carbon emitted, enabling cross-regional comparisons of emission productivity. The coefficient is calculated through a ratio analysis framework:

$$ECC={\displaystyle \frac{G_i}{G}}/{\displaystyle \frac{C_i}{C}}$$

(8)

In the formula, G_i and G are the GDP and total GDP of county i, and C_i and C are the carbon emissions and total carbon emissions of county i. When ECC values exceed unity (ECC > 1), it indicates operation within the “efficiency-dominant regime”, where regional carbon emissions exhibit positive economic productivity—each unit of emitted CO₂ generates greater marginal economic value. Conversely, ECC values below unity (ECC < 1) signify operation in the “inefficiency regime” characterized by diminishing economic returns per unit of carbon emission, suggesting potential decoupling between economic growth and emission intensity.

The Ecological Support Coefficient (ESC) index serves as a normalized bio-capacity indicator that quantifies regional carbon sequestration potential relative to emission pressures. This dimensionless metric integrates two critical dimensions:

$$ESC={\displaystyle \frac{C{A}_i}{CA}}/{\displaystyle \frac{C_i}{C}}$$

(9)

In the formula, CA_i and CA are county i and the total amount of carbon absorption. If ESC > 1, it indicates that the county’s carbon absorption capacity is higher, and vice versa.

2.3.4. Carbon Compensation Partition

Carbon compensation mechanisms represent a policy instrument for internalizing environmental externalities through cross-sectoral resource transfers. These schemes operationalize the “polluter pays principle” (PPP) and “beneficiary pays principle” (BPP) by requiring carbon-emitting entities to compensate sequestration agents via fiscal transfers or in-kind contributions. Firstly, the carbon compensation base value is adjusted, and the carbon emission threshold is set for each region to reduce the difference between regions. The formula is as follows:

$$P_i=ECC\times D_i$$

(10)

In the formula, P_i is the carbon emission threshold of each county, and D_i is the average carbon emissions of each county.

Secondly, to account for temporal and spatial variations in carbon emission intensity, historical data from 2000, 2005, 2010, and 2015 at both county and provincial levels were used to calibrate county-level emissions for 2005–2022. Taking the calculation of 2022 carbon compensation values as an example, the adjustment formula is as follows:

$$C_i^1=C_i\times \left(1+{\displaystyle \frac{G_{t1-i}}{G_{t2-i}}}-{\displaystyle \frac{G_{T1}}{G_{T2}}}\right)\times {\displaystyle \frac{G_{t1-i}}{G_T}}$$

(11)

In the formula, C_i¹ is the corrected carbon emissions (10⁴ t); C_i is the total carbon emissions of county i; G_t1−i and G_t2−i are the carbon emission intensities of county i in 2022 and 2015, respectively (t/CNY 10⁴); G_T1 and G_T2 are the carbon emission intensities of the whole province in 2022 and 2015, respectively (t/CNY 10⁴); and G_T is the average carbon emission intensity (t/CNY 10⁴) of the province’s counties in 2022.

Finally, the corrected carbon compensation reference value is closer to the real situation. The formula is as follows:

$$L_i^1=C_i^1-C_i-P_i$$

(12)

Referring to the previous research methods [58], the method to calculate the carbon compensation value can be obtained. The formula is as follows:

$$V_i=\left|L_i^1\right|\times \varphi \times \omega$$

(13)

In the formula, V_i is the carbon compensation fund (CNY 10⁴) that needs to be paid or obtained by the i-th district and county; φ is the price of unit carbon (CNY 10⁴/10⁴ t); and ω is the carbon compensation coefficient of the i-th county.

$$\phi ={\displaystyle \frac{C_{max}+C_{min}}{2}}\times {\displaystyle \frac{G_1}{G_2}}$$

(14)

Within the formula, C_max and C_min denote the maximum and minimum thresholds, respectively, for domestic carbon sequestration pricing (CNY 10⁴/10⁴ t). Utilizing a volumetric conversion methodology, Zhang et al. [59] determined the optimal carbon sequestration price range for China to be USD 10.11–15.17. Subsequent currency conversion, employing the 2022 exchange rate of CNY 6.9746 per USD as published by the China Foreign Exchange Trading Center, yielded an equivalent price range of CNY 70.51–105.80. Furthermore, G₁ corresponds to the per capita GDP of Shaanxi Province in 2022, while G₂ represents the national per capita GDP for the same period, both quantified in units of CNY 10⁴.

The carbon compensation capabilities vary across districts and counties due to disparities in their economic development levels. Moreover, the actual payment capacity for regional carbon compensation not only depends on aggregate economic output but is also influenced by the stage and structure of economic development. The Pearl growth curve effectively characterizes the nonlinear evolutionary patterns of payment capacity during regional development processes. To address this, an improved Pearl growth curve model [60] is adopted herein to avoid biases arising from simplistic linear proportionality. The formulation is expressed as follows:

$${\omega }_i=A_i/\left(1+a{e}^{-bt}\right)$$

(15)

In the formula, ω_i is the carbon compensation coefficient of the i-th region; A_i is the carbon compensation capacity of the i-th region, that is, the ratio of GDP of the i-th region to the total GDP of Shaanxi Province; a and b are constants here take 1; and t is the Engel coefficient of Shaanxi Province in 2022.

3. Results

3.1. Spatiotemporal Dynamics of Land-Use Carbon Emissions

Land-use transitions in Shaanxi Province between 2000 and 2022 reveal significant ecological restructuring. Forest cover exhibited the most pronounced expansion, with a net gain of 10,663.36 km² (+12.64%) over the study period, and the Conversion of Cropland to Forest Program is the primary driving force for forest land growth. Concurrently, the cultivated land area contracted is 8537.73 km² (−14.04%). Under the constraints of the main functional zoning, the ecological barrier areas strictly restrict cultivated land development, leading to the transfer of cultivated land. Meanwhile, the abandonment of cultivated land and the rigid demand for construction land are also important reasons for the reduction in cultivated land area. Water systems demonstrated the most modest change, with a net increase of 118.86 km² (+21.05%), likely attributable to anthropogenic regulatory interventions and ecological restoration projects. These land-cover transformations directly impact regional carbon budgets through the following: forest expansion effects: increased biomass sequestration potential from matured plantations; agricultural contraction impacts: reduced emissions from fertilizer use and machinery operations. The observed land-use changes demonstrate complex interactions between ecological restoration efforts, demographic pressures, and infrastructure development, with cascading effects on provincial carbon neutrality objectives.

The spatial configuration and evolutionary dynamics of land use in Shaanxi Province are systematically illustrated in Figure 2. Dominated by forest and grass land ecosystems, the provincial land cover exhibits a hierarchical structure where water areas, unutilized land, and construction land occupy relatively marginal proportions. Spatial analysis reveals distinct ecological zonation patterns: forest and grass land formations predominantly occupy the Loess Plateau and Qinba Mountain Ecological Conservation Zone in northern Shaanxi, maintaining contiguous spatial distributions that contribute to regional ecological connectivity. In contrast, cultivated land and construction areas demonstrate pronounced concentration in the Guanzhong Plain, reflecting agricultural intensification and human settlement patterns in this economically active region. Temporal dynamics between 2000 and 2022 indicate significant land-use transitions. Cultivated land area contracted substantially while construction land expanded markedly, with the latter’s growth trajectory indicative of accelerated urbanization processes during the study period. Notably, unutilized land experienced a progressive reduction, undergoing gradual conversion to grass land cover—a transformation directly attributable to sustained ecological restoration initiatives implemented in the Loess Plateau region. These land-use changes occurred within a broader context of evolving carbon cycling patterns, as evidenced by the findings that construction land expansion correlates with increased carbon emissions, while forest areas serve as critical carbon sinks.

Analysis of land-use transitions depicted in Figure 3a,c indicates moderate spatial dynamics across Shaanxi Province, with only 17.7% of the total land area experiencing land-cover category changes while the majority maintain stability. The most substantial transformations involved cultivated land conversions to forest and grass land ecosystems, accounting for 7457.97 km² and 7427.48 km², respectively. Notable bi-directional conversions were also observed between grass land and forest cover, while transitions from construction land/forest to bare land remained minimal (0.47 km² and 0.06 km²). Spatial differentiation in land-use intensity is evident from Figure 3b, with Weiyang, Yanta, Qindu, Baqiao, Gaoling, Yangling, and Weicheng districts exhibiting the highest comprehensive land-use dynamic degrees (>2%) over the 22-year period. These regions, particularly Weiyang (4.71%) and Yanta (4.57%), demonstrate pronounced land-use changes corresponding to their status as economic hubs with rapid urban expansion. Temporal analysis reveals cultivated land reductions of 123.86 km² and 69.07 km², respectively, in these districts, paralleled by construction land increases of 121.13 km² and 68.52 km². Conversely, Ningshan County and Beilin District exhibit the lowest dynamic indices (0.12% and 0.11%), indicative of stable land-use regimes. In these areas, cultivated land only decreased by 12.78 km² and 0.24 km², respectively, while construction land expansion was limited to 3.08 km² and 0.25 km². The predominant land change in Ningshan involved forest expansion (41.19 km²), suggesting ecological restoration impacts in less-developed regions. This spatial heterogeneity in land-use dynamics reflects regional disparities in economic development intensity and ecological conservation priorities across the province.

The carbon sources, carbon sinks, and net carbon emissions in Shaanxi Province from 2000 to 2022 were quantified using established methodologies, with the detailed outcomes presented in

Table 5. Carbon emissions of various types of land in the study area. Unit: 1×10⁴ t.

Year	Carbon Sink						Carbon Source			Net Carbon Emissions
	Cultivated Land	Forest Land	Grass Land	Water	Unutilized Land	Total Carbon Absorption	Cultivated Land	Construction Land	Total Carbon Emissions
2000	−1287.3	−4901.3	−12.06	−1.43	−0.1264	−6202.23	17.49	93,421.34	94,031.65	87,829.43
2005	−1352.5	−5014.9	−12.24	−1.56	−0.083	−6381.28	14.66	108,553.80	109,120.41	102,739.13
2010	−1630.7	−5184.1	−12.28	−1.7	−0.034	−6828.82	14.70	124,763.44	125,330.35	118,501.52
2015	−1502.6	−5317.1	−12.36	−1.69	−0.0109	−6833.75	13.88	155,032.86	155,669.13	148,835.38
2020	−1399.7	−5435.4	−12.22	−1.78	−0.0092	−6849.12	12.84	184,051.49	184,549.74	177,700.63
2022	−1467.4	−5520.8	−11.43	−1.79	−0.0107	−7001.36	12.91	208,300.80	208,857.66	201,855.90

. Over the two-decade period, the province experienced a consistent escalation in both gross carbon emissions and carbon sequestration. Specifically, total carbon emissions exhibited an upward trajectory, increasing from approximately 9.4 × 10⁸ t to 20.89 × 10⁸ t. Concurrently, carbon absorption also demonstrated a gradual rise, ranging from approximately 0.62 × 10⁸ t to 0.7 × 10⁸ t. This consistent growth pattern underscores the province’s evolving carbon dynamics over the observed period.

Temporal analysis of the land-use-associated carbon budget in Shaanxi Province (2000–2022) reveals distinct source–sink dynamics. Construction land consistently serves as the dominant carbon emission source, contributing over 99% of total anthropogenic emissions throughout the study period. This finding aligns with the previously observed land-use transition patterns, where rapid urban expansion correlates with increased fossil fuel consumption and industrial activities. Conversely, terrestrial ecosystems demonstrate significant carbon sequestration capacity, with forest lands accounting for more than 76% of total carbon sinks. Cultivated lands constitute the secondary carbon sink category, though their sequestration performance exhibits a non-linear trajectory characterized by initial growth followed by subsequent decline over the 22-year interval. This biphasic trend in agricultural carbon sequestration may reflect evolving land management practices, climate variability impacts, or the combined effects of land-use intensification and ecological restoration initiatives. The initial phase of the study witnessed the promotion of agricultural technologies and the retirement of inefficient farmland through land conversion, which enhanced the efficiency of remaining cultivated land and contributed to carbon sink growth. Subsequently, carbon sequestration declined due to urbanization encroaching on prime agricultural land and partial farmland abandonment. Meanwhile, prolonged climate warming exacerbated soil respiration processes, thereby reducing the carbon sequestration capacity. This dual-phase evolution of agricultural carbon sequestration reflects the combined impacts of evolving land management strategies, climate variability, and the intricate interplay between land-use intensification and ecological restoration efforts. The pronounced dominance of construction land in emission profiles underscores the critical need for integrating low-carbon urban development strategies into regional spatial planning frameworks.

Referring to the relevant research and combining with the natural breakpoint method, the carbon emissions of land use in Shaanxi Province were divided into low-carbon-emission area (≤0.09 × 10⁸ t), medium-and-low-carbon-emission area (0.09 × 10⁸~0.1 × 10⁸ t), medium-carbon-emission area (0.1 × 10⁸~0.2 × 10⁸ t), high-carbon-emission area (0.2 × 10⁸~0.3 × 10⁸ t), and ultra-high-carbon-emission area (≥0.3 × 10⁸ t).

Spatial analysis of land-use carbon emissions in Shaanxi Province (2000–2022) (Figure 4) reveals pronounced north–south differentiation, forming a distinct gradient of high emissions in the northern regions transitioning to low emissions in the southern areas. In 2000, the provincial carbon emission landscape comprised 91 low-emission zones, 7 moderate–low zones, 9 moderate zones, and 1 high-emission zone, with Yanta District registering the highest emission intensity at 0.2 × 10⁸ tons. Between 2005 and 2020, all administrative units exhibited steady emission growth, with marked spatial reconfiguration observed by 2020: low-emission zones contracted to 30 units, while moderate–low (13), moderate (42), and high-emission (11) zones expanded accordingly. Notably, eleven jurisdictions including Changan, Yanta, Baqiao, and Weiyang districts transitioned into ultra-high-emission categories during this period. By 2022, this polarization intensified with 26 low-emission zones remaining, accompanied by 9 moderate–low, 46 moderate, 9 high, and 17 ultra-high-emission zones. The progressive expansion of ultra-high-emission areas underscores the growing carbon intensity associated with urbanization and industrial development in key economic corridors, particularly in the Guanzhong Plain urban agglomeration. This spatial–temporal evolution pattern necessitates targeted mitigation strategies that address regional disparities in emission profiles while promoting low-carbon transitions in high-intensity zones.

3.2. Multidimensional Analysis of Land-Use Carbon Emission Effects

3.2.1. Economic Contribution Coefficient Dynamics

As depicted in Figure 5, the Economic Contribution Coefficient (ECC) of land-use carbon emissions across counties and districts in Shaanxi Province ranged from 0.07 to 3.66 between 2000 and 2022, underscoring substantial regional disparities in carbon emission intensities and economic output efficiencies. Over this period, the average ECC value increased from 0.73 to 0.87, indicating an elevation in the economic value generated per unit of carbon emissions. Specifically, from 2000 to 2005, there was a notable increase in the number of counties and districts with an ECC exceeding 1. Among these, Beilin District, Weibin District, and Jintai District exhibited particularly high ECC values for land-use carbon emissions, signaling a strong dependency of their economic development on energy consumption. Consequently, carbon emissions in these districts made a more significant contribution to and exerted a greater impact on economic growth. During the subsequent phase from 2005 to 2010, the number of counties and districts with an ECC above 1 remained relatively stable. By 2022, however, only 27 counties and districts retained an ECC greater than 1, with Yanta District, Lianhu District, Fugu County, Shenmu City, and Beilin District surpassing an ECC of 2. These areas continued to demonstrate a substantial economic drive from carbon emissions, albeit predominantly concentrated within advanced industrial hubs and urban centers. Overall, regions with elevated ECC values remain concentrated in economically developed and industrially robust areas, particularly Yanta District, Beilin District, Shenmu City, and Fugu County, where high resource utilization efficiencies translate into significant economic benefits.

3.2.2. Analysis of Ecological Support Coefficient

Temporal assessment of the land-use ecological support coefficient in Shaanxi Province (2000–2022) reveals relatively stable equilibrium conditions characterized by a distinct south–north gradient in carbon emission resilience (Figure 6). Elevated ecological carrying coefficients (>2) were concentrated in southern ecological conservation zones, particularly in Huyi District, Zhouzhi County, Lantian County, Huanglong County, Fuxian County, Taibai County, Xunyi County, and Fengxian County. Notably, Zhouzhi County maintained an exceptionally high carrying capacity (ESC > 6) throughout the study period, indicating robust ecosystem resilience to anthropogenic carbon emissions. While the majority of counties exhibited moderate equilibrium states (ESC ≈ 1), suggesting approximate parity between emission levels and ecological sequestration capacity, certain urbanized regions demonstrated persistent vulnerabilities. Specifically, Yanta District, Baqiao District, Weiyang District, Lianhu District, Xincheng District, Ansai District, Changwu County, and Shenmu City consistently registered low ESC values, reflecting chronic pressure on local ecosystems from excessive carbon emissions relative to their absorption capacity. This spatial heterogeneity in ecological carrying capacity underscores the need for differentiated mitigation strategies that align with regional biophysical conditions and developmental trajectories.

3.2.3. Analysis of Carbon Emission Intensity, Carbon Offset Rate, and Net Carbon Emissions

Temporal analysis of carbon emission dynamics in Shaanxi Province reveals distinct spatiotemporal patterns across key indicators (Figure 7). Carbon emission intensity (t/CNY 10⁴), representing the carbon efficiency of economic output, exhibits a consistent upward trajectory in all 107 administrative units during 2000–2022. This trend indicates increasing reliance on carbon-intensive energy sources and production methods, particularly evident in central and northeastern regions where industrialization levels are relatively high. Conversely, northwestern and eastern areas demonstrate comparatively lower emission intensities, reflecting differences in economic structure and technological advancement.

The Carbon Offset Rate (COR), defined as the ratio of carbon sequestration to emissions, demonstrates significant spatiotemporal variability across the province. Temporal analysis reveals a non-linear trend with initial decline followed by recovery in most regions, suggesting complex interactions between ecological restoration efforts and economic development pressures. Spatially, COR values exhibit pronounced south–north differentiation, with higher ratios in southern ecological conservation zones due to greater forest cover and carbon sequestration capacity. Notably, Zhouzhi County maintains exceptionally high COR values (>6) throughout the study period, reflecting robust ecosystem resilience.

Net carbon emissions, calculated as the difference between anthropogenic emissions and ecological sequestration, show alarming upward trends across all regions. Spatial differentiation is particularly pronounced in northern and eastern Shaanxi, where rapid industrialization and urban expansion have outpaced the carbon sequestration capacity. This spatial–temporal evolution pattern highlights the growing imbalance between carbon metabolism and ecological carrying capacity, necessitating targeted mitigation strategies that integrate low-carbon development pathways with regional ecological conservation objectives. The observed polarization in carbon performance metrics underscores the need for differentiated policy interventions across the province’s diverse ecological–economic zones.

3.3. Spatiotemporal Characteristics of Carbon Compensation Values

Spatiotemporal analysis of land-use carbon compensation values in Shaanxi Province (2005–2022) reveals evolving regional disparities in ecological–economic equilibrium (Figure 8). The results indicate progressive polarization in compensation dynamics: the number of compensated regions declined from 56 in 2005 to 20 in 2022, with severely compensated areas reducing from 46 to 17 during the study period. Concurrently, payment obligations expanded significantly, with the number of payment regions increasing from 51 in 2005 to 87 in 2022, including a marked rise in high-payment zones from 35 to 66. This dynamic reflects growing economic disparities and carbon metabolism imbalances across the province.

Persistent compensation obligations were observed in 11 administrative units (10.28% of total) throughout the study period, indicating structural dependencies on ecological compensation mechanisms. Spatial analysis reveals clear geographic partitioning: high-payment zones predominantly cluster in northern and central regions aligned with industrial corridors and urban agglomerations, while compensation benefits concentrate in southern ecological conservation zones. This pattern underscores the dual challenge of balancing economic development in industrialized regions with ecosystem service preservation in environmentally sensitive areas. The observed trends necessitate targeted policy interventions that integrate carbon pricing mechanisms with regional development strategies to achieve equitable carbon neutrality transitions.

As illustrated in Figure 9, the 2022 carbon compensation framework exhibits spatial variations across county-level jurisdictions in Shaanxi Province, with compensation values categorized into three tiers: heavy-payment area/compensation area (≥100 × CNY 104), moderate-payment area/compensation area (100 × 104~50 × CNY 104), and mild-payment area/compensation area (0~50 × CNY 104), and the payment pressure and compensation degree are expressed by the amount of payment or compensation accounted for the GDP of the county. There are 20 total compensation areas, the amount of compensation is CNY 670 million, and the average degree of compensation is 2.36. High-level-compensation zones, such as Yanta District and Beilin District, dominated the compensation landscape, with Yanta District receiving the highest compensation amount. Conversely, 87 payment zones were designated, the total payment amount is CNY 1.068 billion, and the average payment pressure is 4.35, with Weicheng District bearing the highest payment obligation. From the vantage of major functional zone differentiation, ecological compensation obligations exhibit spatial concentration patterns wherein key development zones constitute the primary compensation recipients, followed by primary agricultural production zones. This distribution stems from the accelerated industrialization and urbanization dynamics inherent to key development zones, which engender an elevated energy consumption intensity and pronounced structural dependency on fossil resources. Some ecological function areas have become carbon compensation payment areas such as Longxian County and Baihe County. Driven by factors such as human activities and tourism development, regional carbon emissions have increased. The compensated areas are mainly key development zones, accounting for 65%. On the whole, the different positioning of the main functional areas is closely related to the distribution of carbon compensation roles. For the main functions and carbon compensation status of different regions, the regional emission reduction and increase policy can be formulated in a targeted manner, and green low-carbon development is not pushed out.

3.4. Carbon Balance Partition

The Economic Contribution Coefficient (ECC) measures the disparity in regional carbon emissions from an economic perspective. Based on ECC assessment results, Yang et al. classified counties in the Yangtze River Economic Belt into five categories and formulated differentiated carbon compensation policies. The Ecological Support Coefficient (ESC) reflects the relative magnitude of the regional carbon sequestration capacity. A Pearson correlation test conducted on ESC and NDVI (Normalized Difference Vegetation Index) across Shaanxi Province’s districts and counties revealed a correlation coefficient of 0.78, indicating a significant positive correlation. This suggests that regions with higher vegetation cover exhibit correspondingly greater carbon sequestration capacity.

Based on three key indicators (ECC, ESC, and COR [61]), this study established a four-tier carbon balance zoning system comprising Low-Carbon Conservation Zones (LCCZs), Economic Development Zones (EDZs), Carbon Intensity Control Zones (CICZs), and High-Carbon Optimization Zones (HCOZs) (Figure 10). Spatially, LCCZs account for 8.4% of the total area, primarily distributed across nine districts and counties within the QinBa Mountain Ecological Barrier, such as Shanyang and Dali. This region exhibits notable carbon sink resource advantages and high potential for carbon sink development. EDZs occupy 34.6% of the study area, characterized by rapid economic growth, high carbon emission intensity, and low energy utilization efficiency. CICZs cover 16.8% of the territory, where carbon emission intensity is lower than in EDZs but demonstrates significant carbon lock-in effects during urbanization. HCOZs constitute 40.2% of the total area, featuring carbon emissions far exceeding carbon absorption, high-carbon-deficit coefficients, and substantial reliance on heavy chemical industries, reflecting entrenched high-carbon-development pathways.

Spatial coupling with MFZ (Figure 10) identified 12 composite functional units, revealing pronounced spatial heterogeneity. Within the Ecological Function Zones (EFZs), gradient characteristics emerged: EF-LCLZ includes four districts/counties with favorable ecological conditions and abundant carbon sink resources (ESC = 4.77, COR = 0.16). EF-EDZs encompass 27 districts/counties like Zhouzhi and Zhenan, where 31.16% of the land generates merely 9.2% of economic output and contributes 13.88% of total emissions. EF-CICZs are singularly represented by Shenmu City (3.66% of provincial land), driving 6.98% of economic growth while producing 3.38% of emissions (ECC = 0.07), indicating severe environmental stress. EF-HCOZs contain eight districts/counties including Huayin, contributing 5.69% of emissions with only 3.93% economic output, showing low ECC (0.63), ESC (0.6), and COR (0.02) values. Agricultural Production Zones (APZs) exhibit spatial differentiation: AP-LCCZs achieve economic–carbon synergy through eco-agriculture models (ESC = 2.6, ECC = 1.4). AP-EDZs serve as a premium agricultural hub with strong ecological carrying capacity (ESC = 3.29) but limited economic contribution (1.48%). AP-CICZs include two districts where 2.05% of land generates 1.6% of emissions, demonstrating moderate economic performance and weak ecological functions (ECC = 1.83, ESC = 0.45). AP-HCOZs span 13 districts, utilizing 11.99% of land to produce 9.64% of emissions with low economic efficiency (ECC = 0.65), highlighting the urgency for traditional agricultural transformation. Key Development Zones (KDZs) display internal disparities: KD-LCCZs comprise two counties with balanced economic growth (1.53% GDP share) and low emissions (1.23%), showing robust ecological–economic synergy (ECC = 1.2, ESC = 1.74). KD-EDZs include five districts with weak economies (ECC = 0.69) but relatively strong carbon sequestration (ESC = 3.03). KD-CICZs span 15 districts where 7.72% of land generates 37.77% of economic output and 23.22% of emissions, coupled with poor carbon sink capacity (ESC = 0.37). KD-HCOZs cover 22 districts, contributing 21.94% of GDP and 32.18% of emissions (ESC = 0.38), indicating severe environmental pressures. The disparities between natural and socioeconomic indicators across the compensation zones underscore the necessity for tailored optimization strategies aligned with specific functional zone characteristics.

4. Discussion

4.1. Spatiotemporal Differentiation Mechanisms of Land-Use Carbon Emissions

The quantitative study on land-use carbon emissions in 107 counties of Shaanxi Province from 2000 to 2022, based on the carbon emission coefficient method, reveals that the total provincial carbon emissions exhibit a sustained growth trend, aligning with the research findings of Gao and Liang Q [62,63]. This growth pattern demonstrates pronounced coupling with regional industrial structure characteristics, where the accelerated industrialization and urbanization processes have been accompanied by the expansion of energy-intensive industries, leading to increased carbon emissions due to substantial energy consumption. Spatially, the carbon emission distribution manifests a “north-high–south-low gradient pattern with diffusion from core areas outward”. High-value zones concentrate in the Guanzhong urban agglomeration and Yulin Energy and Chemical Industry Base, where carbon emission intensity shows a significant correlation with the proportion of secondary industries. Conversely, low-value zones predominantly occur in the Qinba Mountain region, characterized by higher forest coverage rates and enhanced carbon sequestration capacity, forming a natural carbon sequestration barrier. Policy implementation capacity also significantly influences decarbonization outcomes: in coal-dependent counties in northern Shaanxi (e.g., Shenmu and Fugu), identical energy transition policies yielded divergent results—Shenmu achieved an 11.2% reduction in coal’s energy share through rigorous enforcement of efficiency standards, while Fugu only attained a 3.8% reduction due to limited monitoring resources. Governance system reforms have produced demonstrable policy effects: following the 2018 provincial environmental vertical management reform that dismantled local protectionism, the CO₂ intensity reduction rate in Guanzhong urban agglomerations accelerated by 18.7% compared to the pre-reform period. Conversely, counties lacking cross-jurisdictional coordination mechanisms (e.g., the Hanzhong–Ankang border region) continue to experience stagnant carbon compensation efficiency.

4.2. Spatial Differentiation of Regional Carbon Compensation Values

The accounting results of carbon compensation values based on an improved carbon compensation model indicate significant spatial heterogeneity and temporal dynamics in the study area. On the one hand, county-level units consistently identified as compensation-receiving zones play a highly positive role in carbon balance. On the other hand, counties persistently categorized as compensation-paying zones face substantial carbon emission pressures, while the dynamic shifts in carbon compensation roles further reflect the complexities encountered by Shaanxi Province in addressing carbon emission reduction and advancing ecological civilization initiatives. Spatially, compensation-receiving zones are predominantly concentrated in central and southwestern regions, characterized by ecological advantages, agricultural and high-tech industrial dominance, and relatively low energy consumption. These zones receive financial support from compensation-paying zones under the carbon compensation framework. Conversely, compensation-paying zones are primarily distributed in northern and eastern Shaanxi, where secondary industries prevail, energy consumption is intensive, and economic foundations are robust, necessitating their role as financial contributors in carbon compensation mechanisms. The delineation of compensation-receiving and paying zones facilitates resource optimization and promotes green development alongside ecological civilization construction in Shaanxi Province.

4.3. Low-Carbon Zoning Governance Pathways Guided by Major Function Zones

By synthesizing previous studies, Yang et al.’s research on the Yangtze River Economic Belt, though conducted at the county level, did not integrate MFZs and employed horizontal fiscal transfer payments as the policy tool. Xia et al.’s study on the Beijing–Tianjin–Hebei region adopted a relatively broad geographical scale. Xu et al.’s analysis of the Yellow River Basin overlooked agricultural carbon emissions and carbon sequestration. In contrast, this study adopts an “ECC–ESC–COR” three-dimensional evaluation framework combined with MFZ principles to investigate policy implementation at the fundamental administrative unit level (district/county scale). Shaanxi Province is categorized into 12 differentiated sub-regions, and under the guidance of low-carbon development objectives, the following targeted optimization schemes are proposed: EF-LCCZs (e.g., Yangxian County), characterized by superior carbon sink foundations and high ecosystem service values, should implement a negative list system for high-energy-consuming industry access, explore carbon sink option trading market mechanisms, and transform ecological advantages into economic drivers through optimized spatial layouts for eco-tourism industries. EF-EDZs (e.g., Zhashui County), exhibiting low land-use efficiency and imbalanced ecological–economic systems, require a coordinated development model integrating “clean energy substitution with extended specialty forestry-fruit industry chains.” EF-CICZs (exclusively Shenmu City), with high economic density and acute environmental pressures, should establish a “coal-chemicals–CCUS–photovoltaics” coupled system, implement ecological restoration and photovoltaic reclamation in mining areas, and develop digital twin platforms for low-carbon urban management. EF-HCOZs, featuring fragile ecological foundations, must synergize ecological governance with industrial upgrading, advance soil–water conservation projects in loess hilly–gully regions, promote facility agriculture, and upgrade cultural–tourism–health industries. AP-LCCZs, with significant agricultural carbon sink potential, should innovate value realization mechanisms for agricultural carbon sink products, establish carbon-labeled agricultural product systems, and develop carbon sink trading platforms to monetize ecological assets. AP-EDZs, characterized by weak economic contributions but strong ecological carrying capacity, should extend specialty agricultural value chains and promote smart agriculture. AP-CICZs, with degraded ecological functions, need circular agricultural systems, water-efficient facilities, strict fertilizer/pesticide controls, and cropland quality enhancement initiatives. AP-HCOZs, dominated by inefficient agriculture with high emissions, should adopt carbon-sequestration farming practices, establish agricultural pollution monitoring systems, and implement precision emission reduction. KD-LCCZs, balancing economic and ecological performance with low carbon intensity, should strengthen ecological conservation, develop smart agricultural demonstration parks and cultural–ecotourism complexes, and increase renewable energy shares. KD-EDZs, with high carbon sink potential but lagging economies, should launch forest health carbon sink projects, advance digital economies, and enhance value-added chains for traditional Chinese medicinal herbs. KD-CICZs, marked by carbon-intensive industries and weak sequestration capacity, require deep industrial decarbonization, regional carbon market development, and expanded coverage of emission-controlled enterprises. KD-HCOZs, facing acute economy–environment conflicts, should implement systematic pollution–carbon reduction, conduct clean production audits in key sectors, build green transportation systems, and adopt tiered carbon tax policies with revenue reinvested in ecological restoration. The implementation phases are categorized based on the decarbonization urgency–technology maturity matrix as follows: near-term priorities: deploy measures with proven technology and short payback periods (e.g., clean production audits); mid-term deployment: scale up demonstration projects (e.g., CCUS and digital twin platforms); long-term breakthroughs: rely on technology cost reduction (e.g., photovoltaic hydrogen production) or institutional innovation (e.g., carbon sink securitization).

In summary, the carbon emission levels and structural patterns across regions exhibit significant heterogeneity, influenced by multiple factors including regional economic development, industrial structure, energy consumption patterns, and resource endowments. These complexities necessitate diverse and context-specific governance strategies. The pronounced disparities in carbon emissions among counties in Shaanxi Province highlight critical challenges in reconciling future economic growth with ecological civilization construction: (1) Carbon deficit and energy transition: Shaanxi Province faces a substantial carbon deficit, with land-use carbon emissions far exceeding carbon absorption. Energy consumption, particularly coal-dominated energy use, remains the primary carbon source, posing major challenges for emission reduction. Enhancing energy efficiency, strictly controlling energy consumption, and establishing a new energy system are pivotal for achieving green, low-carbon development. (2) Ecological protection and zoned governance: ecological functional zones: strengthen protection efforts to improve carbon sequestration capacity through ecological engineering initiatives to expand vegetation coverage. Planning afforestation to enhance carbon sequestration capacity. Carbon Intensity Control Zones and High-Carbon Optimization Zones: support industrial agglomeration, optimize energy structures, and accelerate clean energy development. Economic Development Zones: reduce reliance on energy-intensive industries while intensifying environmental protection measures and integrating carbon sink spaces such as rooftop greening or vertical forests to enhance carbon sequestration capacity. Low-Carbon Conservation Zones: promote optimized agricultural practices, advance smart agriculture, implement the “crop rotation–straw returning” carbon sequestration model, and rationalize land-use planning to minimize emissions. (3) The carbon transfer mechanism in Shaanxi Province needs to raise funds through multiple channels such as government finance, market transactions, international financing, and corporate capital and combine policy tools such as legal guarantees, economic incentives, technological innovation, and regional coordination. Specifically, the government can provide financial support through vertical transfer payments and horizontal regional compensation mechanisms, such as providing targeted support to carbon sink advantage areas through special ecological compensation funds; the market can raise funds through channels such as carbon trading markets and carbon sink financial products; at the same time, it can actively attract international climate financing and use carbon tax feedback and green financial instruments to attract the participation of enterprises and social capital. In terms of policy pathways, it is necessary to establish a legal framework and standard system, implement economic incentives and regulatory mechanisms, promote technological support and innovation-driven development, and strengthen regional coordination and public participation. Ultimately, a carbon reduction system with shared responsibility and mutual benefit will be constructed to achieve the benign interaction of low-carbon development and ecological protection in Shaanxi Province.

While this study explores the spatiotemporal patterns of carbon emissions and carbon balance characteristics in Shaanxi Province, several limitations remain: (1) Spatial scale sensitivity: regional carbon budgets and compensation mechanisms are highly sensitive to spatial scale variations. Larger spatial scales introduce greater complexity due to heterogeneous natural and socioeconomic conditions, potentially distorting carbon budget and compensation assessments. Additionally, the resolution of the land-use data employed may be insufficient to capture fine-grained micro-level changes within counties. Consequently, more detailed analyses at finer scales such as township or village levels are warranted. (2) While treating Shaanxi Province as an independent ecosystem, the study acknowledges its status as an open system in reality. The influence of external environmental factors on the province’s carbon budget warrants further investigation. (3) The carbon compensation valuation in this study relies on static models and does not account for market fluctuations. Future research should explore dynamic carbon compensation models. (4) Unaccounted emission sources: emissions from solid waste and wastewater treatment processes were not included in this study, potentially introducing biases in carbon budget estimates. Landfills and incineration generate carbon dioxide, with highly concentrated spatial distribution. The amount of carbon dioxide emitted by treatment plants is negatively correlated with the level of urbanization, requiring further in-depth investigation.

5. Conclusions and Implications

This study evaluates the spatiotemporal dynamics of carbon budget capacity across 107 county-level administrative units in Shaanxi Province from 2000 to 2022. Using a multi-model analytical framework that integrates land-use transition data, nighttime light remote sensing, and socioeconomic indicators, we reveal evolving disparities in carbon emissions and sequestration across regions. A horizontal interregional carbon compensation model was developed to delineate differentiated compensation zones and inform region-specific low-carbon development strategies.

(1)

The results confirm a continuous rise in carbon emissions over the study period, with construction land becoming the primary anthropogenic source. Forest ecosystems and cultivated land serve as the main carbon sinks. Spatial autocorrelation analysis identifies a pronounced north–south gradient in emissions, with high-emission clusters in the energy-industrial north and low-emission zones in the agricultural and forested south. Emission patterns demonstrate a centrifugal diffusion trend, radiating from central urban cores to peripheral areas.

(2)

As of 2022, Shaanxi Province exhibits relatively low carbon productivity and inefficient energy use. Only 25% of counties recorded an Economic Contribution Coefficient (ECC) above 1, while 43% had Ecological Support Coefficients (ESCs) greater than 1. ESC values also display centrifugal spatial diffusion. These regional imbalances call for tailored mitigation strategies: emission-intensive zones should improve land use and energy efficiency to boost output per unit of emissions, while ecologically sensitive areas must enhance carbon sequestration to reinforce resilience and contribute to carbon neutrality goals.

(3)

The interregional carbon compensation mechanism estimated CNY 1.068 billion in total payment obligations and CNY 670 million in compensation entitlements in 2022. Model simulations identified 87 counties as net payers—including 66 high-liability areas requiring fiscal intervention—and 20 as compensation recipients, with 17 in need of prioritized ecological investment. This mechanism offers a practical approach for facilitating horizontal carbon transfers consistent with China’s “Dual Carbon” strategy.

(4)

A four-tier carbon zoning system was constructed based on the ECC, ESC, and Carbon Offset Rate (COR): Low-Carbon Conservation Zones (8.4%), Ecological Development Zones (34.6%), Carbon-Intensive Control Zones (16.8%), and High-Carbon Optimization Zones (40.2%). These were further classified into 12 subregions using the Major Function-oriented Zoning (MFZ) framework. Eco-economic coupling analysis reveals complex differentiation: low-carbon maintenance zones combine ecological protection with strong carbon sink capacity, while High-Carbon Optimization Zones face challenges of economic stagnation and emission inefficiency. In agricultural areas, low-carbon subzones achieve synergy between economic growth and carbon sequestration, while high-carbon subzones struggle with entrenched development models. This study’s multi-scale framework balances centralized governance and local adaptation: provinces set carbon targets and equity standards, while counties execute tailored strategies. Shaanxi’s energy–agriculture–ecology gradient offers a model for inland China’s low-carbon shift, exemplified by three pilot zones: Shenmu’s coal CCUS upgrades (industrial decarbonization), Qinba Mountains’ carbon credit banking (ecological value realization), and Guanzhong’s regional carbon trading (urban cluster coordination). During implementation, legal authorization is required, such as the Carbon Compensation Regulations to specify compensation ratio thresholds; the establishment of a Carbon Governance Commission; the integration of carbon compensation into performance evaluation systems; the creation of a special carbon compensation fund (financed by fiscal allocations, carbon taxes, and market financing); and the establishment of environmental arbitration tribunals to resolve disputes. For implementation in other provinces/regions, the MFZ (Major Function Zone) framework should be replaced with local spatial planning systems, and compensation benchmarks should be adjusted accordingly.

(5)

Differentiated carbon compensation policies also present several challenges: payment regions may resist policies due to high compensation expenditures, while recipient regions need to ensure funds are allocated for ecological maintenance rather than misappropriation. To ensure policy feasibility and environmental benefits, it is recommended to establish a “provincially coordinated–municipal/county negotiated” horizontal compensation platform with tiered compensation criteria; implement digital monitoring systems to track compensation fund flows and guarantee their use for ecological restoration; and introduce “carbon sink increment certification” mechanisms to prevent double-counting of existing ecological resources.

The findings of this paper highlight the importance of localized carbon governance strategies that reflect natural endowments and socioeconomic characteristics. A uniform, province-wide approach is insufficient. Instead, precision governance—grounded in spatial differentiation and functional zoning—will be essential for advancing regional carbon peaking and neutrality objectives. The integrated zoning and compensation framework presented here provides a replicable model for other provinces seeking to align regional development with national climate targets.