Spatiotemporal Patterns of Cropland Sustainability in Black Soil Zones Based on Multi-Source Remote Sensing: A Case Study of Heilongjiang, China

Abstract

Sustainable cropland management is essential in maintaining national food security. In the black soil regions of China, which are key areas for commercial grain production, sustainable land use must be achieved urgently. To address the absence of integrated, large-scale, remote sensing-based sustainability frameworks in China’s black soil zones, we developed a comprehensive evaluation system with 13 indicators from four dimensions: the soil capacity, the natural capacity, the management level, and crop productivity. With this system and the entropy weight method, we systematically analyzed the spatiotemporal patterns of cropland sustainability in the selected black soil regions from 2010 to 2020. Additionally, a diagnostic model was applied to identify the key limiting factors constraining improvements in cropland sustainability. The results revealed that cropland sustainability in Heilongjiang Province has increased by 7% over the past decade, largely in the central and northeastern regions of the study area, with notable gains in soil capacity (+15.6%), crop productivity (+22.4%), and the management level (+4.8%). While the natural geographical characteristics show no obvious improvement in the overall score, they display significant spatial heterogeneity (with better conditions in the central/eastern regions than in the west). Sustainability increased the most in sloping dry farmland and paddy fields, followed by plain dry farmland and arid windy farmland areas. The soil organic carbon content and effective irrigation amount were the main obstacles affecting improvements in cropland sustainability in black soil regions. Promoting the implementation of technical models, strengthening investment in cropland infrastructure, and enhancing farmer engagement in black soil conservation are essential in ensuring long-term cropland sustainability. These findings provide a solid foundation for sustainable agricultural development, contributing to global food security and aligning with SDG 2 (zero hunger).

1. Introduction

Cropland provides fundamental resources to ensure agricultural production, food security [1,2], and human survival and development [3]. Sustainable cropland utilization is vital in achieving the zero hunger goal among the United Nations Sustainable Development Goals (SDGs). Beyond direct food provision, cropland also plays a critical role in supporting various regional ecosystem services [4,5], such as soil carbon sequestration [6] and water regulation [7], which are essential for overall environmental well-being and resilience [8]. However, in recent years, climate change, excessive fertilizer use, and unreasonable farming management practices have led to increasingly severe cropland degradation and pollution in China [9], posing a serious threat to sustainable agricultural productivity and influencing food security and the achievement of the SDGs. Therefore, promoting sustainable cropland use and development is a critical issue in agricultural development [10]. To accurately assess the impacts of changes in natural conditions and human activities on the sustainability of cropland, the establishment of a systematic assessment system to support cropland protection and use is crucial.

A cropland sustainability evaluation system offers a comprehensive and systematic approach to analyzing the effects of environmental factors and human activities on cropland [11]. It involves the integration of multiple dimensions, such as soil health [12,13], farm management [13], ecological services [14,15], and intensive agricultural production [16,17]. Among these factors, soil, as the fundamental component of cropland, plays a central role in sustainability assessment and directly impacts the long-term productivity of cropland [18]. Moreover, improper farm management practices can reduce the soil fertility, posing a threat to cropland sustainability [19]. In addition to food production, cropland provides essential ecological services, including soil and water conservation, atmospheric regulation, and carbon sequestration. Researchers have evaluated cropland sustainability from an ecosystem perspective, thereby emphasizing the importance of preserving its ecological functions [20,21]. Additionally, while intensive agricultural production is widely employed to maximize grain yields, excessive intensification can negatively affect soil health and the ecological balance. Therefore, most researchers have focused on assessing intensive agricultural production to guide sustainable agricultural development [16,22]. Overall, the impacts of individual or limited factors on cropland sustainability have been evaluated in existing studies, with fewer comprehensive and systematic evaluation systems for the assessment of cropland sustainability.

To evaluate cropland sustainability comprehensively and accurately, the establishment of an evaluation indicator system that integrates natural conditions, management practices, and crop production factors is essential [23]. The natural characteristics of cropland, including the soil, terrain, and climate, constitute the foundation of agricultural production and significantly influence its sustainability [24]. Among these factors, the soil’s physical and chemical properties are important in assessing soil fertility. For example, the soil organic carbon (SOC) content and soil potential of hydrogen (pH) value directly impact cropland productivity [25,26,27]. Additionally, topographic factors, such as cropland slopes, affect soil and water conservation [26,27]. These phenomena play vital roles in determining the long-term stability of cropland. Climate conditions determine the crop-growing environment, thus influencing farm management measures and the cropland utilization efficiency [28]. Management capabilities also significantly affect cropland sustainability, as they reflect the impact of human activities. For example, irrigation conditions serve as a critical evaluation metric, indicating the adaptability of cropland to climate change [29]. Crop production is the direct embodiment of the social and economic benefits of cropland. Productivity not only directly affects farmers’ income but also reflects the effectiveness and economic feasibility of cropland utilization [30]. Furthermore, cropland is a dynamic system influenced by both natural and human factors, which evolve across temporal and spatial scales [24]. However, many studies lack a comprehensive assessment of cropland sustainability from an integrated spatiotemporal perspective [31,32]. Remote sensing technology is an effective tool to address this challenge. Through the rapid acquisition of ground feature information, remote sensing facilitates the efficient determination of soil properties and crop production characteristics [33,34] and has been widely used in large-scale cropland use evaluations [35,36,37]. By integrating multi-source data, a large-scale cropland sustainability evaluation indicator system can be established, which can not only reveal spatiotemporal differences in large-scale cropland sustainability but also provide paths and suggestions for the accurate implementation of policies.

At present, most methods of evaluating cropland sustainability involve the application of fuzzy comprehensive evaluation models [38,39], the pressure–state–response (PSR) framework [3,40], and the comprehensive index method [24]. Although the fuzzy comprehensive evaluation model can be used to integrate multiple factors to assess cropland sustainability, it is easily affected by subjective influences and involves a complex calculation process [41]. The PSR framework provides a structured approach to analyzing environmental issues, yet its linear causality model neglects the temporal dimension and the complexity of dynamic changes in agricultural systems [11,42]. The comprehensive index method accounts for multiple indicators holistically and is often used in the development of indicator evaluation systems [11,24], which can reflect various aspects of cropland sustainability. Compared to subjective weighting methods such as the analytic hierarchy process (AHP), which often suffer from inconsistency in expert judgment [43], the entropy weight method (EWM) calculates indicator weights objectively based on the intrinsic variability of data, thereby reducing human interference and enhancing the objectivity of the evaluation results [44]. This approach can also provide comprehensive evaluation results rapidly and conveniently [23]. In addition, the integration of remote sensing technology has improved the evaluation process by enabling large-scale and frequent data collection in cropland areas, thereby allowing for the use of extensive sample sizes that improve the robustness of entropy weight calculations by reducing the impacts of individual outliers [24]. This approach provides abundant data support for the comprehensive index method, further enhancing the objectivity and accuracy of the evaluation results [20]. Compared with the fuzzy comprehensive evaluation model, which is prone to subjectivity and computational complexity, and the PSR framework, which is unable to capture the complexity of dynamic systems, the comprehensive index method compensates for these deficiencies, thereby providing a more efficient and objective method for the evaluation of cropland sustainability.

Black soil, referred to as the “giant panda” of cropland, is a valuable agricultural resource. Heilongjiang Province, which is the largest black soil distribution area in China, contains 16 million hectares of black soil cropland, accounting for 50.6% of the total area of China. This renders it a critical region for the protection and sustainable utilization of black soil resources in China. However, owing to long-term overcultivation, improper irrigation and fertilization, and soil compaction caused by mechanical farming practices, the acidification, hardening, and thinning problems of black soil in Heilongjiang Province have become increasingly severe, posing a significant threat to both the productivity and sustainability of black soil [9,45,46]. To address these pressing challenges, the Chinese government has proactively launched a comprehensive series of policies and measures aimed at enhancing black soil protection [45,47,48,49,50]. Moreover, in December 2021, the provincial government of Heilongjiang issued the Heilongjiang Black Soil Protection Project Implementation Plan 2021–2025 [49], in which black soil in Heilongjiang Province was categorized into five zones—plain dry farmland, plain paddy, sloping dry farmland, sloping paddy, and arid windy farmland zones—according to the topographic and natural characteristics. Each zone requires customized management strategies to ensure the effective conservation and sustainable use of black soil. To support these efforts, it is necessary to establish a comprehensive evaluation indicator system for cropland sustainability that is suited to the above five zones and can provide systematic guidance for the protection and use of black soil, thereby ensuring its long-term productivity and ecological stability.

While remote sensing has become indispensable in agricultural monitoring [51,52,53], there remains a research gap in terms of large-scale, integrated studies of black soil regions that utilize the synergistic potential of multi-source remote sensing data. Existing research often focuses on individual indicators or evaluates specific aspects such as productivity [54] or degradation risks [55,56]. These fragmented approaches are insufficient to capture the inherently multi-dimensional and spatially heterogeneous nature of sustainability across diverse black soil zones. Consequently, there is an urgent need for a robust and comprehensive evaluation framework that leverages the complementary strengths of multi-source remote sensing to assess the spatiotemporal dynamics of cropland sustainability in an integrated manner.

To address these limitations in the existing black soil evaluation framework, this study innovates by developing a comprehensive cropland sustainability evaluation system with multi-source remote sensing. This system integrates 13 indicators across four key dimensions: the soil capacity, the natural capacity, the management level, and crop productivity. Furthermore, we apply the EWM to objectively determine indicator weights and, critically, we employ an obstacle factor approach to pinpoint the key limiting factors hindering sustainable agricultural development. Using this framework, we systematically analyze the spatiotemporal characteristics of black soil cropland sustainability in Heilongjiang Province from 2010 to 2020. The aims of this study were to (1) develop a general evaluation indicator system for cropland sustainability in black soil regions; (2) reveal the spatial variations and spatiotemporal dynamics of cropland sustainability in Heilongjiang Province from 2010 to 2020; and (3) provide systematic support for the achievement of sustainable agricultural development by identifying key obstacle factors across different black soil regions and proposing adaptive management strategies (e.g., optimized straw return or soil amendment practices) to increase the crop production capacity and ensure food security.

2. Materials and Methods

2.1. Study Area

Heilongjiang Province is a crucial grain production-dominated region and commodity grain export base in China [45]. Located in the northeast, it extends from 43°26′N to 53°33′N and from 121°11′E to 135°05′E (Figure 1). The topography follows a distinct two high–two low pattern: the northwestern, northern, and southeastern regions are dominated by mountain ranges such as the Greater Khingan, Lesser Khingan, and Zhangguangcai Ranges, whereas the northeastern and southwestern regions feature the low-lying Songnen Plain and Sanjiang Plain. Plains cover more than 60% of the total area of the province, providing advantageous terrain conditions for agricultural activities. Heilongjiang Province has a typical temperate monsoon climate, with an average annual temperature of 8 °C, a ≥0 °C accumulated temperature of 3183.6 °C, and average annual precipitation of 626.9 mm. According to 2020 statistics, the province encompasses approximately 17.18 million hectares of cropland, accounting for 13% of the total cropland area of China and 36% of the land area of Heilongjiang (47.3 million hectares), making it one of the provinces with the largest cropland area in China [57]. Most cropland occurs in the northeastern black soil region of the province, where the soil textures include silty clay loam, silty clay, silty loam, and clay loam, which are suitable for the growth of various grains. The primary crop types in Heilongjiang Province are maize, rice, and soybeans, which accounted for 34.8%, 28.8%, and 21%, respectively, of the cropland area in 2020 [57]. Maize is the primary grain crop, while rice cultivation is concentrated in irrigated areas of the Sanjiang Plain. Moreover, soybean production has expanded in recent years, establishing Heilongjiang as one of China’s key soybean production regions. Therefore, exploring the spatiotemporal evolution of cropland sustainability in this region is important in promoting sustainable agricultural development and ensuring national food security.

2.2. Data Sources and Preprocessing

The cropland data used in this study were derived from the China land use/cover dataset (CLCD) produced on the basis of Landsat Thematic Mapper (TM)/Extended Thematic Mapper (ETM) and Landsat Operational Land Imager (OLI) remote sensing images from 2010 and 2020, with a spatial resolution of 30 m. Google Earth Engine (GEE) was used to process the MOD 16A2 and MOD 13Q1 datasets and extract evapotranspiration and enhanced vegetation index (EVI) data, respectively. ArcGIS Pro (version 3.0) was employed to calculate the slope of the cropland, the annual precipitation and accumulated temperature ≥ 10 °C, the distance between the cropland and the water system, and the distance between the cropland and roads on the basis of digital elevation model (DEM) data, climate data, and hydrological and transportation data, respectively. The centralized contiguity (CC) of cropland was calculated using the contiguity index function in Fragstats (version 4.2) to characterize the contiguity of the cropland concentration. The soil physical and chemical data were derived from the Harmonized World Soil Database dataset (2010) and the High-Resolution National Soil Information Grids of China (2018). All datasets were resampled to a spatial resolution of 250 m × 250 m in ArcGIS Pro, utilizing either the majority or cubic interpolation methods, to ensure spatial analysis consistency. For further details, please refer to

Table 1. Data information.

Criterion Layer	Indicator	Data Source	Year	Resolution
Soil capacity	Soil organic carbon (SOC)	Harmonized World Soil Database (http://www.fao.org/soils-portal, accessed on 28 September 2024) High-Resolution National Soil Information Grids of China (http://www.geodata.cn, accessed on 7 September 2023)	2010/2018	250 m
	Potential of hydrogen (pH)
	Soil texture (ST)
	Cation exchange capacity (CEC)
Natural capacity	River distance (RD)	China 1:100,000 water system dataset (http://www.geodata.cn, accessed on 4 November 2023)	2017	- (Points)
	Slope (S)	The Shuttle Radar Topography Mission (SRTM) digital elevation dataset (https://www.earthdata.nasa.gov, accessed on 15 August 2023)	2000	90 m
	Annual precipitation (AP)	Climate Data (http://data.cma.cn, accessed on 6 September 2023)	2010/2020	- (Points)
	≥10 °C accumulated temperature (AT10)
Management level	Effective irrigation amount (EIA)	MOD 16A2 (https://earthdata.nasa.gov, accessed on 10 July 2024)	2010/2020	250 m
	Road accessibility (RA)	Global Roads Open Access Dataset (https://www.earthdata.nasa.gov, accessed on 8 August 2023) Multi-period road spatial distribution data for 1995/2012/2016/2018/2020 in China (https://www.resdc.cn, accessed on 21 August 2023)	2010/2020	- (Lines)
	Centralized contiguity (CC)	China’s Land Use/Cover Datasets (CLCD) (https://zenodo.org, accessed on 10 August 2023)	2010/2020	30 m
Crop productivity	High cropland productivity (HP)	MOD 13Q1 (https://earthdata.nasa.gov, accessed on 28 June 2024)	2010/2020	250 m
	Stable cropland productivity (SP)

and Table S1.

2.3. Method for Evaluation of Cropland Sustainability

This study proposes a methodology for the evaluation of cropland sustainability (Figure 2). The process commences with the systematic collection of diverse multi-source remote sensing data, encompassing soil property data, remote sensing data, and nature geographic data. This comprehensive data foundation is then subjected to rigorous indicator selection, identifying 13 crucial indicators such as the SOC, soil pH, and high cropland productivity (HP), which are critical in assessing sustainability at the cropland pixel level. Following the assembly of this indicator dataset, the calculation of evaluation indicator weights is performed using the objective EWM. This technique quantitatively determines the importance of each indicator based on its intrinsic variability within the data, ensuring that the weighting reflects the actual information content rather than subjective biases. Finally, these derived indicator weights are applied through weighted summation with the standardized indicator dataset to precisely compute the cropland sustainability score (CSS) for each pixel, thereby providing a comprehensive and data-driven assessment of cultivated land sustainability.

2.3.1. Indicator Selection

On the basis of existing research [11,58,59] and the national cropland quality evaluation standard [60], cropland sustainability was assessed on the basis of four key aspects: natural geographical characteristics, soil physicochemical characteristics, cropland management characteristics, and cropland production characteristics. Through an in-depth analysis of the natural endowments, soil quality, farming system, and cropland productivity, the primary factors influencing sustainable cropland use were identified.

• Soil capacity characteristics

Soil quality is a fundamental component of cropland sustainability and productivity [61]. In this study, we selected the SOC content, soil pH, soil texture (ST), and cation exchange capacity (CEC) as key indicators [62]. The SOC is an important factor in ensuring the cropland production capacity and a core indicator for the evaluation of soil quality [63]. The soil pH plays a pivotal role in soil nutrient transformation and soil fertility, with its suitability being directly related to crop growth and nutrient uptake [64]. The ST, which is a basic soil physical property, determines the capacity of soil for water, nutrient, and gas exchange, retention, and absorption [36]. The CEC reflects the ability of soil to retain and supply nutrients, thus serving as a key indicator of soil fertility [11,65].

2.

Natural capacity characteristics

Natural factors such as the topography, climate, and geographical location play significant roles in shaping cropland productivity. In this study, the slope of cropland (slope, S), the distance from cropland to the river (river distance, RD), the annual precipitation (AP), and the accumulated temperature above 10 °C (≥10 °C accumulated temperature, AT10) were employed as natural capacity characteristic indicators. S is a widely used metric in assessing cropland topography, as steeper slopes exhibit an increased probability of soil erosion and nutrient loss, thus reducing cropland suitability [66]. Water availability is another critical factor for cropland production and can be represented by the RD. Proximity to rivers enhances the irrigation efficiency, thereby reducing the difficulty of drainage irrigation and increasing the irrigation capability. AP is the most important water source for crop growth, and its distribution and adequacy directly impact the crop growth cycle, yield, and quality. In addition, AT10 is a crucial thermal indicator of crop growth, particularly in Heilongjiang, which is located in China’s northernmost region. Cooler temperatures, by slowing crop development and prolonging the growth period, can negatively impact crop yields and quality. Therefore, attention must be given to the effects of the accumulated temperature on crop growth.

3.

Management level characteristics

Notably, the sustainability of cropland, which encompass semi-natural and semi-artificial ecosystems, is inevitably and significantly affected by human activities and socioeconomic factors. In this study, three key indicators were selected to evaluate cropland management: transportation accessibility (road accessibility, RA), the irrigation capacity (effective irrigation amount, EIA), and the cropland spatial characteristics (centralized contiguity, CC). RA serves as an important metric for transportation accessibility. A smaller distance is correlated with an increased agricultural machinery utilization frequency; increased efficiencies in crop planting, field management, and harvesting; and accelerated agricultural product circulation, ultimately increasing the cropland management capacity and economic returns. The irrigation capacity is an important manifestation of cropland infrastructure construction and is directly related to production stability in cropland. The EIA is an important indicator in evaluating the irrigation level of cropland and can be calculated as follows:

$$EIA=ET-P_e,$$

(1)

$$P_e=\sum _{i=1}^n\left\{\begin{array}{c}{\displaystyle \frac{P_i\left(4.17-0.2P_i\right)}{4.17}},P_i<8.3 \mathrm{m}\mathrm{m}\\ 4.17+0.1P_i,P_i\ge 8.3 \mathrm{m}\mathrm{m}\end{array}\right.,$$

(2)

where ET is the cumulative annual actual evapotranspiration derived from MOD 16A2 remote sensing evapotranspiration data, $P_e$ is the cumulative annual effective rainfall, $P$ is the rainfall on day $i$, and $n$ is the number of days in a year. The unit of all parameters is millimeters (mm).

Furthermore, to increase intensive cropland utilization and facilitate agricultural production activities, stakeholders often concentrate and connect cropland. In this study, we adopt the CC, which is a key indicator characterizing the spatial attributes of cropland, to measure spatial connectivity. This indicator can reflect the degree of spatial aggregation of cropland and indicate whether cropland exhibits spatial characteristics suitable for mechanized operation and centralized resource management.

4.

Cropland productivity characteristics

Cropland productivity directly reflects its sustainability and plays an important role in ensuring food security. In this study, cropland productivity was evaluated on the basis of high yield (HP) and yield stability (stable cropland productivity, SP) levels [11,24]. Remote sensing-derived vegetation indices effectively characterize the vegetation growth status and biomass, with long-time vegetation indices capable of representing the cropland productive capacity [34]. Strong vegetation growth is closely associated with grain yields; therefore, it is often employed to estimate and evaluate crop yields. Notably, the EVI can better reveal the spatiotemporal variations in the cropland productive capacity because it reduces atmospheric effects and decouples canopy background signals [67]. Furthermore, among the commonly used vegetation indices, the EVI has the strongest spatiotemporal correlation with actual cropland yields [11]. Therefore, we used the mean annual EVI to quantify HP and the coefficient of variation (CV) of the annual EVI over a continuous five-year period to quantify SP. For HP, a higher average annual EVI corresponds to a higher productivity score. Conversely, for SP, a lower coefficient of variation indicates greater stability, reflecting that cropland with less fluctuation in its annual EVI will receive a higher SP score. Specific calculation details for these scores are provided in Table S1.

2.3.2. Calculation of the Evaluation Indicator Weights

In this study, the EWM was applied to determine the objective weight of each evaluation indicator. The primary strength of the EWM lies in its ability to eliminate the influence of subjective factors and objectively capture the differences between evaluation indicators, making it particularly suitable for the assessment of cropland sustainability across different regions. The EWM assigns weights on the basis of the variability in each indicator. Specifically, when the information entropy $E_j$ is low, this indicates notable variation in the indicator, which provides more valuable information for comprehensive evaluation. As a result, the weight assigned to this indicator increases accordingly. The specific calculation steps of the EWM are as follows.

1.

Standardization of indicator data

To eliminate the impact of different units and dimensions, the min–max normalization method is employed to standardize the raw data. The standardization process is as follows:

$$Y_{ij}=\left\{\begin{array}{ll}{\displaystyle \frac{X_{ij}-X_{jmin}}{X_{jmax}-X_{jmin}}}& (positive indicators)\\ {\displaystyle \frac{X_{jmax}-X_{ij}}{X_{jmax}-X_{jmin}}}& (negative indicators)\\ {\displaystyle \frac{1-\frac{\left|X_{ij}-a\right|}{max\left(\left|X_{ij}-a\right|\right)}-X_{jmin}}{X_{jmax}-X_{jmin}}}& (moderate indicators)\end{array}\right.,$$

(3)

where $X_{ij}$ is the original value of the $i$-th indicator in the $j$-th sample; $X_{jmax}$ and $X_{jmin}$ are the maximum and minimum values, respectively, of the $j$-th indicator; and $a$ is the optimal value of a given moderate indicator. In this study, positive indicators include the SOC, ST, CEC, AP, AT10, EIA, CC, and HP; negative indicators include the RD, S, RA, and SP; and moderate indicators include the pH.

2.

Calculation of the entropy value $E_j$

On the basis of information theory, the information entropy $E_j$ of each indicator can be calculated as

$$E_j=-ln{\left(n\right)}^{-1{\sum }_{j=1}^n{p}_{ij}ln\left(p_{ij}\right)},$$

(4)

where $p_{ij}={\displaystyle \frac{Y_{ij}}{{\sum }_{j=1}^n{Y}_{ij}}}$ denotes the proportion of the standardized value of the i-th indicator in the j-th sample and n is the total number of samples.

3.

Calculation of the indicator weight $W_i$

The weight of each indicator $W_i$ can be calculated as

$$W_j={\displaystyle \frac{1-E_j}{k-{\sum }_{i=1}^k{E}_j}},$$

(5)

where $E_i$ is the indicator entropy of the $i$-th indicator and k is the total number of evaluation indicators.

Through the above calculation process, the weight coefficient of each evaluation indicator was determined. These weight coefficients were used to calculate the CSS, as detailed in

Table 2. Indicator system and weights for evaluation of cropland sustainability.

Target Layer	Criterion Layer	Indicator Layer	Unit	Indicator Interpretation	Weight
Cropland sustainability	Soil capacity	Soil organic carbon (SOC)	g/kg	Content of organic matter containing C	0.0524
		Potential of hydrogen (pH)	-	pH value of soil surface	0.0807
		Soil texture (ST)	-	Composition of mineral particles with different particle sizes	0.0624
		Cation exchange capacity (CEC)	cmol(+)/kg	The ability of soil to adsorb nutrients and eventually release them back into the soil solution	0.0709
	Natural capacity	River distance (RD)	km	Distance from cropland to the river	0.0534
		Slope	°	Steepness or inclination of cropland	0.0474
		Annual precipitation (AP)	mm	Climatic conditions	0.0769
		≥10 °C accumulated temperature (AT10)	°C	Climatic conditions	0.0528
	Management level	Effective irrigation amount (EIA)	mm	Effective irrigation amount of cropland	0.0586
		Road accessibility (RA)	km	Distance from the cropland to the nearest road	0.0480
		Centralized contiguity (CC)	-	Contig landscape index	0.0515
	Crop productivity	High cropland productivity (HP)	-	Mean of EVI for five consecutive years	0.0571
		Stable cropland productivity (SP)	-	CV of EVI for five consecutive years	0.0551

.

2.3.3. Calculation of the Cropland Sustainability Score

In this study, the comprehensive index method was applied to calculate the CSS. The CSS was obtained by multiplying the normalized score of each evaluation indicator by its corresponding weight and summing the weighted scores. The CSS can be calculated as follows:

$$\mathrm{C}\mathrm{S}\mathrm{S}=\sum _{i=1}^n{W}_i\times Y_i,$$

(6)

where $W_i$ is the weight of the $i$-th evaluation indicator, $Y_i$ is the normalized score of the $i$-th evaluation indicator, and $n$ is the total number of evaluation indicators (in this study, $n=13$).

2.3.4. Obstacle Factor Diagnosis Model and Spatial Autocorrelation Analysis

To identify the key factors limiting improvements in cropland sustainability, an obstacle factor diagnosis model was applied. This model evaluates the obstacle degree $O_i$ of each indicator on the basis of three fundamental factors: the contribution degree of the factor $W_i$, the deviation degree of each indicator, and the obstacle degree [68]. The higher the value of $O_i$ is, the greater the negative impact on the improvement in cropland sustainability, thus identifying the indicator as a major obstacle factor. The obstacle degree $O_i$ can be calculated as follows:

$$O_i={\displaystyle \frac{\left(1-I_i\right)\times W_i}{{\sum }_{i=1}^n\left[\left(1-I_i\right)\times W_i\right]}}\times 100\%,$$

(7)

where $O_i$ is the obstacle degree of the $i$-th indicator and $W_i$ is the weight (contribution degree) of the $i$-th indicator. Moreover, $(1-I_i)$ is the deviation degree, which reflects the gap between the highest value and the actual value of the indicator.

To explore the spatiotemporal patterns of cropland sustainability changes in Heilongjiang Province, global and local spatial autocorrelation analyses were conducted in ArcGIS Pro (version 3.0). The global spatial autocorrelation was assessed using Moran’s I, which revealed whether the spatial distribution of cropland sustainability values exhibited clustered, dispersed, or random patterns. The local indicators of spatial association (LISA) method was used to identify local spatial clustering patterns of changes in cropland sustainability, based on the Anselin Local Moran’s I statistic, to identify statistically significant clusters of high values (hot spots) and low values (cold spots). The Anselin Local Moran’s I statistic compares each feature’s attribute value to the weighted average of its neighbors to detect localized spatial associations. A statistically significant positive value indicates the spatial clustering of similar values—either high–high (hot spots) or low–low (cold spots). Statistical significance is typically assessed using p values, with a threshold of 0.05. The hot spots and cold spots of cropland sustainability identified through this research provide a sound foundation for the optimization of land use and development of effective policies.

3. Results

3.1. Spatiotemporal Changes in Cropland Sustainability in Heilongjiang Province

From 2010 to 2020, the comprehensive CSS in Heilongjiang Province increased significantly, by 7.0% (

Table 3. The average scores of criterion layers and indicators from 2010 to 2020.

Criterion Layer	Mean in 2010	Mean in 2020	Indicator Layer	Mean in 2010	Mean in 2020
Soil capacity	16.6087	19.1937	Soil organic carbon (SOC)	6.1525	5.2880
			Potential of hydrogen (pH)	0.4492	2.6912
			Soil texture (ST)	1.1415	4.4843
			Cation exchange capacity (CEC)	8.8279	6.7302
Natural capacity	21.7745	21.4170	River distance (RD)	5.8938	5.8853
			Slope	5.9724	5.9723
			Annual precipitation (AP)	4.8091	4.7577
			≥10 °C accumulated temperature (AT10)	5.0991	4.8016
Management level	17.0984	17.9115	Effective irrigation amount (EIA)	4.9852	5.6250
			Road accessibility (RA)	5.7540	5.8864
			Centralized contiguity (CC)	6.3557	6.3976
Crop productivity	5.5260	6.7632	High cropland productivity (HP)	4.1868	4.3760
			Stable cropland productivity (SP)	1.3392	2.3872
Cropland sustainability score				61.2036	65.4823

, Figure 3a,b), indicating significant progress in achieving cropland sustainability over the past decade. To statistically validate this temporal change, we conducted Welch’s t-tests, which confirmed that the differences in the CSS and its criterion layers between 2010 and 2020 were statistically significant (all p < 0.01; Table S2). The most notable improvements were observed for the soil capacity and cropland productivity capacity, with increases of 15.6% and 22.4%, respectively. This reflects a comprehensive increase in soil quality and the agricultural productivity sustainability levels. Moreover, while an overall improvement was observed across various criterion layers, significant spatial heterogeneity occurred (Figure 3c–j). The central region presented higher soil capacity scores, reflecting greater cropland sustainability, whereas the western and northeastern regions presented lower scores, suggesting an urgent need for soil improvement and conservation in these regions (Figure 3c,d). The changes in the natural capacity characteristics were relatively minor, with lower scores in the western region due to potential disadvantages in terms of climate and terrain, whereas the higher scores in the central and eastern regions were due to favorable natural resource conditions (Figure 3e,f). The management level increased by 4.8%, with minimal changes in spatial patterns, reflecting relatively balanced advancements in agricultural management practices across all regions (Figure 3g,h).

The spatial distributions of cold and hot spots related to changes in cropland sustainability from 2010 to 2020 revealed obvious regional differences (Figure 4). Hot spots for cropland sustainability change (regions with significant improvement) occurred primarily in Suihua and Harbin, whereas cold spots (regions with notable reductions) were distributed mainly in Daqing, Qiqihar, and Jixi (Figure 4a). This pattern highlights the uneven spatial distribution of changes in cropland sustainability, with some regions facing challenges related to cropland quality degradation, thus requiring targeted improvement strategies. The spatial distributions of cold and hot spots in each criterion layer exhibited distinct characteristics. The spatial pattern of soil capacity changes was similar to that of cropland sustainability changes (Figure 4b), with hot spots in Suihua and Harbin reflecting substantial increases in the soil capacity in these regions, whereas cold spots were concentrated in Daqing, Qiqihar, and Jixi, suggesting ineffective improvements or degradation in the soil capacity, thus requiring enhanced soil protection and improvement measures. In terms of natural capacity characteristics, hot spots occurred mainly in Suihua, Daqing, Hegang, Jiamusi, and Harbin, reflecting improved or better-used natural resource conditions in these areas, whereas cold spots were concentrated in Jixi and Qiqihar (Figure 4c), indicating potential adverse changes in climate, terrain, and other natural conditions. The spatial pattern of cropland management level changes was relatively scattered, with hot spots in Suihua and cold spots in Jiamusi and Jixi (Figure 4d). This suggests considerable differences in cropland management level changes among these regions, possibly due to differences in infrastructure development and management practices. The spatial distribution of the changes in cropland productivity differed from those in the other layers, with hot spots in Qiqihar, Daqing, and Jiamusi and cold spots in Shuangyashan and Jixi (Figure 4d). These changes are likely influenced by the crop production patterns and land use, suggesting that more targeted strategies should be implemented to increase the productivity capacity in low-performing regions.

3.2. Spatiotemporal Pattern of Cropland Sustainability in Black Soil Zones of Heilongjiang Province

From 2010 to 2020, the overall CSS increased across all black soil zones in Heilongjiang Province, but the degree of increase in each zone differed. We conducted independent-sample t-tests to assess the statistical significance of these changes. The degrees of improvement were as follows: sloping dry farmland (9.98%) > sloping paddy (9.38%) > plain dry farmland (6.18%) > arid windy farmland (3.70%) > plain paddy (2.81%). For all regions, the t-tests indicated a statistically significant difference in the CSS between 2010 and 2020 (all p < 0.01; Table S3). Among them, the average CSS values in plain dry farmland and arid windy farmland were relatively high, indicating high cropland production potential, and the overall performance was relatively satisfactory (Figure 5).

There were notable spatial variations in the CSS changes across the five major regions. In plain dry farmland, Bayan and Wangkui Counties presented relatively high CSS values, reflecting the better management and utilization of cropland resources in these regions. Conversely, areas such as Youyi, Yi’an, Mingshui, Qinggang, and Nehe County presented relatively low scores, indicating potential issues related to cropland quality degradation or management practices (Figure 6a,b). In the sloping dry farmland areas, Mishan County in the eastern part exhibited a decline in the CSS between 2010 and 2020, suggesting a possible cropland degradation trend in this region (Figure 6c,d). Sloping paddy areas demonstrated a notable increase in scores, with the primary improvements concentrated in the southern part of Qing’an County. This progress may be attributed to advancements in water conservation infrastructure construction, irrigation management, and agricultural technology implementation in the region (Figure 6e,f). The plain paddy areas showed modest improvements, with key enhancements observed in Fujin, Tongjiang, and Beilin Counties. However, a decline occurred in Hulin County (Figure 6g,h), potentially linked to local climate change, adjustments in land use patterns, or other agricultural management challenges. In arid windy farmland, spatial improvements in cropland sustainability were limited but heterogeneous. Certain areas within Fuyu and Longjiang Counties presented relatively low scores, highlighting opportunities for further enhancement (Figure 6i,j). The cropland sustainability in this region is constrained by soil erosion, desertification, and other factors. Therefore, it is important to strengthen soil and water conservation efforts and implement soil improvement measures to promote sustainable development in these areas.

3.3. Obstacle Factors of Cropland Sustainability

From 2010 to 2020, we calculated the obstacle degrees of the cropland sustainability evaluation indicators in Heilongjiang Province using the obstacle factor diagnosis model (Figure 7). This model identifies factors that significantly hinder progress toward sustainability. The results revealed that the SOC, EIA, and AP were the main obstacles limiting cropland sustainability, collectively accounting for 56% of the total obstacle degree. These factors reflect the critical role of soil quality, climate change, and water resource management in increasing cropland sustainability. The main obstacles to cropland sustainability varied across the five major black soil zones, but the SOC, AP, and slope were identified as common limiting factors. Specifically, the AP and SOC were the main obstacle factors, followed by the slope, RA, and RD, collectively accounting for 60% of the total obstacle degree in the sloping paddy zone. These results highlight challenges related to the water supply, soil quality, and infrastructure development in this region, particularly the instability of the AP, which significantly impacts paddies. In terms of regional differences, the most critical factor influencing cropland sustainability in sloping dry farmland, plain dry farmland, and plain paddy fields was the SOC, followed by the AP, SP, slope, RA, and RD, which collectively accounted for 66%, 69%, and 56%, respectively, of the total degree of obstacles in each region. These findings indicate that soil fertility and climatic conditions are key constraints in these areas, whereas a stable water supply and infrastructure construction are essential in enhancing cropland sustainability. In arid windy farmland, the SOC, slope, RA, RD, and AP were the main limiting factors, collectively accounting for 60% of the total obstacle degree. Cropland sustainability in these areas is severely restricted by the natural environment, especially the lack of precipitation and low SOC, which have notable negative impacts on land productivity.

4. Discussion

4.1. Construction of a Comprehensive Cropland Sustainability Evaluation Indicator System for the Black Soil Region

The evaluation of cropland sustainability constitutes the basis for ensuring long-term cropland use and protection. However, traditional evaluation methods often focus on a single perspective, such as soil health [18] or ecological impacts [69], and fail to comprehensively and objectively reflect the overall sustainability of cropland systems. Sustainable cropland development is affected by multiple factors [70], including soil health and fertility, water resource management and the utilization efficiency, topographic features and soil and water conservation, environmental suitability, infrastructure development, and agricultural productivity. Traditional methods neglect the interactions between these factors, thereby limiting the accuracy and comprehensiveness of the sustainability evaluation results. In China’s key agricultural heartland—the black soil region of Heilongjiang—systematic and comprehensive evaluations of cropland sustainability remain scarce. Existing assessment frameworks often utilize indicators that are poorly suited to the province’s distinct climatic conditions. For example, critical thermal indicators such as the accumulated temperature (≥10 °C), which are essential for agricultural assessment in Northern China, are frequently overlooked [11]. In addition, some studies neglect key aspects of agricultural productivity or the natural land conditions [71], thereby limiting their scope.

In large-scale studies, the evaluation of cropland sustainability has predominantly relied on statistical data [31,72], which cannot capture the spatial heterogeneity in cropland sustainability within regions. This limitation is also reflected in recent assessments of Heilongjiang Province. Although some studies report improvements in cropland sustainability in Heilongjiang between 2012 and 2022 [73], these studies primarily rely on coarse-grained statistical data, failing to capture detailed spatial patterns or focus specifically on the province’s black soil zones, which are crucial to national food security. Although field-based measurement methods can offer detailed insights into local conditions, their high cost and operational complexity hinder their applicability at broader spatial scales [18]. Therefore, remote sensing, which can quickly provide large-scale spatial information, has become an ideal approach in assessing cropland sustainability at a regional scale. However, existing studies based on remote sensing often focus on exploring sustainability and analyzing the cropland use intensity [74] or urban land use [75], and they cannot be effectively applied to unique agricultural regions such as the black soil area of Heilongjiang Province.

To address this issue, a comprehensive evaluation indicator system was established for black soil zones, which are known for their high-quality soil resources but are limited by climatic conditions. Using multi-source remote sensing data, such as MOD 16A2 and MOD 13Q1, along with available soil data and climate data, we selected 13 key indicators, including the SOC content, annual precipitation, accumulated temperature ≥ 10 °C, and effective irrigation amount, as well as the soil capacity, management level, natural capacity, and crop productivity, to comprehensively evaluate the spatiotemporal patterns of cropland sustainability in Heilongjiang’s black soil zones. To obtain indicator weights, we implemented both the AHP and EWM. The AHP weights were derived from expert scoring involving 10 specialists. However, inconsistencies were observed in the expert opinions for several indicators (Table S4), reflecting the inherent subjectivity of the AHP approach. In contrast, we considered the entropy weighting method to be more appropriate for the evaluation of indicator weights in this study, as it reduces the reliance on subjective judgment and is better suited to large-scale, data-driven assessments. This approach ensures greater consistency and robustness in capturing regional variability across the black soil zone. The system comprehensively captures cropland sustainability in black soil areas, thus providing theoretical support and practical guidance for the protection and sustainable use of black soil resources.

4.2. Potential Driving Mechanisms of the Spatial and Temporal Changes in Cropland Sustainability

Changes in cropland sustainability are influenced by a variety of factors, including natural factors such as the topography and landform, climate change, and human factors such as unreasonable farming management and agricultural activities, which often interact with and jointly affect changes in cropland sustainability. Especially in Northeastern China, the cold, dry, and windy winters, coupled with hot and rainy summers, accelerate soil erosion, directly impacting soil fertility and quality and posing a threat to cropland sustainability. In addition, long-term intensive tillage without conservation, unsustainable cultivation practices (e.g., deep plowing), fertilization methods (e.g., excessive use of chemical fertilizers), irrigation methods (e.g., overirrigation), and the implementation of a single crop rotation system have caused damage to the soil structure, exacerbated soil degradation, and posed a threat to cropland sustainability. More importantly, the growing demand for cropland resources, driven by high-intensity agricultural production and rapid economic development, has placed increasing pressure on cropland in Heilongjiang Province, where agriculture dominates the economy [46,76]. This pressure is clearly demonstrated by the significant expansion of the cropland area, which increased by 8% or from 15,858.02 thousand hectares in 2010 to 17,180.2 thousand hectares in 2020 [57]. This expansion, coupled with high-intensity agricultural production activities, not only risks resource overexploitation in the black soil region but also leads to declining soil quality, excessive water consumption, and ecosystem imbalance, posing severe challenges to sustainable cropland development.

The obstacle factor diagnosis model results of this study revealed that the SOC content, effective irrigation amount, and annual precipitation were the main obstacle factors influencing cropland sustainability in Heilongjiang Province, collectively accounting for 56% of the total obstacle degree. These findings suggest that improving the soil quality, implementing rational irrigation management, and using water resources efficiently are key strategies in promoting sustainable cropland development. Improving soil quality and fertility can significantly strengthen the soil’s capacity for carbon sequestration, thereby providing crucial climate regulation services to regional ecosystems [77]. Practices like straw return, organic fertilizer application, and green manure incorporation are crucial for this. Moreover, implementing rational irrigation management and efficient water resource utilization ensures stable crop yields while effectively reducing groundwater overextraction [78] and optimizing regional hydrological cycles [79]. This involves using precision irrigation technologies (e.g., drip systems) and optimizing irrigation scheduling based on crop needs and weather conditions [80]. These specific interventions are crucial in addressing the identified challenges and fostering sustainable cropland development.

In practice, the application of scientific and sustainable agricultural technologies has effectively improved cropland sustainability. For example, the Longjiang model, the Lishu model [81], and the Bei’an model [82], implemented by the Chinese Academy of Sciences through the scientific and technological campaign of the Black Soil Granary, have significantly increased the soil fertility and productivity levels [83]. By implementing soil and water conservation engineering, straw crushing or no tillage with high stubble, and grain–forage rotation systems, these models have increased soil water storage and the soil organic matter content in the whole cultivated layer, thereby increasing the soil fertility and achieving high and stable cropland productivity [76]. Specifically, soil and water conservation engineering helps to reduce soil erosion and restore the natural functions of soil. Compared with conventional tillage practices, straw crushing and no tillage with high stubble increase the soil’s structural stability and moisture retention by reducing soil disturbances. Moreover, grain–forage rotation not only helps to increase soil biodiversity but also effectively avoids the excessive consumption of soil nutrients and reduces pest and disease outbreaks [84], thereby strengthening the resilience and stability of regional ecosystems and ensuring the sustainable production capacity of crops [85]. These comprehensive measures enhance cropland productivity and promote the sustainable development of black soil resources, providing valuable experience for their protection and for improvements in agricultural and ecological systems.

In addition to technological innovations, policy support plays a crucial role in regulating human activities and protecting natural resources. A series of black soil protection policies, such as the construction of well-facilitated farmland [86] and the promotion of conservation tillage practices [87], have been issued by the Chinese central government and Heilongjiang provincial government. The implementation of these policies has not only effectively promoted the sustainable development of agriculture but also provided a policy guarantee for soil protection and improvement. By the end of 2023, approximately 12.4 million hectares of well-facilitated farmland had been established across the three northeastern provinces of China [88,89,90]. Through the development of cropland infrastructure, such as water conservancy facilities, irrigation and drainage systems, and field consolidation engineering, well-facilitated farmland establishment has created a more stable agricultural production environment while significantly enhancing the soil quality. Simultaneously, by increasing investment and promoting technological progress, especially in the application of technology in soil and water conservation, soil fertility improvement, and precision agriculture, the production environment and natural conditions of black land have been significantly improved. These measures have not only curbed soil erosion and mitigated soil degradation but also increased the management, protection, and sustainable utilization levels of black soil resources, effectively increasing the capacity for high and stable cropland productivity [19,91].

4.3. Suggestions for Improvements in Cropland Sustainability

Cropland constitutes a key component of national food security and stability. To ensure sustainable cropland development, it is essential to implement the strategy of “storing grain in land and technology”, which not only safeguards the quantity of cropland but also enhances the cropland conservation and quality levels.

Encouraging scientific and technological innovation in cropland conservation and use and promoting the implementation of technological models are necessary. The application of key techniques—such as black soil degradation prevention and control, soil fertility enhancement, and ensuring high yields and efficiency in agricultural activities—can lead to significant improvements in the SOC content, soil fertility, and soil structure while also reducing soil erosion and increasing the water use efficiency, with cost savings and efficiency gains [92,93,94,95]. Given the different production obstacles associated with black soil, it is crucial to integrate technologies from agriculture, biology, information science, etc. Through the integration of multi-disciplinary techniques, technical models and system solutions that are suited to diverse black soil types, geographical conditions, and agricultural production requirements can be developed. Moreover, the establishment of core demonstration zones can serve as a crucial step in accelerating the widespread adoption of these technological models. By employing a point-to-surface expansion strategy, in which innovations are first implemented in core areas and then extended to surrounding regions, large-scale application can be effectively realized [45], which can increase cropland sustainability and ensure the long-term productivity and resilience of black soil agriculture.

Promoting the construction of cropland infrastructure and enhancing the overall production capacity of cropland are essential. Sustainable cropland development cannot be achieved without well-functioning infrastructure. The intensity of field consolidation should be increased, the irrigation and drainage conditions should be improved, mechanized farming roads should be enhanced, and the cropland concentration and connectivity should be increased. Moreover, upgrading water conservancy facilities, promoting the adoption of water-saving irrigation techniques, and rationally developing and using surface water while reducing groundwater extraction are critical measures to ensure the sustainable use of water resources and cropland. Additionally, it is necessary to improve the comprehensive management of sloping farmland, where transitioning from traditional slope planting to the use of mechanical ridges and horizontal planting can effectively control soil erosion, mitigate the thinning of black soil layers, and increase the sustainability of black soil productivity.

Furthermore, establishing a long-term mechanism for the protection and sustainable use of cropland and enhancing farmers’ motivation to conserve black soil are critical. Under the guidance of laws, regulations, plans, and implementation schedules issued by national and local governments, efforts should be directed toward improving overall policy planning and enhancing the integration and coordination of relevant funds from multiple stakeholders. Moreover, reinforcing both horizontal and vertical coordination and clarifying individual responsibilities can help to establish an effective multi-stakeholder mechanism that pools resources and efforts to jointly advance the protection and use of black soil [49]. Additionally, optimizing the compensation system to ensure the fair distribution of benefits among stakeholders, as well as improving the assessment and reward mechanisms, can facilitate the development of a comprehensive compensation and incentive framework for cropland conservation across the integrated three-tier provincial–municipal–county system. Through the successful implementation of demonstration zones, efforts should be intensified to promote comprehensive technical training, increase the effectiveness of black soil protection measures, and increase farmers’ understanding and acceptance of conservation practices. By creating a strong societal atmosphere for black soil protection, a favorable environment can be established in which all sectors actively contribute to the long-term preservation and sustainable use of this critical agricultural resource.

5. Conclusions

This study developed a comprehensive evaluation method to assess cropland sustainability and its spatiotemporal changes in Heilongjiang Province’s black soil region from 2010 to 2020. The method uses multi-source remote sensing data and focuses on four dimensions: the natural capacity, the soil capacity, the cropland management level, and cropland productivity. By integrating 13 remote sensing and ancillary indicators and employing the entropy method for weight assignment and obstacle diagnosis, this research offers a holistic analysis of cropland sustainability. This approach objectively incorporates diverse indicators while preserving their relative importance and interactions, which is conducive to promoting the interaction of factors to evaluate the sustainability of cropland. The results reveal that, from 2010 to 2020, the overall cropland sustainability in Heilongjiang Province increased by 7.0%, with the most significant improvements occurring in the central and northeastern regions. Among the four dimensions, the soil capacity and cropland productivity characteristics exhibited the most significant increases, increasing by 15.6% and 22.4%, respectively, whereas the cropland management characteristics improved by 4.8%. Although natural geographical characteristics did not show improved overall scores, they displayed significant spatial heterogeneity, with better conditions in the central/eastern regions than in the western regions. In terms of the different black soil zones, sloping dry farmland and sloping paddy areas showed the greatest improvements, followed by plain dry farmland and arid windy farmland, with plain paddy areas demonstrating the least improvement. The obstacle factor diagnosis model identified the SOC content and effective irrigation amount as the two primary limiting factors influencing cropland sustainability. Therefore, targeted black soil degradation control and cropland protection strategies should be implemented on the basis of regional characteristics and the identification of key obstacle factors. Specifically, for sloping cropland, policies should focus on the continued promotion of terracing and contour farming to prevent soil erosion and maintain productivity gains. In plain dry farmland and arid windy farmland, efforts should prioritize increasing the organic matter input to address the identified limiting factors of the SOC content and effective irrigation. For plain paddy areas, which showed the least improvement, a more urgent and comprehensive intervention is needed, potentially focusing on improving the soil fertility and optimizing water resource allocation. Furthermore, across all black soil zones, it is crucial to foster a strong societal atmosphere for black soil protection, which can inspire farmers to actively engage in conservation practices. In future research, constructing a multi-scale cropland sustainability evaluation indicator system, accurately obtaining key indicator data, and exploring the synergistic mechanism underlying the influence of multiple factors on sustainable cropland evolution will be critical. This study not only provides decision-making support for black soil protection and sustainable use but also a solid foundation for efforts to ensure food security and achieve the SDG of zero hunger.