Spatio-temporal Dynamics of Vegetation Cover in Hulun Buir from 2000 to 2022 and Its Response to Climate Factors

ZOU Xiang, ZHANG Yuting, XU Lu
School of Geography, Geomatics and Planning, Jiangsu Normal University, Xuzhou 221116, Jiangsu, China

Abstract: Investigating the spatiotemporal variation characteristics of vegetation cover in Hulun Buir provides valuable insights for formulating precise vegetation restoration and ecological environmental protection policies, thereby facilitating ecological civilization construction in the region. Using Sen+Mann-Kendall trend analysis, the Hurst index, linear regression, and partial correlation methods, this study quantitatively analyzes the dynamic distribution characteristics and future trends of vegetation in Hulun Buir from 2000 to 2022 based on monthly maximum value composite remote sensing data combined with temperature and precipitation datasets, and discusses the response of vegetation cover to climatic factors. The results indicate that: (1) Vegetation cover in Hulun Buir increased at a rate of $0.0021 \cdot a^{-1}$ based on the Normalized Difference Vegetation Index (NDVI). Two distinct trend phases were observed during the study period: a stable increasing trend from 2000 to 2010 ($0.00007 \cdot a^{-1}$) and a significant increasing trend from 2010 to 2022 ($0.0031 \cdot a^{-1}$), reflecting that vegetation cover is influenced not only by climatic factors but also significantly by ecological protection policies. (2) Seasonally, the change rates follow the pattern of spring ($0.0031 \cdot a^{-1}$) > winter ($0.0021 \cdot a^{-1}$) > summer ($0.0019 \cdot a^{-1}$) > autumn ($0.0014 \cdot a^{-1}$). Spatially, annual mean NDVI decreases gradually from the central Greater Hinggan Mountains to the eastern hills and western grasslands. (3) The sensitivity of vegetation cover to climate change varies across space and time, showing greater sensitivity to precipitation in the western grassland region spatially, while being more sensitive to temperature in terms of interannual variation. Overall, vegetation cover in Hulun Buir demonstrates an improving trend, although localized degradation has emerged. In the future, forest vegetation in the Greater Hinggan Mountains region may face degradation risks.

Keywords: NDVI; vegetation cover; sustainable development; Hurst index; Hulun Buir

Introduction

Vegetation plays a pivotal role in ecosystems as primary producers and serves as a key indicator for evaluating regional ecological environmental changes. Acting as a medium for interactions among soil, atmosphere, and water, vegetation tightly connects these spheres in nature. Therefore, exploring vegetation changes and their influencing factors across different spatiotemporal scales holds significant importance for sustainable ecosystem development. Vegetation growth is closely related to natural conditions and human production activities, such as climate, hydrology, terrain, and urban construction. Temperature and precipitation play crucial roles in influencing plant growth and development, with climate change altering plant growth environments and consequently affecting vegetation species, distribution, and community structure. Human activities can alter vegetation cover trends and even exert profound impacts on the global ecological environment. Unreasonable anthropogenic activities, including deforestation, over-cultivation and grazing, and land use conversion, cause significant declines in vegetation cover, while afforestation, returning farmland to forest, and improved agricultural irrigation technologies enhance vegetation cover. Therefore, systematically and quantitatively studying the impacts of climate change and social factors on vegetation spatiotemporal variation is essential for providing references for regional or global ecological environment sustainability.

Benefiting from the development of satellite remote sensing monitoring technology, the Normalized Difference Vegetation Index (NDVI) has been widely applied to assess vegetation cover characteristic changes. NDVI is closely related to photosynthetically active radiation absorbed by photosynthetic tissues and can serve as an evaluation index for vegetation response to climate change. Increases or decreases in NDVI indicate enhanced or weakened vegetation cover. Currently, widely used NDVI data products include NOAA/AVHRR, SPOT VGT, and MODIS. Among these, MODIS NDVI data with 250 m spatial resolution have attracted considerable attention from researchers due to their complete time series. Previous studies using NDVI to assess vegetation changes in northern China have achieved remarkable results. Research indicates that summer precipitation is the main factor affecting grassland vegetation growth in northern China, and hydrothermal changes caused by meteorological factors such as temperature and precipitation remain key factors influencing vegetation cover changes. However, some studies suggest that temperature has a more significant impact on vegetation growth than precipitation. Through correlation studies between NDVI and climate factors in Northeast China, vegetation cover shows positive correlations with temperature and precipitation but negative correlation with sunshine intensity, with precipitation being the most important influencing factor. Thus, meteorological factors are important determinants of vegetation growth in northern China.

Hulun Buir is located in a high-latitude inland arid region of China. Under global warming, the temperature and precipitation patterns in this region have changed significantly, leading to reorganization of water cycling processes. Various vegetation types in Hulun Buir, including grasslands and forests, respond to climate drivers to varying degrees. Therefore, assessing the adaptability and response mechanisms of Hulun Buir vegetation to climate change is of great significance. Based on MODIS imagery, temperature, and precipitation data from 2000 to 2022, this study employs major methods for vegetation spatiotemporal change research to explore vegetation distribution patterns, change trends, and relationships with climate factors. This research comprehensively evaluates the ecological environment status of vegetation in Hulun Buir, enhances understanding of vegetation cover responses to climate change, and provides scientific basis for vegetation ecological environmental protection, degradation control, and grassland management.

1 Study Area Overview

Hulun Buir is located in northeastern Inner Mongolia, bordering Mongolia and Russia at the intersection of the three countries, and is considered an important ecological barrier in northern China [FIGURE:1]. Hulun Buir covers approximately 75.97% of Inner Mongolia's total area, with forest and grassland accounting for of its area, making it one of the world's largest natural grassland regions. The overall landform units are distributed in belts: the western plateau grassland, central mountain forest, and eastern hilly farmland. Hulun Buir belongs to the mid-temperate and cold-temperate climate zones, characterized by a temperate continental climate with cold and long winters, short and warm summers, and obvious seasonal precipitation distribution. The region has numerous rivers and lakes, including the Ergun River and Hulun Lake, which play important regulatory roles in the regional ecological environment. As the forests, grasslands, and meadows in Hulun Buir are located in mid-to-high latitude temperate regions [FIGURE:1], they respond sensitively to global climate change. Especially under global warming, Northeast China has been considered one of the most important warming regions in China and East Asia.

2 Data and Methods

2.1 Data Sources

The MODIS/MOD13Q1 satellite data with 250 m spatial resolution were obtained from the NASA Earth Observation System. The data product is generated based on the maximum value composition method and has undergone geometric and radiometric corrections. In this study, if monthly mean values were missing, they were replaced by the mean of adjacent years. Annual mean NDVI was calculated as the average of monthly NDVI values, while seasonal mean NDVI represents the monthly averages for spring, summer, autumn, and winter. The meteorological dataset used in this study was obtained from the National Earth System Science Data Center (http://www.geodata.cn), including annual precipitation and mean annual temperature from 2000 to 2022. All data products were processed using Python for projection transformation, boundary masking, and resampling to obtain annual mean precipitation and temperature, elevation, population density, and other data used in this study. Additionally, land use data for Hulun Buir with spatial resolution were downloaded from the Resource and Environmental Data Platform (https://www.resdc.cn) [TABLE:1].

2.2 Research Methods

This study employed unary linear regression analysis to explore the temporal and spatial trends of NDVI in Hulun Buir from 2000 to 2022. The trend change rate (Slope) indicates vegetation change trends, with positive/negative values representing increasing/decreasing trends, and the absolute Slope value serving as an important indicator for determining the significance of vegetation change. Additionally, the Sen+Mann-Kendall trend analysis was applied based on all pixel values in the raster files. This non-parametric statistical method calculates the median of the sequence to detect trends in time series data. Combining the Sen trend and Kendall test can accurately describe data trends without requiring distribution assumptions, being unaffected by missing values and outliers, yielding relatively scientific and credible results. In this study, pixels were classified into degradation, stability, and improvement categories based on Slope values: Slope < -0.0005, -0.0005 ≤ Slope ≤ 0.0005, and Slope > 0.0005. The significance of trends was determined using the Z statistic: if |Z| ≥ 0.196, the trend passed the significance test; otherwise, it did not. Based on these parameters, vegetation cover was further divided into five categories: severe degradation, slight degradation, stable, slight improvement, and significant improvement. Furthermore, the Hurst index was used to analyze the sustainability characteristics and future trend classification of vegetation cover in Hulun Buir, with detailed methods available in Zhang et al. This study also employed spatial analysis based on raster pixel values to conduct partial correlation analysis between annual NDVI and climate factors. Increases or decreases in partial correlation coefficients indicate enhanced or weakened relationships between vegetation growth conditions and climate factors.

3 Results and Analysis

3.1 Spatial Pattern of NDVI in Hulun Buir

From 2000 to 2022, the annual mean NDVI in Hulun Buir showed a distribution pattern gradually decreasing from the central Greater Hinggan Mountains to the eastern hilly areas and western grasslands [FIGURE:2]. The extremely low NDVI values in the southwestern typical steppe pastoral areas (Xin Barag Left Banner, Xin Barag Right Banner, Manzhouli) were mainly due to water body coverage, specifically Hulun Lake in the southwest and the Nierji Reservoir and Nenjiang River basin in the east. Additionally, high-value NDVI distribution areas existed in the north-central part of Ergun City, central-western part of Oroqen Autonomous Banner, northwestern part of Arun Banner, and central-western part of Zhalantun City. Based on raster pixel value statistics, high-value areas accounted for 64.94% of the total region, while low-value areas accounted for 31.74%. The low NDVI values in western Hulun Buir may be related to human activities such as overgrazing, urban expansion, and mining development. Therefore, the impacts of vegetation destruction and degradation processes caused by human activities require attention from relevant authorities.

3.2 Temporal Variation Characteristics of NDVI in Hulun Buir

From 2000 to 2022, vegetation cover in Hulun Buir showed a fluctuating but stable increasing trend overall (Slope = 0.0021, R² = 0.5341, p < 0.01). Temporally, the period can be divided into two phases: 2000—2010, which remained relatively stable (Slope = 0.00007, p > 0.05), and 2010—2022, which showed a significant increasing trend (Slope = 0.0031, p < 0.01) with a slope greater than the overall trend, indicating rapid vegetation cover increase during this period. Additionally, Mann-Kendall trend analysis revealed a significant upward trend in annual NDVI after 2010, further confirming the rapid increase in vegetation cover after 2010 [FIGURE:3]. Seasonally, the vegetation cover change rates from highest to lowest were spring ($0.0031 \cdot a^{-1}$), winter ($0.0021 \cdot a^{-1}$), summer ($0.0019 \cdot a^{-1}$), and autumn ($0.0014 \cdot a^{-1}$), with the spring change rate exceeding the overall rate, indicating that spring vegetation growth is most influenced by external environmental factors. After the "Twelfth Five-Year Plan," the state strengthened the transformation and upgrading of the livestock industry structure, and the rational development of animal husbandry further promoted significant increases in vegetation cover. This demonstrates that after vegetation cover reached saturation, national and government interventions became important factors influencing vegetation cover in Hulun Buir. This insight suggests that ecological protection policies should be adjusted timely according to environmental changes to promote ecological improvement and development.

3.3 Spatial Trend of NDVI Change in Hulun Buir

Except for the central-eastern part of Oroqen Autonomous Banner, southeastern part of Morin Dawa Daur Autonomous Banner, most areas of Arun Banner, northeastern part of Zhalantun City, and small areas in the northern part of Xin Barag Right Banner and local areas of Xin Barag Left Banner, where the change rate was negative, most areas of Hulun Buir showed positive NDVI change rates [FIGURE:4]. The vegetation types in areas with increasing trends are mainly forests and grasslands located in temperate monsoon zones with abundant precipitation and light-heat conditions. Slope significance classification shows that 71.47% of the region maintained stable vegetation cover, benefiting from government emphasis on ecological environmental protection, mainly distributed in central and northern areas including Yakeshi City, Genhe City, Ergun City, and southern Zhalantun City. Detailed statistics are provided in Table 1. In 2011, the state implemented the "Grazing Withdrawal and Grassland Restoration" policy, and various regions strengthened management of northern grasslands, resulting in significant improvements in grassland ecology and resources. Additionally, the State Council issued "Opinions on Promoting Sound and Rapid Development of Pastoral Areas" in 2011, establishing an ecological priority development strategy for Hulun Buir to maintain grassland ecosystem stability, which is the main reason for the significant improvement trend in vegetation cover. However, areas with significant vegetation degradation may be related to human activities, such as point-like grassland degradation in the southwest likely related to coal mining activities, large-scale degradation in central Oroqen Autonomous Banner closely related to urban expansion, and large-scale vegetation degradation in the southeast possibly due to unreasonable agricultural cultivation. The proportion of urban construction land in Hulun Buir showed an increasing trend from 0.407% to 0.484%, indicating that human activities play an important role in influencing vegetation cover.

3.4 Correlation Between NDVI and Climate Factors

The Hurst index for annual mean NDVI in Hulun Buir from 2000 to 2022 ranged from 0.22 to 0.72 with a mean value of 0.54. In all pixel value statistics, 54.23% of pixels had Hurst index values greater than 0.5, indicating that future vegetation cover in Hulun Buir will show overall anti-persistence with a slight increasing trend. In the future trend prediction [FIGURE:6], pixels showing continuous degradation and continuous improvement accounted for 47.37% and 39.32% of the total area, respectively. Continuous improvement areas are mainly distributed in southwestern grasslands and eastern hilly regions, while continuous degradation areas are concentrated in the central Greater Hinggan forest growth zone. Areas with severe vegetation degradation, including central-eastern Oroqen Autonomous Banner and most areas of Arun Banner, will see further improvement in future development trends. In view of this, we should guard against forest vegetation degradation in Hulun Buir and further strengthen management and protection of this region.

Annual precipitation in Hulun Buir from 2000 to 2022 ranged from 170.97 mm to 571.40 mm, with an average of 416.99 mm. Linear regression analysis showed an increasing trend in annual precipitation at a rate of $5.7547 \cdot mm \cdot a^{-1}$. Annual precipitation showed a southwest-east-northeast banded distribution pattern. Due to the north-south orientation of the Greater Hinggan Mountains, terrain conditions block moisture from penetrating into western areas, creating a spatial pattern of higher precipitation in the northeast and lower in the southwest. The mean annual temperature ranged from -5.89°C to 3.91°C, with an average of -0.97°C. Linear regression analysis revealed an increasing trend in mean annual temperature at a rate of $0.0319 \cdot °C \cdot a^{-1}$, consistent with global climate change trends. Low-temperature areas are concentrated in the northern parts of Ergun City and Genhe City, while high-temperature areas are distributed in the northern parts of Xin Barag Left Banner and Xin Barag Right Banner, as well as northern Zhalantun City and Arun Banner [FIGURE:7].

Partial correlation analysis between annual mean NDVI and annual precipitation showed positive correlation coefficients in 70.35% of the study area, with high values (0.24–0.71) mainly concentrated in southwestern grasslands (37.24% of total area), indicating that precipitation strongly influences NDVI variation in these regions. The southwestern grassland terrain slopes from high in the east to low in the west, perennially affected by continental airflows, with the Greater Hinggan Mountains blocking warm moist air from the east, resulting in relatively low precipitation. Additionally, evaporation significantly exceeds precipitation in the southwestern grassland area, making precipitation a major limiting factor for vegetation growth. Partial correlation analysis between annual mean NDVI and temperature showed positive correlation coefficients in 71.5% of the study area, mainly concentrated in the Greater Hinggan forest distribution zone, consistent with high NDVI value distribution areas. Hulun Buir is a typical temperate continental monsoon region where precipitation occurs mainly in summer (accounting for ), with spring and autumn precipitation being secondary and winter the least. Therefore, favorable hydrothermal conditions during summer months benefit the growth of perennial and drought-tolerant herbaceous plants, making it a critical meteorological period for vegetation growth.

To understand the relationship between NDVI and extreme climate events, correlation analysis was conducted between NDVI and maximum/minimum values of temperature and precipitation [TABLE:2]. Results show that annual mean NDVI has significant correlations with temperature indicators, with the strongest correlation with annual minimum temperature (r = 0.59, p < 0.01), moderate correlation with mean annual temperature (r = 0.56, p < 0.01), and weakest correlation with annual maximum temperature (r = 0.41, p < 0.05). However, annual mean NDVI only showed significant correlation with mean annual precipitation (r = 0.39, p < 0.05), but not with precipitation maximum or minimum values. This indicates that temperature, particularly annual minimum temperature, is the important meteorological factor controlling vegetation cover in high-latitude Hulun Buir, while extreme precipitation events may affect vegetation growth and development but result in smaller changes in vegetation cover.

4 Conclusions

Hulun Buir serves as an important ecological barrier in northern China, with its vegetation cover influenced by multiple factors including natural conditions and human activities, making it a natural experimental field for studying interactions between climate factors and human activities. Using Sen+Mann-Kendall trend analysis, correlation, and partial correlation methods, this study explored the spatiotemporal variation characteristics and driving factors of vegetation cover in Hulun Buir from 2000 to 2022, reaching the following conclusions: (1) Spatially, vegetation cover in Hulun Buir decreases gradually from the center to the southwest and east, consistent with the land use pattern of grassland → forest → farmland from west to east. (2) Temporally, vegetation cover showed an overall increasing trend from 2000 to 2022 with an annual change rate of $0.0021 \cdot a^{-1}$. Seasonally, the change rate was highest in spring, followed by winter and summer, and lowest in autumn. The temporal variation can be divided into two phases: 2000—2010 as a stable fluctuation stage ($0.00007 \cdot a^{-1}$) and 2010—2022 as a significant increase stage ($0.0031 \cdot a^{-1}$), demonstrating that national policies have important impacts on vegetation cover after vegetation growth reaches saturation. (3) Most areas in Hulun Buir showed significant improvement trends in vegetation cover, while degradation occurred in the hilly areas of southeastern Morin Dawa Daur Autonomous Banner, urban construction areas in central Oroqen Autonomous Banner, and mineral extraction areas around Manzhouli. Future predictions indicate forest vegetation degradation in the Greater Hinggan region and improvement in southwestern grasslands, requiring more tailored ecological environmental protection policies. (4) Vegetation cover in Hulun Buir responds significantly to climate factors. Spatially, vegetation cover is sensitive to both temperature and precipitation, particularly showing high dependence on precipitation in the southwestern temperate grassland region. Interannually, vegetation cover is more responsive to temperature, especially annual minimum temperature, while showing no significant response to precipitation extremes.

References

[1] Lou J, Xu G, Wang Z, et al. Multi-year NDVI values as indicator of the relationship between spatiotemporal vegetation dynamics and environmental factors in the Qaidam Basin, China[J]. Remote Sensing, 2021, 13(7): 1240, doi: 10.3390/rs13071240.

[2] Ren Jizhou, Lin Huilong. Study on the simulation methods of grassland soil organic carbon: A review[J]. Acta Prataculturae Sinica, 2013, 22(6): 280-294.

[3] Zhao Wanglin, Luo Tianxiang, Zhang Lin. Relative impact of climate change and grazing on NDVI variations in typical alpine desert grasslands in Tibet[J]. Acta Ecologica Sinica, 2019, 39(22): 8494-8503.

[4] Yu Yuyang, Song Fengyi, Zhang Shijie. Quantitative analysis of temporal and spatial changes of NDVI and its driving factors in Henan Province from 2000 to 2020[J]. Ecology and Environment, 2022, 31(10): 1939-1950.

[5] Ren Y, Zhang F, Zhao C, et al. Attribution of climate change and human activities to vegetation NDVI in Jilin Province, China during 1998—2020[J]. Ecological Indicators, 2023, 153: 110415, doi: 10.1016/j.ecolind.2023.110415.

[6] Zhang R, Zhou Y, Hu T, et al. Detecting the spatiotemporal variation of vegetation phenology in northeastern China based on MODIS NDVI and solar-induced Chlorophyll fluorescence dataset[J]. Sustainability, 2023, 15(7): 6012, doi: 10.3390/su15076012.

[7] Li Jingzhong, Xin Zhenhua, Xie Xiao, et al. Spatio-temporal variations of vegetation cover in semi-arid regions and its response to climate change: A case study of Xilin Gol, Inner Mongolia, China[J]. Chinese Journal of Applied Ecology, 2024, 35(1): 80-86.

[8] Li G, Yu L, Liu T, et al. Spatial and temporal variations of grassland vegetation on the Mongolian Plateau and its response to climate change[J]. Frontiers in Ecology and Evolution, 2023, 11: 1067209, doi: 10.3389/fevo.2023.1067209.

[9] Wu Yunli, Zhang Yu, Tian Jiarong. Impacts by climate change and human activities on NDVI in different vegetation types across the Inner Mongolia Plateau[J]. Chinese Journal of Agrometeorology, 2023, 44(12): 1155-1168.

[10] Chen Y, Zhang Y, Zhang H, et al. Extreme temperatures affect NDVI in Karst ecological fragile area[J]. Theoretical and Applied Climatology, 2024, 155(3): 2017-2029.

[11] Li Xitong, Miao Zhenghong, He Longtao, et al. Dynamic change of vegetation coverage of surface water source based on remote sensing technology[J]. Science Technology and Engineering, 2020, 20(6): 2155-2160.

[12] Chang Wenjing, Cong Shixiang, Wang Rongrong, et al. Quantitative analysis of NDVI changes in Mu Us Sandy Land by climate change and human activities[J]. Arid Land Geography, 2025, 48(1): 63-74.

[13] Shi Song, Li Wen, Ding Yishu, et al. Spatiotemporal evolution and influencing factors of vegetation in northeast China[J]. Environmental Science, 2023, 43(1): 276-289.

[14] Bai Yongfei, Zhao Yujin, Wang Yang, et al. Assessment of ecosystem services and ecological regionalization of grasslands support establishment of ecological security barriers in northern China[J]. Bulletin of Chinese Academy of Sciences, 2020, 35(6): 675-689.

[15] Yin Zhenliang, Feng Qi, Wang Lingge, et al. Vegetation coverage change and its influencing factors across the northwest region of China during 2000—2019[J]. Journal of Desert Research, 2022, 42(4): 11-21.

[16] Liang L, Li L, Liu Q. Precipitation variability in northeast China from 1961 to 2008[J]. Journal of Hydrology, 2011, 404(1-2): 67-76.

[17] Xu Xinliang, Liu Jiyuan, Zhang Shuwen, et al. China multi-period land use remote sensing monitoring dataset (CNLUCC)[EB/OL]. [2018]. http://www.resdc.cn/DOI.

[18] Zhang Fengyuan, Su Yuanhang, Liu Binhui. Response characteristics of NDVI to asymmetric diurnal temperature increase and precipitation changes during the forest growing season in northeast China[J]. Journal of Beijing Forestry University, 2023, 45(2): 34-48.

[19] Wu Xinhao. Study on spatiotemporal variation of land cover and ecological quality in the Yellow River Delta[J]. Geographical Science Research, 2023, 12(3): 463-475.

[20] Zhou Lihua, Hou Caixia. Key problems of grassland utilization and the graze ban policy in farming-pastoral ecotone of northern China[J]. Arid Land Geography, 2019, 42(2): 354-362.

[21] Wang Hongbo, Han Aihui. Study on grassland management policy in China[J]. Forest Resources Management, 2015, 24(11): 1799-1807.

[22] Qi Xiaohui, Gao Bo, Wang Haichun, et al. The study on the compensation and award standards for forage-livestock balance and grazing prohibition based on herders' perspective of grassland ecological protection subsidies and incentives policies: Take Xilin Gol League as an example[J]. Journal of Arid Land Resources and Environment, 2016, 30(5): 30-35.

[23] Yin Hang, Zhang Zezhong, Zhang Weijie, et al. Using vegetation health index to calculate spatiotemporal variation in drought and its determinants in Inner Mongolia[J]. Journal of Irrigation and Drainage, 2023, 42(8): 80-89, 105.

[24] Fu Yuanyuan, Zhao Jianjun, Zhang Hongyan, et al. Spatiotemporal variation of vegetation phenology in the Daxing'an Mountains stratified by eco-geographical regions[J]. Chinese Journal of Applied Ecology, 2016, 27(9): 2797-2806.

[25] Wan Huawei, Gao Shuai, Liu Yuping, et al. Grassland degradation monitoring and spatiotemporal variation analysis of the Hulun Buir ecological function region[J]. Resources Science, 2016, 38(8): 1443-1451.