Zhengzhou City is located in the middle north of Henan Province, the boundary between the middle and lower reaches of the Yellow River, and the total area of the city is 7,567 square kilometers. The general trend of the terrain is high in the southwest and low in the northeast (Figure 1), which belongs to the temperate continental monsoon climate, with an average annual temperature of 15°C and an average annual precipitation of 577.70 ∼ 691.60 mm. There are 124 rivers of various sizes in the territory, spanning the two major river basins of the Yellow River and the Huaihe River (Peng et al., 2019). Zhengzhou is the capital city of Henan Province and the core development area of the Central Plains urban agglomeration. At the same time, it is a national central city in central China and an important hub of transportation in China, as well as an important node city for ecological protection and high-quality development in the Yellow River Basin (Cai et al., 2022a).

The land use data from 2000 to 2020, night lights, population density, NDVI and soil type data used in this paper are from the Resource and Environmental Science and Data Center of Chinese Academy of Sciences (https://www.resdc.cn/). Railway, highway and other location data come from the OpenStreetMap (https://openmaptiles.org/). The DEM data come from the Geospatial Data Cloud Platform (https://www.gscloud.cn/), and the slope and slope direction data come from the extraction and analysis of the DEM data. Soil root oxygen content data are from Harmonized World Soil Database (HWSD) (http://webarchive.iiasa.ac.at/Research/LUC/External-World-soil-database/).2030-2050 cmip6 land use data are from Global 0.25° × 0.25° Land-Use Harmonization (LUH2) dataset (https://luh.umd.edu/data.shtml). Precipitation and temperature data are derived from the National Earth System Science Data Center (https://www.geodata.cn/data/) (Table 1).

As the capital of Henan Province, Zhengzhou City is the political, economic and cultural center of the whole province. At present and or a long period of time in the future, Zhengzhou City will be the key to accelerate the urbanization, industrialization and agricultural modernization of the surrounding areas. Understanding the current situation of land use, rationally adjusting the layout of land use, and realizing the intensive and efficient use of land are the inevitable choices to meet the development strategy of Zhengzhou and promote the economic development of the whole city and even the whole province.

The types of land use in Zhengzhou are divided into six categories, with the percentage of land use from largest to smallest being arable land, construction land, forest land, grassland, water area, and unused land. In this paper, ArcGIS is used to extract, cut and classify the data to get the land use status map of Zhengzhou. The analysis on land use changes in 2000, 2010 and 2020 gives Figure 2. From the figure, it can be seen that the construction land with the most obvious changes is mainly distributed in the lots centered on Zhongyuan District, Jinshui District, Guancheng Hui Autonomous District. And during this period, the construction land area has been continuously expanding, presenting a remarkable change from a single center with multiple scattered dots in 2000 to a single center with a radial spider-web-like distribution in 2010 and 2020.

According to the analysis of Table 5, it can be seen that the proportion of cultivated land in Zhengzhou is the largest, but with the passage of time, the proportion of cultivated land shows a trend of decreasing. From 67.44% accounted for in 2000 to 56.98% in 2020, the area decreased by 791.245 km² in 20 years. The area of forest land and grassland also showed different degrees of decline during this period, which decreased by 198.949 km² and 285.746 km², respectively. Water area and construction land showed an upward trend during this period, in which the construction land area increased by 1165.207 km², an increase of 2.38 times. From the land use transfer chord diagram (Figure 3), it can be seen that the increase of these two land use categories mainly comes from the transfer of arable land, which is also closely related to the implementation of the policy of returning fields to lakes and prioritizing economic development in Zhengzhou. The area of unutilized land was zeroed out between 2000 and 2010, which is also related to the intensive economic development of Zhengzhou, the increase in land utilization rate and the strengthening of the degree of utilization.

From the perspective of space, the high-value area of ecosystem carbon storage in Zhengzhou during 2000–2020 is mainly distributed in the western and southern areas of Zhengzhou, that is, the woodland area, while the high-value area in the eastern area is scattered (Figure 7). The medium ecosystem carbon storage region is evenly distributed throughout the whole area, similar to the distribution of cropland in the region. In the past 20 years, with the increase of construction land, the low-value area of ecosystem carbon storage in the study area has gradually increased. The change is more pronounced in the west during the period, with a large number of high-value areas shifting to areas with relatively low ecosystem carbon storage, which coincides with the transfer of forested land to cropland and construction land over the two decades. The distribution of regional ecosystem carbon storage in 2030–2050 is similar to that in the historical period, the high value area decreases and the low value area increases in the SSP245 and SSP585 scenarios, while the high value area increases in the SSP126 scenario. At each time node, the SSP585 scenario had the least high-value area, which corresponded with the least woodland area in this scenario. This also side by side shows the close relationship between land use type and ecosystem carbon storage, there is a strong consistency between the two, and its changes have a significant effect on the ecosystem carbon storage in the study area (Sun X. X. et al., 2023).

From the quantitative point of view, the ecosystem carbon storage in Zhengzhou has decreased by a total of 1.0092 × 10⁷t between 2000 and 2020, and the ecosystem carbon storage in 2020 is 6.0131 × 10⁷t (Figure 8). Under all three scenarios, the ecosystem carbon storage in Zhengzhou in 2050 shows different degrees of decline. Among them, the SSP126 scenario shows the smallest decline, while the SSP585 scenario shows the largest decline, which is 5.7762 × 10⁷t and 5.3464 × 10⁷t, respectively, and decreasing by 0.2369 × 10⁷t and 0.6667 × 10⁷t each from 2020. Looking at the ecosystem carbon storage values, it can be seen that although the SSP126 scenario has the most construction land, it still results in the scenario having the largest ecosystem carbon storage of the three due to the largest increase in its woodland area.

Taking 2020 as the base period and comparing the ecosystem carbon storage at different time nodes from 2030 to 2050 in each scenario relatively, the ecosystem carbon storage increases and decrease map of the research area is obtained. According to Figure 9, it can be seen that the areas of ecosystem carbon storage increase in all three scenarios are around the forest land, and the region of ecosystem carbon storage increase in the SSP126 scenario is the largest among the three. The areas with reduced ecosystem carbon storage are mainly distributed in the western central region of the research area. Due to topographical reasons, this region has a relatively lower level of development compared to the central area of Zhengzhou, and there is more encroachment on it in future development. It is mainly manifested as the erosion of arable land by construction land, and the conversion of forest land to arable land. The areas of decreasing ecosystem carbon storage in the SSP245 and SSP585 scenarios are more concentrated compared to the SSP126 scenario. The areas of increasing and decreasing ecosystem carbon storage in the SSP126 scenario are relatively evenly distributed throughout the study area.

In order to explore the driving factors of ecosystem carbon storage changes in the region, based on the relevant studies and the actual situation in the study area. The study selected seven influence factors from both natural and social aspects, which are natural factor DEM (X3), annual average temperature (X4), annual precipitation (X5), and NDVI (X6), social factors GDP (X1), population density (X2), night lighting (X7) (Zhang et al., 2024). The ArcGIS software is used to create a fishing net, which is used to extract the influencing factor data, and discretize the data to meet the operational needs of the Geographic detector (Li P. C. et al., 2023). The results of the single-factor analysis of the geographic detector are shown below (Table 6), and the p-values of the seven driving factors are all less than 0.001, which passes the test of significance and is able to accurately explain the changes in ecosystem carbon storage in Zhengzhou City. The degree of influence of the factors on ecosystem carbon storage in the study area in descending order is annual mean temperature (X4), DEM (X3), NDVI (X6), night light (X7), GDP (X1), population density (X2), annual precipitation (X5). Among them, the explanatory power of annual average temperature (X4), DEM (X3) and NDVI (X6) are all greater than 0.2, which is the main reason for the change of ecosystem carbon storage in the study area. Night light (X7), GDP (X1), and population density (X2) have relatively low explanatory power and are minor influencing factors. It can be seen that natural factors play a dominant role in explaining the spatial and temporal distribution of ecosystem carbon storage.

Through the analysis of the interactive detection results, we can see that all the interactive factor detection results are double-factor enhancement or nonlinear enhancement, indicating that the distribution of carbon storage is affected by many factors (Figure 10). The interactions between dem (X3) and mean annual temperature (X4) and factors were all greater than 0.29. And the interactive explanatory power between DEM (X3) and annual precipitation (X5) is the largest, which is 0.35736, followed by dem (X3) and NDVI (X6), which is 0.34876. Changes in ecosystem carbon storage are the result of both natural and anthropogenic effects, and the interaction between various factors should be fully considered in the design and planning, taking into account both natural and anthropogenic factors in a comprehensive manner. Different targeted development plans should be formulated for regions with different natural conditions to ensure sustained and coordinated regional development.

By mining the drivers of land use change, it was found that DEM and slope contributed the most to land use change, especially to the distribution of woodland and grassland. Higher elevations and slopes have a relatively lower intensity of human activity, and the lower the level of disturbance to woodlands and grasslands, the more favorable it is for its growth and expansion. In addition, GDP is also a major driver of changes in cropland, forest land and construction land, with the smallest area required to produce a unit of GDP being construction land, presenting the phenomenon of high GDP requiring a high level of building area, which leads to the emergence of the erosion of arable land and other areas by construction land. Population density is also a major potential driver of change in woodland, watersheds, and construction land, with higher population densities increasing demand for housing, workplaces, etc., which in turn leads to an expansion of construction land and a squeeze on woodland and other areas (Figure 11). From the perspective of land use change, from 2000 to 2020, the arable land area of Zhengzhou decreases from 5103.204 km² to 4311.959 km², the construction land area increases from 843.915 km² to 2009.122 km², and the construction land is constantly expanding outward with the built-up area as the center. Ecosystem carbon storage showed a continuous downward trend from 7.0223 × 10⁷t to 6.0131 × 10⁷t during the two decades, and the conversion of arable land with strong carbon sequestration capacity to construction land with weak carbon sequestration capacity was the main reason. The future increase of construction land area in Zhengzhou is limited and the increase rate tends to slow down, especially in the center of Zhengzhou (Cai et al., 2022b). Future urban additions are more likely to occur in the western part of Zhengzhou, derived from encroachment on arable, grassland, among others.

Comparing the ecosystem carbon storage values of Zhengzhou under the three scenarios, it can be seen that in the future, if Zhengzhou city wants to minimize the loss of ecosystem carbon storage while ensuring economic development. It should be based on the SSP126 scenario, the sustainable utilization of natural resources should be strengthened, an ecological restoration planning system should be gradually constructed, and a carbon-sequestration and sink-enhancement mechanism led by the government, with the participation of many parties and local linkages should be established, so as to mitigate and adapt to climate change. Pay attention to the protection of ecological land, wide entry and strict exit. Intensive and economical use of land, delineation of the red line of ecological protection and strict management of urban development boundaries, reduction of the negative impact of man-made activities on regional ecology, implementation of the regulation of mountainous, forested and grassland areas, and giving full play to the agglomeration effect and territorial advantages. Strictly follow the territorial spatial planning scheme, unified implementation of territorial space use control, at the same time should also pay attention to low-carbon oriented land use pattern optimization, strengthen ecological protection, and restoration to actively play the carbon sink function of ecological land (Cao and Zhang, 2025).

The land use simulation and carbon stock prediction in the study area are in agreement with the trend of Li et al. study from the perspective of Henan Province as a whole (Li et al., 2024). The evolution of land-use patterns all show the expansion of construction land and the decrease of ecological land. The regional carbon stocks are all highest in the SSP126 scenario, followed by the SSP245 scenario. Furthermore, the trend of change obtained in the study is also similar to that in the study of Shaanxi Province, which is adjacent to the study area by Cheng et al. (2024). The change trend of ecosystem carbon storage in the future obtained by the study is the same as that of Sun F. H. et al. (2023), but the value of ecosystem carbon storage has some differences. This is mainly due to differences in the methods used to determine future land-use requirements. Sun et al. set two scenarios of natural development and ecological protection, while this paper uses three different scenarios provided by CMIP6 to make predictions. In addition, it is known through the previous studies that the dead organic matter of vegetation in an area is equal to 1/10th of the aboveground carbon pool in that area (Delaney et al., 1998; Fan et al., 2023). The article integrates and complements the carbon density data, and thereby determines the carbon density data for the study area, based on the previous study (Sun Y. F. et al., 2023). These two reasons lead to discrepancy in the ecosystem carbon storage values obtained by the two studies.

The results of the geographic detector show that the average annual temperature, DEM and NDVI have significant effects on the spatial differentiation of ecosystem carbon storage in the study area, and the driving factors such as population density, GDP and NPP have certain effects on the spatial differentiation of ecosystem carbon storage. Suitable temperature is suitable for plant growth, and certain topographic relief can reduce the interference of human activities, and then form a high vegetation cover area. NDVI reflects the extent of vegetation cover. Areas with high vegetation cover (such as woodland, grassland) have better carbon sequestration capacity than other land types (such as construction land). The interaction analysis shows that natural factors and human factors interact together to enhance the influence on the spatial distribution of ecosystem carbon storage, which indicates that the spatial differentiation of ecosystem carbon storage is the product of the interaction of natural environment and human activities. Different factors have different directions of action, and the future development of Zhengzhou city will be governed differently based on the characteristics of the factors, taking into account the actual situation. While comprehensively considering the natural conditions and socio-economic development of the region in order to ensure that the selected model fits the characteristics of the region, it is highly vigilant in preventing unreasonable human interventions from synergizing with other factors and exerting excessive pressure on the regional ecosystems, so as to achieve sustainable land use and ecological conservation (Zhang et al., 2024).

Although the PLUS model used in the article is higher in simulation accuracy than previous models such as Flus model, there exist a number of parameters that need to be determined in conjunction with historical data and expert opinion. In addition, the carbon pool used in the study is supplemented and improved on the basis of previous studies, and there is no field sampling, which may have some differences between the carbon density and the actual study area. Therefore, in the subsequent study, we can try to use fuzzy logic or Bayesian method for reference to quantify the uncertainty of the model (Mutale and Qiang, 2024).