Spatiotemporal Evolution Characteristics and the Climatic Response of Carbon Sources and Sinks in the Chinese Grassland Ecosystem from 2010 to 2020

Abstract

With the increase in global carbon dioxide emissions, China has put forward the goals of a carbon peak and carbon neutrality (double carbon) and formulated an action plan to consolidate and enhance the carbon sink capacity of the ecosystem. The Chinese grassland ecosystem (CGE) is widely distributed and is the key link for China to achieve the double carbon objectives. However, there is a relative lack of research on carbon sources and sinks in the CGE, so it is urgent to integrate and analyze the carbon sources and sinks in the grassland ecosystem on the national scale. Based on the refined grid data, the net ecosystem productivity (NEP) of the CGE was estimated by coupling the vegetation production model and soil respiration model. The results showed that the cumulative carbon sequestration of the CGE was 14.46 PgC from 2010 to 2020. In terms of spatial distribution, this shows that the differentiation characteristics are high in the northwest of China and low in the southeast of China, which strongly corresponds with the 400 mm isohyet and 0 °C isotherm of China. The results of the correlation analysis showed that the NEP of the CGE was positively correlated with precipitation and negatively correlated with temperature; that is, precipitation mainly promotes the accumulation of NEP, and temperature mainly inhibits it. The coupling effect of temperature and precipitation jointly affects the spatial change of carbon sources and sinks of the CGE. This study can provide a scientific basis for government departments to formulate targeted policies to deal with climate change, which is of great significance for China to improve ecosystem management, ensure ecological security and promote the realization of China’s double carbon goal.

1. Introduction

Since the 21st century, global environmental change and sustainable development have become two important scientific issues and challenges in the world [1,2]. Carbon dioxide (CO₂) is one of the main greenhouse gases (GHGs), and its increasing emissions year by year have a great impact on global climate change, triggering a series of environmental problems, such as rising temperatures, melting glaciers and frequent extreme weather. According to the Global Energy Review: CO₂ Emissions in 2021 released by the International Energy Agency, the energy-related CO₂ emissions grew to 36.3 Gt in 2021 [3]. In order to cope with global warming and explore and solve the increasingly prominent problem of climate change, governments and researchers all over the world are actively taking measures and putting forward a series of important scientific issues and research directions, such as ″can we stop global climate change?″ and ″where do we put all the excess carbon dioxide?″ [4,5,6].

In 2015, nearly 200 parties to the United Nations Framework Convention on Climate Change reached the Paris Agreement at the Paris climate change conference. The Paris Agreement unified arrangements for global action against climate change after 2020 and set the goal of achieving net zero emissions in the second half of this century [7]. In order to implement the Paris Agreement, China incorporated the goal of addressing climate change into the 14th Five-Year Plan in 2020 [8,9]. In September of the same year, China reiterated its commitment at the United Nations General Assembly to ″strive to achieve carbon peak in 2030 and carbon neutrality in 2060″, and then determined China’s action plan for a carbon peak and carbon neutrality [10]. The plan clearly pointed out the need to ″improve the quality and stability of the ecosystem and increase the carbon sink of the ecosystem″ [11]. Strengthening the protection and restoration of the ecosystem and implementing the consolidation and improvement of the ecosystem’s carbon sequestration capacity have become the key measures for China to achieve the double carbon goal and an effective way to slow down global warming.

A terrestrial ecosystem (TE) has the functions of fixing CO₂, purifying air, maintaining biodiversity, preventing wind and fixing sand, soil and water conservation, etc. [2,12]. The carbon cycle can comprehensively reflect the response of the ecosystem to climate change and the impact of human activities and has become one of the important aspects of global climate change research [4,13]. The TEs of China are complex, diverse and widely distributed. By 2021, the area of TEs accounted for about 87.14% of the total land area in China, of which about 56.1% was grassland ecosystems (GEs), and the carbon storage of the GE was 41.67 PgC, which plays an important role in the TE [14,15]. Therefore, estimating the carbon sources and sinks of the CGE can help to provide a scientific basis for government departments to formulate targeted policies to deal with climate change. At the same time, it is of great significance for China to improve ecosystem management, ensure ecological security and promote the realization of China’s double carbon goal.

In recent years, there have been many studies estimating carbon sources and sinks in ecosystems, with various methods. Gao et al. [16] searched the core collection database of the Web of Science for studies on grassland carbon sequestration from 2010 to 2019 through CiteSpace software. The results showed that the research hotspots of grassland carbon sequestration in recent years are mainly ″climate change″, ″sequestration″ and ″soil organic carbon″. After 2016, the research on ″ net primary productivity″ (NPP), ″ecosystem CO₂ exchange″ and ″carbon balance″ gradually increased. Based on the NPP data, Zhang et al. [17] estimated the carbon storage and carbon sink of the GE in Sanjiangyuan and analyzed the spatial distribution pattern and change trend of carbon storage and carbon sinks of the GE in Sanjiangyuan. Pan et al. [18] used the CASA model to analyze the NPP of the TE in the arid region of northwest China from 2001 to 2012 through linear trend analysis, standard deviation analysis and the Hurst index method and concluded that the NPP of the ecosystem in the arid region of northwest China has a strong relation with seasonal variation and divided the regions of NPP changes based on a stability analysis in the study area. Based on MODIS-Normalized vegetation index data, Yun et al. [19] estimated the carbon sinks of the GE in the Shiyang River Basin, analyzed the spatial distribution characteristics of the carbon sources and sinks and divided the carbon source area and carbon sink area.

Although the above studies have provided reliable methods, data and results for the research on ecosystem carbon sources and sinks, the following problems still need to be further discussed and solved. In terms of research scale, the current studies on carbon sources and sinks in the CGE are mostly concentrated in Inner Mongolia, Qinghai and Xizang, and the results of carbon sources and sinks in the GE estimated by different study methods are also quite different. There is an urgent need for an integrated analysis of carbon sources and sinks in the GE at the national scale, which is not only the only way for China to achieve the double carbon goal but is also urgently needed for the effective global management of GHGs and an active response to climate change. In terms of research methods, the important role of soil carbon sinks in the GE has not been fully considered. The research shows that most of the carbon in the GE is stored in soil, and the carbon cycle of the GE mainly depends on soil carbon sinks. In terms of research content, there are no estimation results of the spatiotemporal evolution characteristics of carbon sources and sinks in the CGE in a long-time series at a fine scale, and there is a lack of a horizontal comparative analysis of carbon sources and sinks in the CGE, as well as the exploration of different regional difference mechanisms, which cannot provide reliable support for the state to formulate restoration strategies, quality improvement and carbon increase measures of the GE according to local conditions.

Net ecosystem productivity (NEP) refers to the net absorption or net storage of carbon in the ecosystem, which is used to quantitatively describe the carbon source and sink capacity of the ecosystem [20,21]. When NEP > 0, it indicates that the ecosystem plays the role of a carbon sink; otherwise, it is a carbon source. As a key parameter to characterize vegetation activities, the accurate estimation of NEP is helpful not only to measure the health degree of the ecosystem, but also to quantitatively analyze the carbon sequestration status and potential of the regional ecosystem. In this study, the gross primary productivity (GPP) and soil respiration of the GE were comprehensively considered. The remote sensing inversion datasets of temperature, precipitation and photosynthetically active radiation were obtained by using GEE tools. The NPP of the CGE was calculated by using the vegetation production model (VPM). Then, based on the soil respiration model (SRM), we constructed an NEP estimation model for the fine grid-scale of the GE in China and discussed the spatial and temporal characteristics of carbon source and sink distribution and the interannual variation trend of the GE in different regions. Furthermore, this study established a spatial analysis model based on pixels to quantitatively analyze the coupling response between the spatiotemporal evolution of NEP and climate change in GE, and further revealed the internal driving factors of regional differences in the CGE. It should be noted that although human activities, topographical factors and surface evapotranspiration play an important role in the carbon cycle of ecosystems, the focus of this study is to calculate the NEP of vegetation and soil respiration in CGE, and they have not been deeply considered in this study.

The main objectives and innovations of this study are as follows: Based on the fine grid data, a national scale carbon source and sink estimation model was established. In this study, the spatiotemporal variability and evolution trend of carbon sources and sinks in the CGE were evaluated, the stability of carbon sources and sinks in the CGE was analyzed and the response mechanism of carbon sources and sinks to different meteorological factors was discussed by analyzing the correlation between NEP and temperature and precipitation in the CGE. The results of this study can provide a scientific basis for decision-making departments in the Chinese government to formulate differentiated policies to improve the carbon sink capacity of the GE according to local conditions, which is of great significance for China to achieve the double carbon objectives and also to provide effective technical methods and data support for further research by other scholars in this field.

2. Study Area and Data Acquisition

2.1. Study Area

China has some of the richest grassland resources in the world. The total area of the GE is about 2.86 million km², accounting for 5.71% to 9.34% of the grassland area in the world and containing 3.59% to 15.98% of global grassland carbon [14]. The CGE ranges from the northeast of the Greater Khingan Mountains to the south of the Qinghai–Tibet Plateau (Figure 1). Due to its wide distribution, the CGE spans five temperature zones: the tropical zone, subtropical zone, temperate zone, subtemperate zone and frigid temperate zone. Among them, the northwest region is densely distributed and is mainly grassland, accounting for about 84% of the total area of the CGE. The distribution in southeast China is relatively sparse, mainly dominated by meadow species, accounting for about 16% of the total area of the CGE [22]. Based on the national development needs, this study systematically calculated the spatial and temporal distribution patterns and evolution trends of carbon sources and carbon sinks in the CGE using fine-grained grid data. This study explores the response relationship between spatiotemporal variability and climate change in order to provide a reliable implementation path for China to achieve a carbon peak and carbon neutrality.

2.2. Data Acquisition

The grassland ecosystem boundary of China in this study was extracted from GlobeLand30 2020 [23].

The data used to calculate the NEP of the CGE included land surface water index (LSWI), photosynthetically active radiation (PAR), enhanced vegetation index (EVI) temperature and precipitation. LSWI was calculated with MOD09A1 V6 surface reflectance data on the GEE platform with reference to the study of Xiao et al. [24]. The Moderate Resolution Imaging Spectroradiometer (MODIS) sensor onboard the NASA EOS Terra satellite has visible, near-infrared and shortwave infrared bands, and the principle of this method is to take into account that the absorption rate of vegetation in the near-infrared band is low, while that in the short infrared band is high.

PAR is the MCD18A2 product, and it represents a part of the solar radiation spectrum from 0.4 to 0.7 µm that is absorbed, transferred and stored within ecosystems [25]. The data were generated by the Terra and Aqua MODIS combined surface radiation algorithm. The algorithm starts by estimating surface reflectance from a set of multi-temporal data inputs and evaluates the PAR flux for each of them.

EVI is provided by MOD13A2 V6 products to select the best available pixel values from all collections over a 16-day period [26]. It is calculated by bidirectional surface reflectance with atmospheric correction and screened for water, clouds, heavy aerosol and cloud shadow. A blue band was used to remove EVI smoke and sub-pixel thin clouds caused by residual air pollution. The standards for use were low clouds, low viewing angles and highest EVI values.

Temperature data were from the National Earth System Science Data Center (http://www.geodata.cn, accessed on 15 March 2020), which is based on monthly average temperature data from 824 benchmark meteorological stations on the ground in China from 2010 to 2020. The monthly grid data with a horizontal resolution of 1 × 1 km in China were generated by spatial interpolation using a DEM with a resolution of 1km in China as a covariable. After data acquisition, the monthly temperature was superimposed and averaged to obtain the annual average temperature dataset.

The meteorological data used in this study were temperature and precipitation. The temperature data were consistent with the temperature data for calculating NEP. Precipitation data were from the National Earth System Science Data Center. (http://www.geodata.cn, accessed on 15 March 2020) These data were generated for the China area by downscaling the delta spatial downscaling scheme based on the global 0.5° climate data released by CRU and the global high-resolution climate data released by WorldClim. Data from 496 independent meteorological observation points were used to verify the reliability of the results.

The vector data of Chinese administrative regionalization used in this study were from the National administrative regionalization information query platform. (http://xzqh.mca.gov.cn/map, accessed on 15 March 2020). The data are mainly used for statistical analysis and horizontal comparison among administrative divisions.

In this study, all obtained data were preprocessed with ArcGIS 10.5 software for the unified coordinate system and spatial resolution. The data coordinate system was converted to GCS_WGS_1984. The datum was D_WGS_1984, the prime meridian was Greenwich and the angular unit was degrees. The spatial resolution of all raster data was 500 × 500 m. The null value region was spatially interpolated or assigned a value of 0 depending on the type and meaning of the data.

3. Methodology

3.1. Research Framework

In this study, the following steps were used to analyze the spatiotemporal evolution characteristics and driving factors of carbon sources and sinks in the CGE:

• Step 1: NEP estimation of the CGE. Based on the VPM and SRM, the NEP of the CGE on interannual grid scale from 2010 to 2020 was estimated;

• Step 2: Analysis of the spatial distribution characteristics of carbon sources and sinks in the CGE. The spatial distribution characteristics of carbon sources and sinks of the CGE from 2010 to 2020 were analyzed from the grid scale and provincial administrative division scale;

• Step 3: Analysis of the spatiotemporal evolution trend of carbon sources and sinks in the CGE. By constructing the univariate linear regression equation and variation coefficient formula of NEP interannual variation in the CGE, the interannual variation rate and spatiotemporal variation rate with its spatial grid were calculated, and the spatiotemporal evolution trend and stability of carbon sources and sinks in its ecosystem were discussed.

• Step 4: Study of the response relationship between NEP and climate change in the CGE. By calculating the correlation coefficient and partial correlation coefficient between NEP and temperature and precipitation, this study explores the correlation mechanism between NEP and meteorological elements in the CGE and analyzes the response process of NEP to climate change in the CGE.

3.2. Estimation Model of NEP

NEP is an important indicator to measure the carbon sequestration capacity of the ecosystem [27,28]. The calculation of the NEP of the GE can more intuitively reflect the carbon cycle relationship between ecosystem, soil and atmosphere. In this study, the VPM and SRM were used to estimate the NPP and carbon consumption of soil microbial respiration (R_H) in the CGE on an interannual scale, and NEP was obtained by subtracting R_H from NPP [29,30]. When NEP > 0, it indicates that there is net carbon sequestration in the ecosystem; that is, it belongs to carbon sink area. When NEP < 0, it means that the carbon sink of the ecosystem is less than the carbon source, which belongs to the carbon source area.

$NEP\left(x,t\right)=NPP\left(x,t\right)-R_H\left(x,t\right)$

where NEP (x, t) is the net ecosystem productivity of grid x in year t; NPP (x, t) is the net primary productivity of grid x in year t and R_H (x, t) is the soil microbial respiration of grid x in year t.

NPP was obtained based on GPP calculated using the VPM [31,32,33,34]. The description of the VPM can be found in the Supplementary Materials.

$NPP\left(x,t\right)=GPP\left(x,t\right)\times r_{NPP/GPP}$

where GPP (x, t) is the total primary productivity of grid x in year t, and r_NPP⁄GPP is the carbon utilization rate, which represents the efficiency of vegetation in the ecosystem for converting productivity into biomass and storing it in the ecosystem.

R_H was obtained from the regression equation of soil carbon emission established by meteorological data such as temperature and precipitation [35].

$R_H\left(x,t\right)=0.22\times \left(\mathit{exp}\left(0.0912·T\left(x,t\right)\right)+\mathit{ln}\left(0.3145·P\left(x,t\right)+1\right)\right)\times 30\times 46.5\%$

where T (x, t) is the average annual temperature of grid x in year t, and P (x, t) is the average annual precipitation of grid x in year t.

3.3. Interannual Variation Rate of NEP

In order to analyze the spatiotemporal variation characteristics of carbon sources and sinks in the CGE, a univariate linear regression equation was constructed to analyze the fluctuation law and spatial variation characteristics of NEP. The interannual variation rate of NEP in the CGE was obtained by calculating the slope of the multi-year regression trend line grid by grid [36].

${\theta }_{slope}=\frac{n\times {{\displaystyle \sum }}_{i=1}^{n}i\times NE{P}_i-{{\displaystyle \sum }}_{i=1}^{n}i·{{\displaystyle \sum }}_{i=1}^{n}NE{P}_i}{n\times {{\displaystyle \sum }}_{i=1}^n{i}^2-{({{\displaystyle \sum }}_{i=1}^{n}i)}^2}$

where n represents the number of years. The estimated years of this study were 2010–2020; that is, n was 11. NEP_i is the annual average NEP of year i. θ_Slope is the slope of the trend, θ_Slope > 0 indicates that NEP increases year by year; otherwise, it decreases year by year.

According to the calculation results, the CGE was divided into five areas in this study: θ_Slope < −5 indicates significant deterioration, −5 ≤ θ_Slope < −1 indicates slight deterioration, −1 ≤ θ_Slope < 2 indicates remaining unchanged, 2 ≤ θ_Slope < 6 indicates slight improvement and θ_Slope ≥ 6 indicates obvious improvement.

3.4. Stability Analysis of NEP

The coefficient of variation equation based on grid scale was established to measure the stability of NEP in the CGE from 2010 to 2020 [37].

$C_v=\frac{\sigma }{\overline{x}}$

where σ is the standard deviation; $\overline{x}$ is the average value and C_V is the coefficient of variation, which reflects the degree of dispersion of NEP. The smaller the C_V, the less the data fluctuate with time and have high stability, and vice versa.

According to the calculation results, the CGE was divided into five areas: 0 ≤ C_V ≤0.5 indicates low, 0.5 < C_V ≤ 1.5 indicates relatively low, 1.5< C_V ≤ 2.5 indicates medium, 2.5 < C_V ≤ 5 indicates relatively high and C_V > 5 indicates high.

3.5. Correlation between NEP and Climate Factors

In this study, in order to further explore the correlation mechanism between the spatiotemporal change law of NEP and the change of climate factors in the CGE, the spatial analysis method based on the spatial grid, including the correlation coefficient and partial correlation coefficient, was used to calculate the correlation between the change in the NEP spatial pattern and temperature and precipitation and to analyze the response mechanism between NEP and climate change.

The correlation coefficient was calculated by the following formula [38,39]:

$R_{xy}=\frac{{{\displaystyle \sum }}_{i=1}^n\left[\left(x_i-x_p\right)\left(y_i-y_p\right)\right]}{\sqrt{{{\displaystyle \sum }}_{i=1}^n{\left(x_i-x_p\right)}^2{{\displaystyle \sum }}_{i=1}^n{(y_i-y_p)}^2}}$

where R_xy represents the correlation between variables x and y; x_i is the NEP in year i; y_i is the value of variable y in year i; x_p is the average value of NEP from 2010 to 2020; y_p is the average value of variable y from 2010 to 2020 and n is the estimated number of years.

The partial correlation coefficient was obtained by the following formula to analyze the response of NEP to another variable when one variable in precipitation or temperature was fixed [40].

$r_{xy,z}=\frac{r_{xy}-r_{xz}{r}_{yz}}{\sqrt{{\left(1-r_{xz}\right)}^2{\left(1-r_{yz}\right)}^2}}$

where r_xy, r_xz and r_yz represent the cross-correlation coefficient between the variation x and y, x and z and y and z, respectively, and r_xy,z is the cross-correlation coefficient between x and y when the value of z is fixed.

In addition, the T-test method was used to test the significance of the correlation coefficient and partial correlation coefficient of each variable [41]:

$t=\frac{r_{xy,z}}{\sqrt{1-r_{xy,z}^2}}\sqrt{n-m-1}$

The significance level was taken as α= 0.05, and the degree of freedom was 11. According to the significance level, results were divided into significant positive correlation ($R$ > 0, p < 0.05), insignificant positive correlation ($R$ > 0, p ≥ 0.05), significant negative correlation ($R$ < 0, p < 0.05) and insignificant negative correlation ($R$ < 0, p ≥ 0.05). Then, the CGE was divided into four different areas according to different significance levels.

4. Results and Analysis

4.1. Spatial Distribution Characteristics of Carbon Sources and Sinks in the CGE

Figure 2 shows the spatial distribution of the perennial average NEP of the CGE from 2010 to 2020. As can be seen from Figure 2, the spatial distribution pattern of the CGE could generally be characterized as high in the northwest of China and low in the southeast of China, but the carbon sink capacity of the CGE showed an opposite trend. The annual average NEP value of the CGE from 2010 to 2020 was 143.71 gC·m⁻²·a⁻¹. The annual average carbon sequestration was 411.48 TgC/a. Figure 3 shows the perennial average distribution of NEP in the CGE of different provinces in China from 2010 to 2020. As can be seen from Figure 3, Guizhou province had the highest average NEP in China with a value of 478.60 gC·m⁻²·a⁻¹, followed by Yunnan Province (381.21 gC·m⁻²·a⁻¹) and Chongqing Municipality (359.377 gC·m⁻²·a⁻¹). Shanghai had the lowest NEP, with an annual average of 51.39 gC·m⁻²·a⁻¹ from 2010 to 2020. The perennial mean NEP for other regions is shown in Figure 3.

Figure 4 shows the spatial distribution of the cumulative NEP of the CGE from 2010 to 2020. It can be seen from Figure 4 that the cumulative NEP value of the CGE from 2010 to 2020 was 5051.27 gC·m⁻², and the cumulative carbon sequestration was 14.46 PgC. Based on the accumulated NEP values of the CGE from 2010 to 2020, carbon source regions and carbon sink regions of different provinces in China were further extracted, as shown in Figure 5. In Figure 5, the carbon sink region of the CGE (NEP > 0) from 2010 to 2020 was about 2.45 millions km², accounting for 85.82% of the total area of the CGE. The cumulative NEP was 5189.16 gC·m⁻², and the total carbon sequestration was 12.75 PgC. The carbon source region of the CGE (NEP < 0) covered 0.46 million km², accounting for 14.18% of the total area of the CGE, with a cumulative NEP of −304 gC·m⁻² and total carbon emissions of 0.12 PgC.

4.2. Interannual Variation of Carbon Sources and Sinks in the CGE

Figure 6 shows the general trend of the interannual change of NEP of the CGE from 2010 to 2020. It can be seen from Figure 6 that the average annual NEP of the CGE from 2010 to 2020 increased year by year. However, the NEP decreased from 2014, and reached the maximum decline in 2015, which was 13.66 gC·m⁻² lower than that in 2013. The main reason may be the decrease in precipitation from 2014 to 2016; the annual precipitation of the CGE was 416.19 mm in 2013, 387.49 mm in 2014 and 386.84 mm in 2015. The precipitation gradually recovered to the level of 2013 after 2016. The decrease in precipitation led to the reduction in area, the weakening of soil respiration, the reduction in primary productivity, the increase in grazing pressure and the increase in fire prevalence, which led to a decrease in NEP. From 2018 to 2020, the interannual mean value of NEP in the study area recovered and remained stable at the level of 2013.

Figure 7 shows the grid-by-grid spatial distribution of the interannual change rate of NEP in the CGE. As can be seen from Figure 7, the interannual change rate of NEP in the CGE showed trend of obvious improvement for a large area, accounting for about 45% of the total area of the CGE. About 15% of the total area of the CGE had a trend of slight improvement. About 22% of the total area of the CGE had a trend of remaining unchanged. About 7% of the total area of the CGE had a trend of slight deterioration, and about 8% of the total area of the CGE was in significant deterioration. It can be seen that the annual average NEP change in the CGE tended to have good prospects, and the overall trend saw an increase.

Affected by climate factors and policies related to environmental protection, the obvious improvement regions were mainly between the 200 mm isohyet and the 400 mm isohyet in China, and there was also more distribution in the northern part of Xinjiang. In addition to obvious improvement and slight improvement areas, significant deterioration areas could also be found in the southern grassland of China which is south of the 400 mm isohyet. In northwest China to the north of the 200 mm isohyet, most of the grasslands distributed on the Qinghai–Tibet Plateau and the western basin were in the range of slight deterioration and have remained unchanged under the influence of many factors, such as the climate factor.

Figure 8 shows the proportions of different trends in the annual average NEP of the CGE in different provinces. It can be seen that the two provinces with the largest proportion of obvious improvement areas were Inner Mongolia and Sichuan, both accounting for 72% of their respective grassland area. The provinces and municipalities with a large proportion of obvious improvement were all located in northwest China. The province or municipality with the most areas of slight improvement was Xizang (32.08%), followed by Qinghai. Macao had the largest proportion of regions remaining unchanged (100%), followed by Hong Kong (72.06%) and Shanghai (69.62%). The larger the proportion of unchanged, slight improvement and obvious improvement regions, the higher the degree of environmental protection in the region to some extent. The proportion of slight deterioration in all regions of the study area was less than 20%, but the proportion of significant deterioration was not low. The proportion of significant deterioration regions exceeded that of obvious improvement regions in the follow provinces: Beijing, Fujian, Guizhou, Hainan, Hubei, Hunan, Jiangxi, Taiwan, Tianjin, Hong Kong, Zhejiang and Chongqing. If the number of significant deterioration regions exceeds that of obvious improvement regions for a long time, it will lead to the decline of NEP and the carbon sink capacity of the region, which is not conducive to sustainable development and even worse for the realization of the double carbon goal.

4.3. Spatiotemporal Variability of Carbon Sources and Sinks in the CGE

In this study, the variation coefficient of NEP was further calculated to analyze the stability of the spatial variation of NEP. The greater the coefficient of variation is, the more frequent and volatile the NEP value is. Figure 9 shows the spatial distribution of the NEP variation coefficient in the study area. It can be seen from Figure 9 that the mean variation coefficient of NEP in the CGE was 0.027, which shows good stability. Most areas in the study area were at a ″low″ degree of variation, which means high stability, accounting for about 82.42% of the total grassland. They were mainly distributed in areas west of 400 mm isohyet, mainly in the Inner Mongolia steppe and the Qinghai–Tibet Plateau, which are less affected by human activities. The ″relatively low ″ degree of variation accounted for 11.04% of grassland, mainly distributed near the 400 mm isohyet and partially distributed in the south areas. The areas with a ″medium″ degree accounted for 2% of grassland and were distributed near the 400 mm isohyet. The area with a ″relatively high″ degree of variation accounted for about 3%, and the distribution was scattered in a small range in all regions. Regions with ″high″ levels of variation accounted for about 0.2%.

4.4. Correlation between the Spatiotemporal Evolution of Carbon Sources and Sinks and Climatic Factors in the CGE

Figure 10 shows the correlation coefficient and partial correlation coefficient between NEP and climatic factors (temperature and precipitation) in the CGE. It can be seen from Figure 10 that the average correlation coefficient between the NEP and precipitation in the CGE was 0.09, and the NEP of the CGE was positively correlated with precipitation. The regions with a significant positive correlation between NEP and precipitation accounted for 1.60% of the total area of the CGE, mainly located in the Qinghai–Tibet Plateau and Inner Mongolia grassland. The NEP of the CGE in these regions increased with the increase in precipitation. The regions with a significant negative correlation between NEP and precipitation accounted for 35.27% of the total area of the CGE. In these regions, with the increase in precipitation, soil respiration carbon consumption increased and the NEP of the CGE decreased. The average correlation coefficient between NEP and temperature in the CGE was −0.06, and the NEP of the CGE was negatively correlated with temperature. The regions with a significant negative correlation between NEP and temperature accounted for 53.72% of the total area of the CGE, which were mainly located in the south of China, indicating that the increase in temperature hinders the accumulation of NEP. This may be because the increase in temperature reduces the relative humidity, and the transpiration of vegetation will remain high, resulting in the non-synchronization of water and heat required for vegetation growth, resulting in less total NPP accumulated in the grassland ecosystem. At the same time, the increase in temperature increases soil respiration and consumes a large amount of organic carbon, so the NEP of the CGE decreases. The regions with a significant positive correlation between NEP and temperature accounted for 2.25% of the total area of the CGE and were mainly distributed in the north of Qinghai, Gansu, Inner Mongolia and other regions, which further fully shows that temperature was mainly an obstacle to NEP in the study area.

Under the conditions of constant temperature, the partial correlation analysis between NEP and precipitation showed that the significant positive correlation area accounted for 4.78% of the total area of the CGE. When the influence of precipitation on NEP was controlled, the proportion of areas with a significant negative correlation between NEP and temperature in the total area of the CGE decreased to 52.67%, and the area with a significant positive correlation increased slightly. This shows that climatic factors are mainly coupled to the CGE, and precipitation and temperature jointly regulate the change and accumulation of NEP in the CGE. The overall results of correlation and partial correlation analysis between NEP and climate factors in the CGE show that precipitation mainly promotes NEP accumulation, while temperature mainly suppresses it. The influence range of temperature on NEP in the CGE is significantly higher than that of precipitation, and the coupling effect of precipitation and temperature on NEP is significant.

Table 1 shows the partition proportion of annual NEP of the CGE in relation to climate factors. Regardless of the correlation coefficient or partial correlation coefficient, NEP was mainly negatively correlated with temperature and mainly positively correlated with precipitation. Table 1 further illustrates the coupling effect of precipitation and temperature on NEP.

5. Discussion

Guided by the goal of a carbon peak and carbon neutralization in China, this study estimated the fine grid scale NEP of the CGE from 2010 to 2020 by coupling the VPM and SRM. The results showed that the cumulative carbon sequestration of the CGE from 2010 to 2020 was 14.46 PgC. This study identified the carbon source area and carbon sink area; the total carbon sequestration in the carbon source area was 12.75 PgC, and the total carbon emission in the carbon sink area was 0.12 PgC. In short, the carbon sink capacity of the CGE shows the spatial differentiation characteristics of being high in the northwest of China and low in the southeast of China, and this distribution characteristic strongly corresponds with the 400 mm isohyet and 0 °C isotherm of China, which indicates that there is a strong correlation between GE distribution and precipitation and temperature. This study further analyzed the general trend of the interannual variation of NEP in the CGE from 2010 to 2020. It was found that it has an obvious synergistic relationship with the change characteristics of precipitation in China, which may show that the decline of precipitation led to the reduction in the distribution area of the CGE and the weakening of soil respiration, which led to the decrease in NEP.

In order to explore the response relationship between NEP and temperature and precipitation in the CGE, the correlation coefficient and partial correlation coefficient of the NEP and temperature and precipitation in the CGE were further analyzed. The results showed that the NEP of the CGE was positively correlated with precipitation (average correlation coefficient 0.09) and negatively correlated with temperature (correlation coefficient −0.06). When a certain factor was controlled unchanged, the partial correlation coefficient showed that climate factors are mainly coupled to the CGE, and precipitation and temperature jointly regulate the change and accumulation of NEP in the CGE. In addition, the study also found that the impact of temperature on NEP was significantly higher than that of precipitation in the study area.

In short, based on the fine grid data, this study realized the estimation of carbon sources and sinks of the CGE from 2010 to 2020 and quantitatively discussed the driving relationship between the spatiotemporal change of carbon source and sink of the CGE and temperature and precipitation. This study is of great significance for China in achieving the double carbon goal. However, there are still some deficiencies. The NEP reveals the characteristics and laws of the spatial distribution of carbon sources and carbon sinks in the CGE, but the ecosystem carbon cycle and soil respiration are complex ecological processes, and there are certain uncertainties in the estimation of models. When estimating NEP, all indicators were calculated on the scale of ″year″ in this study, which erased the influence of seasonal variation on NEP to a certain extent. This study only discussed the impact of temperature and precipitation on the NEP. In addition to temperature and precipitation, the impact of human activities, topographic factors, surface evapotranspiration and other factors on the spatial distribution of NEP was not considered. Therefore, further research should widely collect actual monitoring data to estimate more accurately the ecosystem’s carbon sources and sinks and consider the seasonal variation factors, as well as the fluctuation impact of human activities, meteorological factors and topographic factors on the CGE, so as to have a deeper understanding of the contribution mechanism of increasing the sinks and reducing the sources of carbon cycles in the CGE.

6. Conclusions

Based on the VPM and SRM, this study accounted for NEP in the CGE on a national scale. The results showed that the CGE has great carbon sequestration potential. The carbon sequestration area was about 2.30 million km², accounting for 80.44% of the total area of the CGE. From 2010 to 2020, the average annual carbon sequestration of the CGE reached 411.48 TgC/a. It was found that the CGE presents the spatial differentiation characteristics of being high in the northwest of China and low in the southeast of China, which strongly corresponds with the spatial variation characteristics of temperature and precipitation. Further correlation analysis between NEP and temperature and precipitation in the CGE also confirmed that the change of NEP in the CGE was positively correlated with precipitation and negatively correlated with temperature. The overall change is the result of the coupling effect of temperature and precipitation, and both of them jointly regulate the change and accumulation of NEP in the CGE. The results of this study can provide a decision-making basis for policy makers to formulate ecosystem quality and carbon enhancement policies according to local conditions, provide a reliable implementation path for China to achieve the goal of carbon peak and carbon neutralization and provide effective technical methods and data support for further research by other scholars in this field.

Author Contributions

Conceptualization, X.L. and G.L.; writing—original draft preparation, methodology, data curation, X.L.; software, G.L.; validation, D.J. and J.F.; formal analysis, Y.W.; writing—review and editing, G.L. and J.F.; visualization, J.F.; funding acquisition, D.J., G.L. and J.F. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available on request from the author.

Acknowledgments

We would like to thank the editors and anonymous reviewers for their helpful remarks.

Conflicts of Interest

The authors declare no conflict of interest.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures and Table

Figure 1. Location of the study area (the boundaries, names and designations are for illustrative purposes only and do not represent an official endorsement).

Figure 2. Spatial distribution of the perennial mean NEP of the CGE from 2010 to 2020.

Figure 3. Average NEP of the CGE in different provinces of China from 2010 to 2020.

Figure 4. Spatial distribution of the cumulative NEP of the CGE from 2010 to 2020.

Figure 5. Spatial distribution of carbon sources and carbon sinks of the CGE from 2010 to 2020.

Figure 6. Interannual variation trend of NEP in the CGE from 2010 to 2020 in China.

Figure 7. Interannual change rate distribution of NEP in the CGE.

Figure 8. The proportion of variation trends of NEP in the CGE of different provinces.

Figure 9. Hierarchical distribution of the NEP variation coefficient in the study area.

Figure 10. Correlation between NEP and precipitation and temperature in the CGE. (a) Correlation coefficient between NEP and precipitation; (b) partial correlation coefficient between NEP and precipitation; (c) correlation coefficient between NEP and temperature; (d) partial correlation coefficient between NEP and temperature.

Table 1

The partition proportion of annual NEP of the CGE in relation to climate factors.

Type	Significant Positive Correlation (%)	Insignificant Positive Correlation (%)	Insignificant Negative Correlation (%)	Significant Negative Correlation (%)
Correlation coefficient between NEP and temperature	2.25	44.04	-	53.72
Correlation coefficient between NEP and precipitation	1.60	63.12	-	35.27
Partial correlation coefficient between NEP and temperature	2.48	44.84	-	52.67
Partial correlation coefficient between NEP and precipitation	4.78	60.83	-	34.39

Supplementary Materials

The Supplementary Material can be downloaded at: https://www.mdpi.com/article/10.3390/su14148461/s1. Document S1: Vegetation Production Model. References [31,32,33,34] are also cited in the supplementary materials.

References

1. Saier, M.H. Climate Change, 2007. Water Air Soil Pollut.; 2007; 181, pp. 1-2. [DOI: https://dx.doi.org/10.1007/s11270-007-9372-6]

2. Solomon, S.; Plattner, G.K.; Knutti, R.; Friedlingstein, P. Irreversible climate change due to carbon dioxide emissions. Proc. Natl. Acad. Sci. USA; 2009; 106, pp. 1704-1709. [DOI: https://dx.doi.org/10.1073/pnas.0812721106] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19179281]

3. Sanders, S. 125 Questions: Exploration and Discovery; Science/AAAS Custom Publishing Office: Washington, DC, USA, 2021.

4. International Energy Agency. Global Energy Review: CO₂ Emissions in 2021; IEA Publications: Paris, France, 2021.

5. Schleussner, C.F.; Lissner, T.K.; Fischer, E.M.; Wohland, J.; Perrette, M.; Golly, A.; Rogelj, J.; Childers, K.; Schewe, J.; Frieler, K. et al. Differential climate impacts for policy-relevant limits to global warming: The case of 1.5 °C and 2 °C. Earth Syst. Dyn.; 2016; 7, pp. 327-351. [DOI: https://dx.doi.org/10.5194/esd-7-327-2016]

6. Schulze, E.D.; Wirth, C.; Heimann, M. Managing Forests After Kyoto. Science; 2000; 289, pp. 2058-2059. [DOI: https://dx.doi.org/10.1126/science.289.5487.2058] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11032555]

7. Sun, W.J.; Zhao, Y.; Li, Z.; Yin, Y.T.; Cao, C.L. Carbon Emission Peak Paths Under Different Scenarios Based on the LEAP Model—A Case Study of Suzhou, China. Front. Environ. Sci.; 2022; 10, 905471. [DOI: https://dx.doi.org/10.3389/fenvs.2022.905471]

8. Chai, Q.M.; Guo, H.Y.; Liu, C.Y.; Dong, L.; Ju, L.X.; Liu, C.S.; Chen, Y.; Chen, H.B.; Zhuang, G.Y. Global Climate Change and China’s Action Scheme: Climate Governance of China in the 14th Five-Year Plan Period from 2021 to 2025 (Conversation by Writing). Yuejiang Acad. J.; 2020; 12, pp. 36-58.

9. Lin, B.Q. The Period of Carrying out Energy Revolution to Promote Low Carbon Clean Development in China. China Ind. Econ.; 2018; 6, pp. 15-23.

10. Liu, J.B. Carbon neutralization: Building a community of human and natural life together. Fujian For.; 2021; 4, 1.

11. Central People’s Government of the People’s Republic of China. Action Plan for Carbon Dioxide Peaking before 2030. Available online: http://www.gov.cn/zhengce/content/2021-10/26/content_5644984.htm (accessed on 6 May 2022).

12. Intergovernmental Panel on Climate Change. IPCC Climate Change 2021: The Physical Science Basis; Cambridge University Press: London, UK, 2021.

13. Shi, P.J.; Ye, Q.; Han, G.Y.; Li, N.; Wang, M.; Fang, W.H.; Liu, Y.H. Living with global climate diversity-suggestions on international governance for coping with climate change risk. Int. J. Disaster Risk Sci.; 2012; 3, pp. 177-184. [DOI: https://dx.doi.org/10.1007/s13753-012-0018-y]

14. Wang, S.Z.; Fan, J.W.; Liu, S. A Comprehensive Analysis of Difference in Carbon Stock Estimation in the Grasslands of China. Acta Agrestia Sin.; 2017; 25, pp. 905-913.

15. Ni, J. Carbon storage in grasslands of China. J. Arid. Environ.; 2002; 50, pp. 205-218. [DOI: https://dx.doi.org/10.1006/jare.2001.0902]

16. Gao, Y.J.; Shi, J.; Li, X. Knowledge mapping analysis of grassland carbon sink research based on CiteSpace. Acta Prataculturae Sin.; 2020; 8, pp. 195-203.

17. Zhang, J.P.; Liu, C.L.; He, H.G.; Sun, L.; Qiao, Q.; Wang, H.; Ning, Y.C. Spatial-temporal Change of Carbon Storage and Carbon Sink of Grassland Ecosystem in the Three-River Headwaters Region Based on MODIS GPP/NPP Data. Ecol. Environ. Sci.; 2015; 24, pp. 8-13.

18. Pan, J.H.; Li, Z. Temporal-spatial change of vegetation net primary productivity in the arid region of Northwest China during 2001 and 2012. Chin. J. Ecol.; 2015; 34, pp. 3333-3340.

19. Yun, Y.J.; Zhao, J. Spatial Pattern of Vegetation Carbon Sinks Based on MODIS-NDVI Data: A Case Study in Shiyang River Basin, China. Mt. Res.; 2018; 36, pp. 644-653.

20. Xinhua News Agency. National Carbon Emissions Trading Market Online Trading Officially Launched. Available online: http://www.gov.cn/xinwen/2021-07/17/content_5625625.htm#1 (accessed on 6 May 2022).

21. Ministry of Natural Resources, PRC. 2021 China Mineral Resources. Available online: http://mnr.gov.cn/sj/sjfw/kc_19263/zgkczybg/202111/t20211105_2701985.html (accessed on 6 May 2022).

22. Ding, Y.; Chun, L.; Sun, J.J.; Wu, Z.N.; Yun, X.J.; Li, F.; Jia, D.Z.; Lai, Y.N. China Grassland. For. Hum.; 2020; Z1, 20–39+12–19

23. Chen, J.; Ban, Y.; Li, S. China: Open access to Earth land-cover map. Nature; 2014; 514, 434.

24. Xiao, X.; Boles, S.; Liu, J.; Zhuang, D.; Fronlking, S.; Li, C.; Salas, W.; Ill, B.M. Mapping paddy rice agriculture in southern China using multi-temporal MODIS images. Remote Sens. Environ.; 2005; 95, pp. 480-492. [DOI: https://dx.doi.org/10.1016/j.rse.2004.12.009]

25. Wang, D.D.; Liang, S.L.; Zhang, Y.; Gao, X.Y.; Bron, M.G.L.; Jia, A.L. A New Set of MODIS Land Products (MCD18): Downward Shortwave Radiation and Photosynthetically Active Radiation. Remote Sens.; 2020; 12, 168. [DOI: https://dx.doi.org/10.3390/rs12010168]

26. Didan, K. MOD13A2 MODIS/Terra Vegetation Indices 16-Day L3 Global 1 km SIN Grid V006. NASA EOSDIS Land Processes DAAC. Available online: https://lpdaac.usgs.gov/products/mod13a2v006/ (accessed on 6 May 2022).

27. Dai, E.F.; Huang, Y.; Wu, Z.; Zhao, D.S. Spatial-temporal features of carbon source-sink and its relationship with climate factors in lnner Mongolia grassland ecosystem. Acta Geogr. Sin.; 2016; 71, pp. 21-34.

28. Huang, X.J.; Zhang, X.Y.; Lu, X.H.; Wang, P.Y.; Qin, J.Y.; Jiang, Y.C.; Liu, Z.M.; Wang, Z.; Zhu, A.X. Land development and utilization for carbon neutralization. J. Nat. Resour.; 2021; 36, pp. 2995-3006. [DOI: https://dx.doi.org/10.31497/zrzyxb.20211201]

29. Lee, M.; Nakane, K.; Nakatsubo, T.; Koizumi, H. Seasonal changes in the contribution of root respiration to total soil respiration in a cool-temperate deciduous forest. Plant Soil; 2003; 255, pp. 311-318. [DOI: https://dx.doi.org/10.1023/A:1026192607512]

30. Tang, X.; Fan, S.; Qi, L.; Guan, F.; Du, M.; Zhang, H. Soil respiration and net ecosystem production in relation to intensive management in Moso bamboo forests. Catena; 2016; 137, pp. 219-228. [DOI: https://dx.doi.org/10.1016/j.catena.2015.09.008]

31. Chen, J.Q.; Yan, H.M.; Wang, S.Q.; Gao, Y.N.; Huang, M.; Wang, J.B.; Xiao, X.M. Estimation of gross primary productivity in Chinese terrestrial ecosystems by using VPM model. Quat. Sci.; 2014; 34, pp. 732-742.

32. Xiao, X.M.; Zhang, Q.Y.; Braswell, B.; Urbanski, S.; Boles, S.; Wofsy, S.; Berrien, M.; Ojimac, D. Modeling gross primary production of temperate deciduous broadleaf forest usingsatellite images and climate data. Remote Sens. Ensironmen; 2004; 91, pp. 256-270. [DOI: https://dx.doi.org/10.1016/j.rse.2004.03.010]

33. Xiao, X.; Hollinger, D.; Aber, J.; Goltz, M.; Davidson, E.A.; Zhang, Q.; Moore, B., III. Satellite-based modeling of gross primary production in an evergreen needleleaf forest. Remote Sens. Environ.; 2004; 89, pp. 519-534. [DOI: https://dx.doi.org/10.1016/j.rse.2003.11.008]

34. Raich, J.W.; Rastetter, E.B.; Melillo, J.M.; Kicklighter, D.W.; Steudler, P.A.; Peterson, B.J.; Grace, A.L.; Moore, B.; Vorosmarty, C.J. Potential net primaryproductivity in South America: Application of a global model. Ecol. Appl.; 1991; 1, pp. 399-429. [DOI: https://dx.doi.org/10.2307/1941899] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27755669]

35. Pei, Z.; Ouyang, H.; Zhou, C.; Xu, X. Carbon Balance in an Alpine Steppe in the Qinghai-Tibet Plateau. J. Integr. Plant Biol.; 2009; 51, pp. 521-526. [DOI: https://dx.doi.org/10.1111/j.1744-7909.2009.00813.x]

36. Wang, Q.; Zhang, T.B.; Yi, G.H.; Chen, T.T.; Bie, X.J.; He, Y.X. Tempo-spatial variations and driving factors analysis of net primary productivity in the Hengduan mountain area from 2004 to 2014. Acta Ecol. Sin.; 2017; 37, pp. 3084-3095.

37. Pan, J.H.; Huang, K.J.; Li, Z. Spatio-temporal variation in vegetation net primary productivity and its relationship with climatic factors in the Shule River basin from 2001 to 2010. Acta Ecol. Sin.; 2017; 37, pp. 1888-1899. [DOI: https://dx.doi.org/10.1080/10807039.2017.1400373]

38. Schober, P.; Boer, C.; Schwarte, L.A. Correlation Coefficients. Anesth. Analg.; 2018; 126, pp. 1763-1768. [DOI: https://dx.doi.org/10.1213/ANE.0000000000002864] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/29481436]

39. Xu, J.H. Mathematical Methods in Contemporary Geography; 2nd ed. Higher Education Press: Beijing, China, 2002.

40. Liu, F.; Zeng, Y.N. Analysis of the spatio-temporal variation of vegetation carbon source/sink in Qinghai Plateau from. Acta Ecol. Sin.; 2021; 41, pp. 5792-5803.

41. Wu, S.S.; Yan, Z.J.; Jiang, L.G.; Wang, R.; Liu, Z.F. The Spatial-Temporal Variations and Hydrological Effects of Vegetation NPP Based on MODIS in the Source Region of the Yangtze River. J. Nat. Resour.; 2016; 31, pp. 39-51.