Nitrogen Deposition Enhances the Positive Contribution of CO₂ Fertilization to China’s Carbon and Water Use Efficiencies

Abstract

With climate change, Nitrogen deposition (ND) and carbon dioxide (CO₂) have increased significantly, but the effects of ND and CO₂ fertilization on carbon-use efficiency (CUE) and water-use efficiency (WUE) are still unclear. Therefore, the effects of ND and CO₂ on CUE and WUE are studied by means of trend analysis and partial derivatives. The results show the following: (1) From 2000 to 2020, the mean CUE in China was 0.533, and the WUE was 1.235 g C kg⁻¹ H₂O. (2) In the study area, a 53.8% CUE and 61.8% WUE showed an upward trend, and ND enhanced the positive contribution of CO₂ to CUE and WUE by 15.9% and 19.3%, respectively. (3) ND explained 48.1% and 56% of the increase in CUE and WUE, respectively, and CO₂ explained 45.8% and 50.1% of the increases in CUE and WUE, respectively. (4) The areas dominated by ND and CO₂ accounted for 43.9% and 12.7% of the total area, respectively, highlighting the importance of ND in the driving mechanism of CUE and WUE. The research results are helpful to further research in CUE and WUE.

1. Introduction

Vegetation carbon-use efficiency (CUE) and water-use efficiency (WUE) are useful indicators for assessing carbon-sequestration and plant water-use patterns, which are important parameters for forming ecosystem carbon and water storage [1,2]. The value of CUE is the ratio of net primary productivity (NPP) to total primary productivity (GPP) [3]. WUE connects the cycles of water and carbon, which is a key parameter for understanding the two [4,5]. Water resources are regarded as the biggest obstacle to regional high-quality development, and improving the utilization efficiency of water resources is an effective way to deal with the increasingly serious water shortage [6].

In recent years, the increase of CO₂ concentration, temperature [7,8], drought, and other climate changes have seriously affected the carbon and water exchange of terrestrial ecosystems [9]. CO₂ fertilization can regulate the carbon allocation of the biomass of each component of the vegetation and change the demand for autotrophic respiration (Ra) of the vegetation, thereby increasing the CUE [10,11]. In addition, nitrogen deposition (ND) also affects the CUE [12,13]. More and more studies show that CUE changes with the changes of vegetation type, climate, nitrogen deposition, and other factors [14,15]. The eco-physiological characteristics of leaves show that WUE increases with an increase in atmospheric CO₂ concentration, which is called the “CO₂ fertilizer effect” [16,17]. Increasing the CO₂ concentration increases WUE by decreasing transpiration rate [18]. On the leaf scale, WUE depends on the supply of water and nitrogen in addition to the CO₂ in the atmosphere [19]. ND is an important factor in promoting WUE [20,21]. An increase in ND decreases stomatal conductance, leading to reduced transpiration, and its influence on the total primary productivity (GPP) determines the change in WUE [22,23,24]. ND and CO₂ have an important influence on CUE and WUE, however, it is still unclear how much of an impact they have on CUE and WUE. Therefore, it is necessary to quantitatively evaluate the impact of ND and CO₂ fertilization on CUE and WUE.

The purpose of this study aims to investigate the responses of CUE and WUE in China to ND and CO₂ from 2000 to 2020. The specific objectives are as follows: (1) Analyze the spatial-distribution and temporal-trend characteristics of CUE and WUE in China. (2) Elucidate the effects of ND and CO₂ on CUE and WUE. (3) Reveal the dominant areas and contribution rate of ND and CO₂ and the residual factors for CUE and WUE. This study is of great significance for enriching the relevant research on CUE and WUE.

2. Data and Methods

2.1. Study Area

Based on the characteristics of geographic location, physical geography, and human geography, the research area was divided into four regions: the southern region, Qinghai–Tibet region, northern region, and northwestern region (Figure 1). The Qinling Mountains and the Huaihe River are the dividing line between the northern and southern regions, whereas the Daxinganling-Yinshan-Helan Mountain is the dividing line between the northern region and the northwestern region. The four Qinghai–Tibet, northwest, north, and south areas of the study area are roughly divided by the first and second ladders. The northern region is the northern part of China’s monsoon climate zone, while the southern region is the southern part of the monsoon climate zone. Northwest China is a non-monsoon region, located west of the 400 mm annual isoprecipitation line, and the Qinghai–Tibet region is a unique geographical unit with a high altitude and cold climate.

2.2. CUE and WUE Calculations

Vegetation CUE describe the relationship between photosynthesis and respiration [25,26]. WUE refers to the carbon assimilation rate per unit water loss [27]. The calculation formulas for both are as follows:

$$\mathrm{CUE}=\frac{\mathrm{NPP}}{\mathrm{GPP}}$$

(1)

$$\mathrm{WUE}=\frac{\mathrm{GPP}}{\mathrm{ET}}$$

(2)

where CUE (dimensionless) is carbon-use efficiency, and WUE (g C kg⁻¹ H₂O) is water-use efficiency. GPP (g C m⁻²), NPP (g C m⁻²), and ET (kg⁻¹ H₂O m⁻²) are gross primary productivity, net primary productivity, and evapotranspiration, respectively.

2.3. Data Sources

The 2000–2020 GPP and NPP data were obtained from MOD17A3 with a resolution of 1 km, an interval of 8 days, and a scale factor of 0.1 (Figure S1), with data from NASA (https://search.earthdata.nasa.gov/search, accessed on 12 October 2021). MODIS data are widely used in studies related to gross primary productivity and net primary productivity [28,29]. NPP site data were compiled by the Global Primary Production Data Initiative (GPPDI) from the website: https://daac.ornl.gov/cgi-bin/, (accessed on 21 August 2021). GPP site data come from the website: https://fluxnet.fluxdata.org/, (accessed on 5 August 2021), and the eddy covariance measurements at the basis of this product were collected and harmonized in the LaThuile FLUX NETwork (FLUXNET) synthesis database. Evapotranspiration data were collected from the monthly evapotranspiration dataset at 0.25° resolution from the GLDAS Noah 2.1 dataset of the Goddard Geoscience Data and Information Service (GES DISC) (https://disc.gsfc.nasa.gov/datasets, accessed on 5 January 2022). GLDAS data combine satellite and ground observations to provide multiple types of high-quality datasets using land-surface models (LSM) and data-assimilation techniques [30].

Nitrogen dioxide (NO₂) column concentration is based on satellite observations using differential optical-absorption spectroscopy technology, and the Royal Netherlands Meteorological Institute combined modeling, retrieval, assimilation methods data from three sensors: GOME (Global Ozone Monitoring Experiment), SCIAMAHY (Scanning Imaging Absorption Spectrometer for Atmospheric Mapping), and GOME-2 (METOP-A) (Global Ozone Monitoring Experiment-2) (http://www.temis.nl/airpollution/no2.html, accessed on 20 January 2022). The CO₂ concentration comes from the Copernicus Atmosphere Monitoring Service (CAMS) global inversion-optimised greenhouse-gas fluxes and concentrations (https://ads.atmosphere.copernicus.eu/cdsapp#!/dataset/cams-global-greenhouse-gas-inversion?tab=overview, accessed on 11 December 2021). DEM elevation data were calculated using a digital elevation model (https://topex.ucsd.edu, accessed on 10 February 2022) (Figure 1), the spatial resolution is 1000 m × 1000 m.

2.4. Methods

2.4.1. Nitrogen Deposition Model

We referred to the model of Lu et al. [31] and established a pixel-by-pixel spatial linear-regression model based on the existing ND data and NO₂ column concentration, precipitation, temperature, relative humidity, wind speed, etc. The reference formula is as follows:

$${\mathrm{N}}_{\mathrm{d}\mathrm{e}\mathrm{p}}={\mathrm{a}}_{\mathrm{o}}\mathrm{T}+{\mathrm{a}}_1\mathrm{P}+{\mathrm{a}}_2\mathrm{H}+{\mathrm{a}}_3\mathrm{W}+{\mathrm{a}}_4\mathrm{C}+\mathrm{b}$$

(3)

where N_dep is nitrogen deposition, a₀, a₁, a₂, a₃, a₄, b are the regression coefficients of the linear equation; T is the surface temperature; P is precipitation; H is relative humidity; W is wind speed; and C is the concentration of NO₂ column.

2.4.2. Trend Analysis

We used ordinary least-squares regression for each cell to determine the slope value: linear-trend estimation was used to analyze the time dynamics of CUE, WUE, and various factors, and the statistical test of the correlation coefficient was used to evaluate the significant-change trend [32,33]. Its linear-trend value was estimated using the least-squares method.

$$\Delta \mathrm{CUE}=\frac{{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}(\mathrm{i}\times {\mathrm{CUE}}_{\mathrm{i}})-\frac{1}{\mathrm{n}}{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}\mathrm{i}\times {\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{CUE}}_{\mathrm{i}}}}}}{{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{i}}^2-\frac{1}{\mathrm{n}}{\left({\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{t}}_{\mathrm{i}}}\right)}^2}}$$

(4)

where n is the time span of 21 years from 2000 to 2020, ${\mathrm{CUE}}_{\mathrm{i}}$ is the CUE value of year i, i = 2000, 2001, …, 2020. At the same time, we used the same method to analyze the changing trend of factors such as $\Delta \mathrm{WUE}$, $\Delta \mathrm{ND}$, $\Delta {\mathrm{CO}}_2$, etc. If the correlation coefficient of the regression equation passed a significant confidence level of 0.05 (p < 0.05), it falls or increases to a significantly very important level.

2.4.3. Partial Derivatives of ND and CO₂ Effects on CUE and WUE

The 2000–2020 CUE and WUE responses to Nitrogen deposition (ND), carbon dioxide (CO₂), and Residual factor (Resi) were preliminarily calculated. They were expressed as the partial derivative of multiple linear regression [34,35].

$$\Delta \mathrm{CUE}=\mathrm{N}{\mathrm{D}}^{\mathrm{C}\mathrm{U}\mathrm{E}}+{{\mathrm{CO}}_2}^{\mathrm{C}\mathrm{U}\mathrm{E}}+{\mathrm{Resi}}_{\mathrm{con}}$$

(5)

where is the $\mathrm{N}{\mathrm{D}}^{\mathrm{C}\mathrm{U}\mathrm{E}}$, ${{\mathrm{CO}}_2}^{\mathrm{C}\mathrm{U}\mathrm{E}}$, and Resi_con, respectively, represent the contribution of ND, CO₂, and Resi to the change of CUE. Derived from Equation (5), the changes in ND, CO₂, and Resi factors related to the long-term change of CUE can be expressed by the following formula:

$$\mathrm{N}{\mathrm{D}}^{\mathrm{C}\mathrm{U}\mathrm{E}}=\frac{\partial \mathrm{C}\mathrm{U}\mathrm{E}}{\partial \mathrm{N}\mathrm{D}}\times \frac{\mathrm{d}\mathrm{N}\mathrm{D}}{\mathrm{d}\mathrm{t}}$$

(6)

Among them, $\frac{\mathrm{d}\mathrm{N}\mathrm{D}}{\mathrm{d}\mathrm{t}}$ represents the linear trends of nitrogen deposition [36] and $\frac{\partial \mathrm{C}\mathrm{U}\mathrm{E}}{\partial \mathrm{N}\mathrm{D}}$ represents the impact of ND on CUE from 2000 to 2020. To compare the impact of other drivers on CUE, Equation (6) was also applied to the sensitivity and long-term trend estimates of annual mean ND, CO₂, and Resi. Therefore, Formula (6) can be further rewritten as [37]:

$$\Delta \mathrm{C}\mathrm{U}\mathrm{E}=\frac{\partial \mathrm{C}\mathrm{U}\mathrm{E}}{\partial \mathrm{N}\mathrm{D}}\times \frac{\mathrm{d}\mathrm{N}\mathrm{D}}{\mathrm{d}\mathrm{t}}+\frac{\partial \mathrm{C}\mathrm{U}\mathrm{E}}{\partial \mathrm{C}{\mathrm{O}}_2}\times \frac{\mathrm{d}\mathrm{C}{\mathrm{O}}_2}{\mathrm{d}\mathrm{t}}+{\mathrm{Resi}}_{\mathrm{c}\mathrm{o}\mathrm{n}}$$

(7)

Here $\frac{\partial \mathrm{C}\mathrm{U}\mathrm{E}}{\partial \mathrm{N}\mathrm{D}}\times \frac{\mathrm{d}\mathrm{N}\mathrm{D}}{\mathrm{d}\mathrm{t}}$ is the contribution of ND on CUE, $\frac{\partial \mathrm{C}\mathrm{U}\mathrm{E}}{\partial {\mathrm{C}\mathrm{O}}_2}\times \frac{\mathrm{d}\mathrm{C}{\mathrm{O}}_2}{\mathrm{d}\mathrm{t}}$ is the contribution of CO₂ on CUE, and ${\mathrm{Resi}}_{\mathrm{con}}$ is the contribution of the residual factors to CUE. The influence of other factors on WUE was calculated in the same way, with (5) quantizing each pixel on a 1 km × 1 km spatial window. Therefore, in this study, we mainly considered the contributions of ND and CO₂ to CUE and WUE, and all other factors that would affect CUE and WUE were classified as residual factors.

2.4.4. Analysis of CO₂ Enhancement by Nitrogen Deposition

We calculated the area proportion of net ND enhanced CO₂ to CUE (S_NC). That is, ND (positive contribution) enhances the contribution of CO₂ (positive contribution) to the change in CUE [38]. The same approach was used for the contribution of ND enhanced CO₂ to WUE, calculated as follows:

$${\mathrm{S}}_{\mathrm{N}\mathrm{C}}=\left|{{\mathrm{S}}_{\mathrm{N}\mathrm{D}}}_{\mathrm{Increasing}}\right|÷\left|{{\mathrm{S}}_{{\mathrm{C}\mathrm{O}}_2}}_{\mathrm{Increasing}}\right|\times 100\%$$

(8)

$${{\mathrm{S}}_{\mathrm{N}\mathrm{D}}}_{\mathrm{Increasing}}=\left|{\mathrm{S}}_{\mathrm{N}\mathrm{D}}-{{\mathrm{S}}_{\mathrm{N}\mathrm{D}}}_{\mathrm{Decreasing}}\right|$$

(9)

$${{\mathrm{S}}_{{\mathrm{C}\mathrm{O}}_2}}_{\mathrm{Increasing}}=\left|{\mathrm{S}}_{{\mathrm{C}\mathrm{O}}_2}-{{\mathrm{S}}_{{\mathrm{C}\mathrm{O}}_2}}_{\mathrm{Decreasing}}\right|$$

(10)

where S_NC is the proportion of ND enhanced CO₂ contribution to CUE, ${\mathrm{S}}_{\mathrm{N}\mathrm{D}}$ is the contribution area of ND to CUE, ${\mathrm{S}}_{{\mathrm{C}\mathrm{O}}_2}$ is the contribution area of CO₂ to CUE, ${S_{\mathrm{ND}}}_{\mathrm{Increasing}}$ is the positive contribution area of ND to CUE change, ${{\mathrm{S}}_{{\mathrm{C}\mathrm{O}}_2}}_{\mathrm{Increasing}}$ is the positive contribution area of CO₂ to CUE, ${{\mathrm{S}}_{\mathrm{N}\mathrm{D}}}_{\mathrm{Decreasing}}$ is the negative contribution area of ND on CUE, and ${{\mathrm{S}}_{{\mathrm{C}\mathrm{O}}_2}}_{\mathrm{Decreasing}}$ is the negative contribution area of CO₂ impact on CUE.

2.4.5. Assessing the Relative Contributions of ND and CO₂

$\mathrm{Con}.\mathrm{ND}$ is the contribution rate of ND to CUE, $\mathrm{Con}.{\mathrm{CO}}_2$ is the contribution rate of CO₂ to CUE, and $\Delta {\mathrm{ND}}_{\mathrm{con}}$, $\Delta {\mathrm{CO}}_{2\mathrm{con}}$, and $\Delta {\mathrm{Resi}}_{\mathrm{con}}$ represent the contributions of ND, CO₂, and Resi to CUE, respectively. The same method was used to calculate the contribution rate of each factor to WUE. In addition to ND and CO₂ affecting CUE and WUE, factors such as climatic factors, human activities, and extreme weather all affect CUE and WUE. Therefore, in this study, all other influencing factors were classified as residual factors [39].

$$\mathrm{Con}.\mathrm{ND}=\frac{\left|\Delta {\mathrm{ND}}_{\mathrm{con}}\right|}{\left|\Delta {\mathrm{ND}}_{\mathrm{con}}\right|+\left|\Delta {\mathrm{CO}}_{2\mathrm{con}}\right|+\left|\Delta {\mathrm{Resi}}_{\mathrm{con}}\right|}\times 100\%$$

(11)

$${\mathrm{Con}.\mathrm{CO}}_2=\frac{\left|\Delta {\mathrm{CO}}_{2\mathrm{con}}\right|}{\left|\Delta {\mathrm{ND}}_{\mathrm{con}}\right|+\left|\Delta {\mathrm{CO}}_{2\mathrm{con}}\right|+\left|\Delta {\mathrm{Resi}}_{\mathrm{con}}\right|}\times 100\%$$

(12)

$$\mathrm{Con}.{\mathrm{Resi}}_{\mathrm{con}}=100-{\mathrm{ND}}_{\mathrm{con}}-{\mathrm{CO}}_{2\mathrm{con}}$$

(13)

3. Results and Analysis

3.1. Spatial and Temporal Distribution Characteristics

From a spatial point of view, the average CUE in China from 2000 to 2020 is 0.533, which is high in the west and low in the east, and the high-value area is distributed in the Qinghai–Tibet region at 0.62 (Figure 2a). Mainly because the environment in the Qinghai–Tibet Plateau is relatively cold and dry, plants in this environment have a strong carbon-storage capacity, and energy consumption for plant growth is low [40]. In addition, the vegetation types in the Qinghai–Tibet Plateau mainly consist of alpine meadows, alpine grasslands, alpine shrubs, etc. Generally, the CUE of shrub and herb ecosystems is higher than that of forests, so the CUE of this region is higher than that of other regions [41]. The average value of WUE in China is 1.235 g C kg⁻¹ H₂O, which is low in the west and high in the east, and the low-value area is mainly distributed in the Qinghai–Tibet region of 0.656 g C kg⁻¹ H₂O (Figure 2b). Low CO₂ partial pressure, low air temperature, and a high altitude-limited plant growth lead to lower WUE [42]. The high-value area of WUE is mainly distributed in the southern area of 1.714 g C kg⁻¹ H₂O, because the vegetation in this area is relatively lush, which can intercept more solar radiation and utilize more soil, thereby, promoting plant growth [43]. It is worth noting that the high-value area of CUE is exactly the low-value area of WUE. From the perspective of latitude, the high-value areas of CUE are mainly distributed in the mid-latitude areas, and the low-value areas of WUE are mainly located in the mid-latitude areas, mainly because the Qinghai–Tibet region is located in the mid-latitude areas, resulting in higher CUE and lower WUE (Figure 2c,d).

During the study period, the CUE in China showed an increasing trend of 0.2 × 10⁻³ yr⁻¹ (Figure 3a), and the most obvious increase trend was 1.5 × 10⁻³ yr⁻¹ in the southern region, which was nearly 7 times the growth trend of the entire study area (Table S1). A total of 56.9% of the regions of CUE passed the significance test (Figure 3c). The CUE of 53.8% showed an increasing trend, and the CUE of 46.2% showed a decreasing trend. China’s WUE showed an increasing trend of 3.3 × 10⁻³ g C kg⁻¹ H₂O yr⁻¹ (Figure 3b), north (12 × 10⁻³ g C kg⁻¹ H₂O yr⁻¹) and northwest (13 × 10⁻³ g C kg⁻¹ H₂O yr⁻¹) showed an increasing trend, which was nearly 4 times that of China (Table S1). Fifty-four-point-two percent of WUE passed the significance test (Figure 3d). The WUE of 61.8% showed an increasing trend, and the WUE of 38.2% showed a decreasing trend (

Table 1. CUE and WUE trends in different areas. Qinghai–Tibet Plateau: QT; South area: SA; North area: NA; Northwest area: NW.

Trend	CUE Increase	CUE Decrease	WUE Increase	WUE Decrease
QT	9.43%	11.26%	15.83%	7.81%
SA	13.61%	20.06%	10.65%	22.12%
NA	18.55%	10.15%	23.07%	4.02%
NW	12.17%	4.77%	12.25%	4.25%

). Significant increases in WUE in temperate and boreal forests in the northern hemisphere followed past increases in atmospheric-carbon-dioxide concentrations [5]. Elevated CO₂ concentration increases plant water-use efficiency by reducing transpiration rate [18].

3.2. Effects of ND and CO₂ on CUE and WUE

Here, we analyzed the correlations of ND and CO₂ with CUE and WUE (Figure 4). Fifty-two percent of ND was positively correlated with CUE, while 56% of ND was positively correlated with WUE, and the regions with positive correlation between the two were mainly distributed in the northern region (Figure 4a,b). ND in eastern China has a great impact on CUE and WUE, which is consistent with previous studies [44]. Temperate forests are generally considered nitrogen-limited, and nitrogen additions can significantly increase productivity [45]. A Nitrogen increase is the main reason for the obvious impact on CUE and WUE in areas with high nitrogen demand [46]. Fifty-four-point-two percent of CO₂ is positively correlated with CUE and mainly distributed in the southern region. Fifty-point-one percent of CO₂ was positively correlated with WUE and mainly distributed in northern and Qinghai–Tibet regions. Increases in atmospheric CO₂ have strong effects on ecosystem processes in temperate forests [5]. Although CO₂ in the northern region was negatively correlated with CUE, ND complemented the negatively correlated region, making the ND in the northern region positively correlate with CUE. CO₂ was negatively correlated with WUE in most of the southern regions, but ND also complemented the negatively correlated regions, making ND positively correlated with WUE.

3.3. ND and CO₂ for CUE and WUE Main Control Area

Contributions of ND and CO₂ to CUE and WUE show significant spatial differences (Figure 5). The positive contribution of ND to CUE accounted for 48.1% of the total area, mainly in the northern region, and the positive contribution of CO₂ to CUE accounted for 45.8% of the total area, mainly in the southern region. Fifty-six percent of the total area of the region where ND contributed positively to WUE was mainly distributed in the northern region, and the region with a positive contribution of CO₂ to WUE accounted for 50.1% of the total area, mainly in the Qinghai–Tibet and northern regions.

To determine the main factors controlling the variation of CUE and WUE in China, here, we considered the effect of the residual factors on CUE and WUE. As shown in Figure 6, among the effects of the three factors on CUE, the areas dominated by ND_con, CO₂_con, and Resi_con accounted for 43.9%, 12.7%, and 43.4% of the total area, respectively (

Table 2. Main control area and the rate of contribution of ND, CO₂, and Resi in different areas to CUE and WUE. Qinghai–Tibet Plateau: QT; South area: SA; North area: NA; Northwest area: NW.

Main Control Area	QT	SA	NA	NW	Contribution Rate	QT	SA	NA	NW
ND_CUE	8.49%	11.76%	14.62%	9.01%	ND_CUE	8.43%	11.56%	14.32%	9.39%
CO2_CUE	5.65%	3.18%	2.47%	1.38%	CO2_CUE	5.74%	4.52%	4.36%	2.48%
Resi_CUE	9.5%	17.81%	10.04%	6.09%	Resi_CUE	8.45%	16.83%	9.32%	4.6%
ND_WUE	8.55%	13.93%	18.18%	10.22%	ND_WUE	7.65%	12.87%	16.11%	7.51%
CO2_WUE	.08%	1.7%	3.2%	2.13%	CO2_WUE	7.79%	2.1%	3.63%	2.52%
Resi_WUE	8.07%	17.09%	5.71%	4.14%	Resi_WUE	9.58%	17.35%	6.25%	6.72%

). It shows that ND is the main factor affecting the change of CUE; ND mainly controls the CUE in the north and northwest regions. In addition, the influence of CO₂ on the CUE in the Qinghai–Tibet region is very important. As far as WUE is concerned, the areas dominated by ND_con, CO₂_con, and Resi_con account for 50.9%, 14.1%, and 35% of the total area, respectively (Table 2). ND has the largest control area of WUE in the entire region, and CO₂ controls WUE mainly in the Qinghai–Tibet region.

3.4. Contribution of Rates ND, CO₂, and Residual Factors to CUE and WUE

The contribution rate of ND to CUE and WUE is 43.7% and 44.05%, respectively; the contribution rate of CO₂ to CUE and WUE is relatively small, 17.1% and 16.04%, respectively; the contribution rate of the remaining factors to CUE and WUE also accounts for a large proportion of 39.2% and 39.9%, respectively (Table 2). The contribution rates of ND and residual factors to CUE and WUE have the same distribution trend; while contribution rates in the whole region are relatively high, the contribution rates in Qinghai–Tibet and a small part of the south are relatively low. The contribution rate of CO₂ in the whole region is relatively low, but the contribution rate is relatively high in Qinghai–Tibet and a small part of the south (Figure 7). We found that the high-value area of ND contribution to CUE and WUE is exactly the low-value area of CO₂ contribution to CUE and WUE. The increase in CUE caused by ND enhanced the area of the CO₂-driven increase by 15.9% over the same period (Figure 7g). In addition, ND enhanced the proportion of CO₂ contribution to WUE by 19.3% (Figure 7h). From the above analysis, it can be concluded that ND and CO₂ have a complementary relationship in many regions, that is, the region where ND contributes more to CUE has a smaller contribution rate of CO₂ to it.

4. Discussion

4.1. Comparison with Other Related Research

To enhance the reliability of the results, we compared the findings of this study with other results (

Table 3. Mean values and trends of CUE by region.

Location	Method	Period	CUE	CUE (yr−1)	References
Globe	MODIS	2000–2009	0.52	-	[52]
China	MODIS	2000–2015	0.537	3.51 × 10−5	[53]
China	MODIS	2000–2014	0.54	-	[54]
Songnen Plain	MODIS	2001–2015	0.516	-	[55]
West Sichuan Plateau	MODIS	2001–2018	0.846	3.1 × 10−2	[47]
Tibetan Plateau	MODIS	2000–2010	0.646	-	[48]
China	MODIS	2000–2020	0.533	2 × 10−4 yr−1	This Study

and

Table 4. Mean values and trends of CUE by region.

Location	Method	Period	WUE (g C kg−1 H2O)	WUE (g C kg−1 H2O yr−1)	References
China	MODIS	2000–2014	1.227	10 × 10−3	[50]
China	MODIS	1979–2012	1.48	7.32 × 10−4	[56]
Tibetan Plateau	MODIS	2001–2020	1.17	0.8 × 10−3	[51]
Songnen Plain	MODIS	2001–2015	1.319	-	[55]
China	MODIS	2000–2020	1.235	3.3 × 10−3	This Study

). CUE and WUE in most areas of China showed an upward trend, indicating that the carbon-sequestration capacity of terrestrial ecosystems has increased over the past 21 years. The average vegetation CUE is 0.846, indicating that the vegetation in the western Sichuan Plateau grows well and the carbon-sequestration efficiency is high [47]. In addition, we also found that the value of CUE increases with increasing altitude [48]. The eastern margin of the Qinghai–Tibet Plateau and the Qilian Mountains have the largest CUE, indicating that the horizontal and vertical zonality of CUE and the geographical distribution of latitude, longitude, and altitude are highly correlative. Climate warming and drying changes may contribute to an increase in the CUE of the Qinghai–Tibet Plateau [49], which is similar to the findings of this study; the CUE of the Qinghai–Tibet region is higher than that of other regions. The WUE in China is 1.227 g C kg⁻¹ H₂O [50], which is similar to the findings of this study (1.235 g C kg⁻¹ H₂O). The mean WUE in the Qinghai-Tibet Plateau region was 1.17 g C kg⁻¹ H₂O, the lowest among all regions [51]; the Qinghai-Tibet region also had the lowest WUE value among the four subregions in this study.

From the scatter plots of the NPP, GPP, CUE, and WUE site and raster data, it can be seen that the distribution of the two datasets at the same longitude are consistent (Figure 8). The correlation coefficients between the measured data at NPP and GPP sites and the MODIS raster data are 0.757 and 0.879, respectively, (Figure 8a,b), which pass the 95% significance test. The results show that the measured data and raster data have good correlation (Figure S3). In addition, we used NPP and GPP site and raster data to calculate the correlation coefficients of CUE and WUE as 0.773 and 0.838, respectively (Figure 8c,d).

4.2. Interactive Effects of ND and CO₂ on CUE and WUE

In all categories, the areas where ND increases CUE and WUE while CO₂ reduces CUE and WUE are mainly distributed in the northern region, accounting for the largest proportion at 41.65% (Figure 9). For areas where ND has a positive effect on CUE and WUE, the CO₂ shows a negative effect on CUE and WUE. As soil-nutrient concentrations increase due to ND, transpiration decreases to match the effectiveness of nutrient mass flow caused by reduced stomatal conductance [22]. The areas where ND and CO₂ have a positive impact on WUE is larger than the area where both of them have a positive impact on CUE (Figure 10). ND induced increases in LAI increases canopy transpiration, resulting in decreased soil evaporation [57]. Therefore, the effect of ND on GPP is greater than that of ET, which is consistent with the finding that the combined effect of CO₂ fertilization and N addition on WUE is greater than the effect of CO₂ alone [23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58]. The areas where ND reduces CUE and WUE while CO₂ increases CUE and WUE are mainly distributed in Qinghai–Tibet and southern regions, accounting for 35.33% (Figure 9). In areas where ND has a negative effect on CUE and WUE, the CO₂ shows a positive effect on CUE and WUE. It shows that ND and CO₂ have an interactive relationship, and the affected area accounts for 76.98% of the total area.

4.3. Effect of Residual Factors on CUE and WUE

From 2000 to 2020, ND (15.24 mg N m⁻² yr⁻¹) and CO₂ (2.21 ppm yr⁻¹) showed an increasing trend, which promoted CUE and WUE. Besides ND and CO₂ (Figure S2), various other factors also have profound effects on CUE and WUE at different temporal and spatial scales. The temperature and precipitation factors also showed an upward trend; rising air temperature decreased the CUE ratio, and increased precipitation led to an increase in the CUE ratio [52,53,54,55,56,57,58,59]. In environmental conditions, temperature, precipitation, solar radiation, etc, are all key climate-change factors that affect the WUE of ecosystems [60]. Photosynthetically active radiation, soil nutrition, etc, also affect CUE and WUE [61]. In cold regions at high latitudes or across the Tibetan Plateau, this limitation will be removed to some extent with increasing precipitation and temperature, which may stimulate more carbon sequestration [62]. High altitude and extreme drought conditions limit plant growth and lead to a decrease in the ratio between T and ET, which explains why the lowest value of WUE occurs in the Qinghai–Tibet region [63,64]. Vapor-pressure deficit (VPD) is the most important environmental factor affecting WUE, and many studies at the ecosystem level have found that VPD is significantly negatively correlated with WUE [65]. Increased humidity in the air promotes plant WUE [43]. At the individual ecosystem level, mild water-stress increases WUE, while extreme water-stress decreases WUE [66,67].

In addition, land-use change can have an impact on CUE and WUE [1,68,69]. There are also studies showing that CUE and WUE vary with vegetation type [15,70]. At the same time, vegetation functional types and soil-nutrient content are also important factors affecting the changes of CUE and WUE [14,41]. WUE continued to rise for all plant types in the presence of increased aerosol loads [44]. Healthy ecosystems enhance ecosystem services, contributing to CUE and WUE [57].

4.4. Uncertainty and Future Outlook

This paper studies the temporal and spatial pattern of CUE and WUE in China and the impact of ND and CO₂ on CUE and WUE, which effectively improves the understanding of vegetation CUE and WUE. However, our study still has some uncertainties and limitations. Firstly, in terms of remote-sensing data of GPP and NPP, we did not select more remote-sensing data for comparative analysis, and this uncertainty affects the comprehensive understanding of CUE and WUE to a certain extent. The change of GPP data makes the calculated NPP change, which affects the value of CUE and also increases the uncertainty of research. Secondly, the data accuracy of ND and CO₂ also has certain deficiencies. In terms of content, our study mainly analyzed the responses of CUE and WUE to ND and CO₂. The study found that the residual factors still accounted for a large proportion of the changes in CUE and WUE, and we classified all other influencing factors as residual factors but did not quantitatively analyze the impact of the residual factors (climate factors, human activities, and extreme climate conditions). In future research, the impact of other factors on CUE and WUE should be thoroughly analyzed to enrich the understanding of CUE and WUE.

5. Conclusions

In this study, the method of trend analysis and partial derivatives was used, revealing the effects of ND and CO₂ on vegetation CUE and WUE. We clarified the main control areas of ND, CO₂, and residual factors for CUE and WUE, quantitative analyzed the different proportions of ND interacting with CO₂ contributing to CUE and WUE. The main results are as follows:

(1)

From 2000 to 2020, 53.8% of China’s CUE showed an upward trend of 0.2 × 10⁻³ yr⁻¹, with the most obvious increase in trend in the southern region (1.5 × 10⁻³ yr⁻¹); 61.8% WUE showed an upward trend of 3.3 × 10⁻³ g C kg⁻¹ H₂O yr⁻¹, and the most obvious increase in trend was 13 × 10⁻³ g C kg⁻¹ H₂O yr⁻¹ in the northwest region.

(2)

Approximately 48.1% and 56% of ND were positively correlated with CUE and WUE, and the proportions of CO₂ positively correlated with CUE and WUE were 45.8% and 50.1%, respectively.

(3)

Approximately 52% and 56% of ND with CUE and WUE were positive contributions, and the proportion of CO₂ that contributes positively to CUE and WUE were 54.2% and 50.1%, respectively.

(4)

The proportions of ND, CO₂, and residual factors to the main control area of CUE were 43.9%, 12.7%, and 43.4% respectively, and proportion of the main control area of WUE were 50.9%, 14.1%, and 35% respectively.

(5)

ND enhances the contribution of CO₂ to CUE by 15.9%, ND enhances the contribution of CO₂ to WUE by 19.3%.