A Database of High-Resolution Meteorological Drought Comprehensive Index Across China for the 1951–2022 Period

Abstract

Drought events exacerbated by global climate change occur frequently in China. Currently, high-spatiotemporal-resolution gridded meteorological drought index datasets are generally available for single time scales (e.g., 30, 60, 90, and 150 days) and do not fully account for seasonal differences in the impact of drought on vegetation, thus limiting their accuracy when monitoring drought in different regions of China. To compensate for the limitations of existing drought index datasets, a Chinese regional daily meteorological drought comprehensive index (MCI) dataset covering 1951–2022 with a spatial resolution of 0.1 degrees was developed, and standardized precipitation index (SPI) and standardized precipitation evaporation index (SPEI) datasets at 30- and 90-day scales were constructed based on ERA5-Land datasets. Compared with the existing SPI and SPEI datasets, the generated dataset exhibits a high degree of consistency with those in eastern part of China (R² > 0.5; the average biases were close to 0 and significantly smaller than RMSEs of the fitting). Additionally, the MCI dataset can more accurately reflect the changes in shallow soil moisture in the eastern part of China in a timely manner (R² > 0.7 for the 0–7 cm depth), thus providing notable empirical support for research on drought development in different ecosystems.

Dataset: https://doi.org/10.5281/zenodo.14741907, https://doi.org/10.5281/zenodo.14760329, and https://doi.org/10.5281/zenodo.14795177.

Dataset License: The dataset is made available under license Creative Commons Attribution 4.0 International.

1. Introduction

Drought, as a global natural disaster, poses a serious threat to water resources and food security [1,2,3]. It is estimated that global economic losses due to drought amount to $6–8 billion annually, accounting for more than half of all meteorological disasters, with the number of affected people exceeding that of any other meteorological disaster [4]. China, with its vast territory, has an extremely heterogeneous distribution of water resources, and drought has become a key factor affecting the country’s grain production [5,6,7]. In recent years, the trend of global warming has intensified, leading to a significant increase in the frequency and intensity of drought events in China, which in turn has resulted in severe economic losses [8,9]. Therefore, conducting drought monitoring at high spatial and temporal resolutions is crucial for gaining a deep understanding of the formation and development mechanisms of drought, and the results can be used to help develop and implement preventive measures through drought risk management and early warning provision.

On the basis of the impact of drought on various aspects of the water cycle, drought can be classified into meteorological drought, agricultural drought, hydrological drought, and socioeconomic drought [10,11]. Long-term precipitation deficiency can manifest as meteorological drought. Moreover, if meteorological drought persists for an extended period, it typically triggers agricultural drought, hydrological drought, and socioeconomic drought [12,13,14]. Therefore, monitoring meteorological drought and providing timely warnings are highly important for mitigating the damage caused by drought to agriculture, water resources, and socioeconomic activities. To assess meteorological drought accurately, various drought indices have been proposed [15,16,17,18,19,20,21]. Moreover, to measure precipitation deficits over different time scales, researchers have proposed the standardized precipitation index (SPI), which allows for the assessment of drought intensity at various time scales via the normalization of precipitation data in the same period over various years [22]. Owing to its simplicity and practicality, this index has become one of the most widely used drought indices. However, the SPI does not consider the impact of potential evapotranspiration on drought, which is related to factors such as temperature. Additionally, the SPI at a single time scale has certain limitations for accurately assessing drought intensity. Considering the impact of potential evapotranspiration, Vicente-Serrano standardized the difference between precipitation and potential evapotranspiration in the same period over multiple years and established the standardized precipitation evapotranspiration index (SPEI) [23]. Compared with the SPI, the SPEI is more sensitive to changes in evapotranspiration caused by temperature [24]. Furthermore, to address the limitations of single-time-scale drought indices in assessing drought intensity in China and to fully account for the impact of potential evapotranspiration, the National Climate Center of China proposed a comprehensive meteorological drought index (CI) based on the long-term operational service experience of China’s meteorological drought [25]. This index integrates the SPI at short time scales (30 days) and long timescales (90 days), as does the relative humidity index, which reflects crop water deficit conditions. This index has been incorporated into the national standard of China for meteorological drought level classification. However, in daily drought monitoring practice, the excessive sensitivity of the CI to precipitation events leads to discontinuities in the reflected drought intensity. Additionally, the CI does not respond significantly to severe drought events caused by long-term precipitation deficits [21,24]. To address this issue, the National Climate Center of China proposed the meteorological composite drought index (MCI) [26,27] and revised the national standard of China for meteorological drought classification on this basis. This index builds upon the CI by incorporating a standardized weighted precipitation index over the past 60 days, thereby improving the unreasonable leap phenomenon in the development process of drought. Additionally, the MCI incorporates the SPI over the past 150 days, thereby considering the impact of precipitation over the last 150 days, making its response to major drought events more pronounced. Moreover, the MCI values are adjusted according to the impact of drought on vegetation in different regions and seasons to make the drought monitoring services more targeted by introducing a seasonal adjustment factor. Currently, the MCI has been widely applied in meteorological drought monitoring services in China and has played a significant role in reflecting the development and evolution of drought [28,29,30].

Long-term drought index datasets with high spatiotemporal resolutions (HSTRs) are crucial for exploring the processes of drought initiation and development. However, constrained by the spatial coverage limitations of meteorological station data, meteorological drought index products derived from such observations often struggle to meet the demands of high spatiotemporal resolution drought monitoring in China, particularly across the vast and sparsely populated western regions. The development of meteorological reanalysis datasets like ERA5-Land has provided new opportunities for conducting high-resolution precision drought analysis. Currently, several datasets based on meteorological reanalysis data provide HSTR grid-based meteorological drought indices, such as the SPI and SPEI, with spatial resolutions of up to 0.1 degrees and temporal resolutions of up to 1 day. The time scales range from 30 to 360 days, and these datasets cover the region of China since the founding of the People’s Republic of China (PRC) (1949–2022) [31,32,33,34]. Nevertheless, the existing datasets use the SPI and SPEI, which are drought indices for a single time scale and do not comprehensively consider the impacts of precipitation deficits and high temperatures at different time scales, as well as the differences in the impacts of drought on vegetation across different seasons. Given the vast territory of China, which has diverse climates and terrain types, SPI and SPEI datasets, which are based on a single temporal scale and do not account for seasonal factors, are characterized by a certain degree of uncertainty in drought monitoring. In addition, recent studies emphasize that climate change has induced non-stationarity in drought drivers (e.g., precipitation anomalies coupled with rising temperatures), which challenge the application of traditional drought indices [17,19,35]. The development of non-stationary indices, such as the Nonstationary Standardized Precipitation Evapotranspiration Index (NSPEI) [19], can better capture drought risks under global warming. The MCI used in this study, which integrates multi-timescale drought indices and seasonal adjustments, responds to this need by dynamically adapting to regional climate non-stationarity across China. However, to date, there is no publicly available grid-based MCI drought index dataset for the region of China, making the development of a daily MCI drought index dataset with high spatial resolution for the Chinese region since 1949 particularly important.

The core objective of this study is to construct a daily multi-timescale SPI and SPEI dataset covering the region of China from 1951 to 2022, with a spatial resolution of 0.1 degrees, and to further develop a daily MCI dataset. Specifically, the three main research objectives are as follows: (1) By performing a comparative analysis of the high-spatial-resolution SPI and SPEI datasets produced in this study with existing drought index datasets, the reliability of these datasets and the ERA5-Land precipitation data and potential evapotranspiration data used to calculate these indices across different regions of China are assessed; (2) the correlations among the different drought index datasets (MCI, SPI and SPEI at different time scales) and the daily cumulative precipitation and soil moisture content data at various depths are compared to further evaluate the characteristics and advantages of the MCI drought index for reflecting drought conditions at various time scales in the region of China; and (3) finally, the constructed MCI dataset is used to analyse the drought variation characteristics in China since the 1950s and explore the temporal and spatial distributions of key drought events in the mid-lower Yangtze Basin in 2022. Compared with existing research findings, the effectiveness of the MCI dataset produced in this study for drought identification and drought variation analysis is evaluated.

2. Materials and Methods

2.1. Overview of the Research Area

This research area covers China, which is located in the eastern part of Asia and on the western coast of the Pacific Ocean. China’s terrain distribution presents a three-level pattern, with the western region being high and the eastern region being relatively low (as shown in Figure 1). The Chinese territory features a rich variety of landforms, such as plains, plateaus, mountains, hills, and basins. The plain areas are distributed mainly in the eastern part of China, such as on the Northeast China Plain, on the North China Plain, and in the middle–lower Yangtze Basin. These areas are the main grain production bases of China. These regions are densely populated, with fertile soil and abundant water resources, creating favourable conditions for the large-scale development of agriculture. In contrast, the western part of China is dominated by plateaus and basins. Owing to its high altitude, the climate is cold and dry, with few large-scale crop cultivation areas. In terms of climate, the regional climate characteristics of China are greatly influenced by the thermal differences between land and sea, resulting in a significant monsoon climate and continental climate features [36]. This monsoon climate leads to distinct dry and wet seasons, as well as four clear seasons in the eastern regions, where winters are dry and cold and summers are humid and rainy. The influence of terrain also results in a pattern in which the eastern coastal regions receive more rainfall than other regions do, whereas the western inland regions are arid and receive less rainfall [37,38]. In the eastern regions, the rainy season and high-temperature season begin earlier and last longer in the southern region, whereas in the northern region, they start later and last for a shorter period. This climatic difference results in different agricultural production characteristics between the southern and northern regions of China. Moreover, with global climate change, the originally wet and hot summers in southern China may experience delayed or shortened rainy seasons, leading to summer droughts, which in turn affect the yield of autumn crops such as rice and corn [39]. Therefore, conducting real-time and accurate drought monitoring for China’s regions, as well as analysing the development trends of the intensity and spatial range of droughts, is extremely important for agricultural and socioeconomic development [40]. Figure 1 shows the meteorological and geographical primary zoning of China by the China Meteorological Administration, which divides the region of China into 10 areas, including five regions in the eastern part of the country with low terrain and dense populations: Northeast China, North China, Central China, East China, and South China.

2.2. Overview of Dataset Production and the Technical Verification Process

The dataset production and technical verification process in this study involved 3 steps (Figure 2): meteorological data preprocessing, drought index calculation and the validation of drought index datasets. (1) In the process of preprocessing meteorological data, the hourly precipitation and potential evapotranspiration data from ERA5-Land were converted into daily cumulative data, which were then converted into daily cumulative precipitation data and potential evapotranspiration data for the last 30, 60, 90, and 150 days; (2) in the process of calculating drought indices, the output results from the previous step were employed to obtain multi-scale SPI and SPEI datasets, which were subsequently used to establish MCI datasets; (3) in the process of validating the drought index datasets, two previously published drought index datasets were employed to evaluate the accuracy of the drought index datasets produced in this study. In addition, soil moisture content data for different depths from ERA5-Land were adopted to evaluate the performance of the different drought indices established in this study in measuring drought severity. The key data and methods in the dataset production and technical verification processes are detailed in the following sections.

2.3. Precipitation and Potential Evapotranspiration Data

In this study, hourly precipitation data and potential evapotranspiration data were obtained from ERA5-Land for the period between 1950 and 2022, with a spatial resolution of 0.1 degrees. As a fifth-generation meteorological reanalysis dataset launched by the European Centre for Medium-Range Weather Forecasts, ERA5-Land utilizes a vast amount of historical observation data and combines data assimilation with modelling techniques to achieve precise estimations of global meteorological parameters on a per-pixel basis [41]. Numerous studies have confirmed that the meteorological data, such as precipitation data, provided by ERA5-Land are highly accurate in the northern and eastern–central regions of China [42,43,44], and these data have been widely applied to generate drought index products such as the SPI and SPEI for China [32,45], which demonstrates the reliability of ERA5-land data in producing regional drought index products in China. In this study, we converted the hourly precipitation and potential evapotranspiration data from ERA5-Land into daily cumulative data and then calculated average daily cumulative precipitation data and potential evapotranspiration data for the past 30, 60, 90, and 150 days to facilitate the calculation of drought indices such as the SPI, SPEI, and MCI.

2.4. Verification Dataset

To ensure the reliability of the SPI and SPEI datasets produced in this study, we used the daily SPI and SPEI datasets based on the China meteorological forcing dataset (CMFD) meteorological reanalysis dataset for the Chinese region from 1979 to 2018 developed by Zhang et al. [33] (hereafter referred to as SPI_Zhang and SPEI_Zhang), as well as the daily global SPEI dataset based on ERA5 dataset from 1982 to 2021 developed by Liu et al. [32] (SPEI-GD). Both datasets were released in recent years, cover two time scales of 30 and 90 days, and have high spatial and temporal resolutions. The spatial resolution of SPI_Zhang and SPEI_Zhang is 0.1 degrees, whereas that of SPEI-GD is 0.25 degrees, which is similar to the spatial resolution of the ERA5-Land meteorological reanalysis data used in this study. For a comparative analysis with SPEI-GD, we resampled the spatial resolution of the SPEI dataset generated in this study to 0.25 degrees by using pixel aggregation.

2.5. Soil Moisture Content Data

To assess the correlation between the MCI and soil moisture content at different depths and compare it with the SPI and SPEI at various time scales, hourly soil moisture data from ERA5-Land were collected for the period between 1950 and 2022. The data cover three different depth layers, 0–7, 7–28, and 28–100 cm, which represent the moisture contents in the surface, middle, and deep soil layers, respectively.

2.6. Calculation of the SPI

The SPI is calculated by computing the gamma probability distribution of precipitation over a certain period and then normalizing it to a standard normal distribution, the steps of which are as follows [26]:

• The probability density function of the gamma distribution for cumulative precipitation (x) over a certain period is as follows:

  $$f\left(x\right)={\displaystyle \frac{1}{\beta \mathsf{\Gamma }\left(\gamma \right)}}{x}^{\gamma -1}{e}^{-x/\beta }x>0,$$

  (1)

  where $\beta$ > 0 and $\gamma$ > 0 are the scale and shape parameters, respectively, which can be calculated via the maximum likelihood estimation method.

For the precipitation in a certain year $x_0$, the probability of occurrence of an event for which the random variable $x$ is less than $x_0$ can be obtained as:

$$F\left(x<x_0\right)={\int }_x^{x_0}f\left(x\right)dx,$$

(2)

2.

The probability of occurrence of an event for which the precipitation ($x$) is 0 can be represented as:

$$F\left(x=0\right)=m/n,$$

(3)

where the number of samples with a precipitation value of 0 is denoted by $m$ and the total number of samples is denoted by $n$.

3.

The probability values calculated from Equations (2) and (3) are substituted into the standard normal distribution function, the formula for which is as follows:

$$F\left(x<x_0\right)={\displaystyle \frac{1}{\sqrt{2\pi }}}{\int }_x^{x_0}{e}^{-z^2/2}dx,$$

(4)

The approximate solution of Equation (4) yields:

$$Z=S\left\{t-{\displaystyle \frac{\left(c_{2}t+c_1\right)t+c_0}{\left[\left(d_{3}t+d_2\right)t+d_1\right]t+1.0}}\right\},$$

(5)

where $t=\sqrt{ln{\displaystyle \frac{1}{F^2}}}$ and $F$ is the probability calculated from Equation (2) or Equation (3).

When $F$ > 0.5, the value of $F$ is set as 1 − $F$, and $S$ = 1; when $F$ ≤ 0.5, $S$ = −1. Additionally, $c_0$ = 2.515517, $c_1$ = 0.802853, $c_2$ = 0.010328, $d_1$ = 1.432788, $d_2$ = 0.189269, and $d_3$ = 0.001308.

In this study, daily multi-timescale cumulative precipitation data from 1951 to 2022 were used to calculate the daily SPI at 30-, 60-, 90-, and 150-day scales (referred to as SPI30d_{HSTR_China}, SPI60d_{HSTR_China}, SPI90d_{HSTR_China} and SPI150d_{HSTR_China}, respectively).

2.7. Calculation of the SPEI

The SPEI is constructed on the basis of the difference between precipitation and potential evapotranspiration, and a log-logistic probability density function that includes three parameters is used to describe this variation. The calculation steps are as follows [26].

1.

The difference between the accumulated precipitation and potential evapotranspiration $D_i$ is calculated over a specific time scale.

2.

The log-logistic probability density function is used to calculate the SPEI, and the formula for the log-logistic distribution probability density function for period $D_i$($x$) is as follows:

$$F\left(x\right)={[1+{({\displaystyle \frac{\alpha }{x-\gamma }})}^{\beta }]}^{-1},$$

(6)

where $\alpha$, $\beta$ and $\gamma$ represent the scale, shape and origin, respectively, and can be fitted via linear moments.

3.

The cumulative probability density $P$ is normalized as follows:

$$P=1-F\left(x\right),$$

(7)

When the cumulative probability $P$ is less than or equal to 0.5,

$$W=\sqrt{-2ln\left(P\right)},$$

(8)

$$SPEI=W-{\displaystyle \frac{c_0-c_{1}W+c_2{W}^2}{1+d_{1}W+d_2{W}^2+d_3{W}^3}},$$

(9)

where $c_0$ = 2.515517, $c_1$ = 0.802853, $c_2$ = 0.010328, $d_1$ = 1.432788, $d_2$ = 0.189269, and $d_3$ = 0.001308.

When the cumulative probability $P$ is greater than 0.5, $P$ = 1 − $P$, and the formula for the SPEI is as follows:

$$SPEI=-(W-{\displaystyle \frac{c_0-c_{1}W+c_2{W}^2}{1+d_{1}W+d_2{W}^2+d_3{W}^3}}),$$

(10)

In this study, daily multi-timescale cumulative precipitation and potential evapotranspiration data from 1951 to 2022 are used to calculate the daily SPEI at 30-, 60-, 90-, and 150-day scales (referred to as SPEI30d_{HSTR_China}, SPEI60d_{HSTR_China}, SPEI90d_{HSTR_China} and SPEI150d_{HSTR_China}, respectively).

2.8. Calculation of the MCI

Given that drought is caused by the cumulative effect of recent and long-term precipitation deficits, the MCI is used to comprehensively assess evapotranspiration over the past 30 days and precipitation over the past 60 days, quarter and 150 days. Furthermore, considering the characteristics of the crop growth season, the MCI incorporates a seasonal adjustment factor, making it more suitable for monitoring drought conditions in crop-growing areas than are other indices. The formula for the MCI is as follows [26]:

$$MCI=Ka\times \left(a\times SPI{W}_{60}+b\times M{I}_{30}+c\times {SPI}_{90}+d\times {SPI}_{150}\right),$$

(11)

where $SPI{W}_{60}$ is the standardized precipitation weighting index at the 60-day scale, $M{I}_{30}$ is the relative humidity index at the 30-day scale, ${SPI}_{90}$ is the SPI at the 90-day scale, and ${SPI}_{150}$ is the SPI at the 150-day scale. $a$, $b$, $c$ and $d$ are weight coefficients that vary among different provinces in southern and northern China, of which the setting scheme is shown in

Table 1. Setting scheme of the weight coefficients for the MCI.

Meteorological and Geographical Region of China	a	b	c	d
Northeast China	0.3	0.5	0.3	0.2
Inner Mongolia	0.3	0.5	0.3	0.2
North China	0.3	0.5	0.3	0.2
East China	0.5	0.6	0.2	0.1
Central China	0.5	0.6	0.2	0.1
South China	0.5	0.6	0.2	0.1
Northwest China	0.3	0.5	0.3	0.2
Southwest China	0.3	0.5	0.3	0.2
Xinjiang	0.3	0.5	0.3	0.2
Tibet	0.3	0.5	0.3	0.2

. $Ka$ is the seasonal adjustment coefficient, which is determined based on the sensitivity of the main vegetation to drought in different seasons across various regions of China. The setting scheme for $Ka$ is shown in

Table 2. Setting scheme of seasonal adjustment factor in the MCI.

Agro-Climatic Zone	Province (Autonomous Region, Municipality)	Month
		1	2	3	4	5	6	7	8	9	10	11	12
Wheat and maize zone	Beijing, Tianjin, Hebei, Shanxi, Shandong, Shaanxi, Gansu	0.4	0.8	1.0	1.2	1.2	1.2	1.2	1.0	1.0	0.8	0.6	0.4
	Henan	0.6	0.8	1.0	1.2	1.2	1.2	1.2	1.1	1.0	0.8	0.6	0.4
	Ningxia	0.4	0.8	1.0	1.0	1.0	1.2	1.2	1.0	0.9	0.8	0.6	0.4
Maize zone	Inner Mongolia, Jilin, Heilongjiang	0	0	0	0.6	1.0	1.2	1.2	1.0	0.9	0.4	0	0
	Liaoning	0	0	0	0.8	1.0	1.2	1.2	1.0	0.9	0.4	0	0
Maize and grassland zone	Qinghai, Xinjiang, Tibet	0	0	0	0.6	1.0	1.2	1.2	1.0	0.9	0.4	0	0
Wheat, maize, rice zone	Sichuan, Chongqing, Guizhou, Yunnan	1.0	1.0	1.1	1.2	1.0	1.2	1.2	1.2	1.0	1.0	1.0	1.0
Winter wheat and rice zone	Hubei, Anhui, Jiangsu	1.0	1.0	1.1	1.2	1.0	1.2	1.2	1.2	1.0	1.0	1.0	1.0
Rice zone	Zhejiang, Hunan, Jiangxi, Fujian, Guangdong, Guangxi, Hainan	0.9	0.9	1.0	1.0	1.2	1.2	1.2	1.2	1.0	1.0	0.9	0.9

.

The calculation steps for $SPI{W}_{60}$ are as follows: (1) Calculate the daily weighted cumulative precipitation ($WAP$) over the past 60 days via the following formula:

$$WAP={\sum }_{n=0}^N{a}^n{P}_n,$$

(12)

where $N$ is the time scale, which is 60 days; $a$ is the weight, which is 0.85; and $P_n$ is the precipitation on the nth day before the current day. The value of $a$ is empirically set to emphasize the influence of recent precipitation on drought. (2) Substitute $WAP$ into the SPI formula to obtain $SPI{W}_{60}$.

The formula for $M{I}_{30}$ is as follows:

$$MI={\displaystyle \frac{P-PET}{PET}},$$

(13)

where $P$ represents the cumulative precipitation over a 30-day period and $PET$ represents the cumulative potential evapotranspiration over the same 30-day period.

In this study, daily cumulative precipitation and potential evapotranspiration data for the past 30, 60, 90, and 150 days from 1951 to 2022 are used to calculate the daily MCI (namely, MCI_{HSTR_China}). The meteorological drought classification based on the MCI is shown in

Table 3. Meteorological drought classification based on the MCI.

Level	Type	MCI
1	No drought	−0.5 < MCI
2	Light drought	−1.0 < MCI ≤ −0.5
3	Moderate drought	−1.5 < MCI ≤ −1.0
4	Severe drought	−2.0 < MCI ≤ −1.5
5	Extreme drought	MCI ≤ −2.0

.

2.9. Technical Validation

To evaluate the consistency between the SPI_{HSTR_China} and SPEI_{HSTR_China} datasets generated in this study and the existing SPI_Zhang, SPEI_Zhang, and SPEI-GD datasets, as well as their differences across various regions in China, a pixel-by-pixel analysis using the coefficient of determination (R²), root mean square errors (RMSEs) of the fitting and average biases were conducted to assess the consistency between the SPI_{HSTR_China}/SPEI_{HSTR_China} datasets and the existing datasets across China. Additionally, the differences in the correlations of these datasets across different regions in China were compared to evaluate the reliability of the SPI_{HSTR_China} and SPEI_{HSTR_China} datasets generated in this study, as well as the reliability of the ERA5-Land precipitation data and potential evapotranspiration data used to calculate these indices in various regions of China. To assess the representativeness of the data in the correlation analysis, a significance analysis was conducted via hypothesis testing.

Furthermore, the soil moisture content is a key indicator for measuring drought severity, and the soil moisture content at different depths can reflect drought conditions over varying time spans. In this study, the daily mean values of the MCI and the 30- and 90-day SPI and SPEI for the eastern part of China and its various meteorological and geographical subregions were calculated and used to establish the daily average drought indices in these regions. Within the eastern part of China and its various meteorological and geographical subregions, the correlations between the multiyear MCI_{HSTR_China} dataset (and the SPI_{HSTR_China} and SPEI_{HSTR_China} datasets at different time scales) generated in this study and the ERA5-land soil moisture content data at different depths were subsequently analysed on the basis of R² values. Notably, the characteristics and advantages of the MCI_{HSTR_China} dataset over the SPI and SPEI at different time scales in reflecting drought conditions over different time spans were assessed. Finally, via intergenerational statistical analysis, the variations of the frequency and intensity of droughts since the 1950s in both time and space were revealed. In addition, using a drought event in the mid-lower Yangtze Basin in 2022 as a case study, the application effect of the MCI in drought identification and variation analysis was demonstrated.

Specifically, we initially analysed and evaluated the correlations between the MCI or SPI/SPEI at various time scales and soil moisture at different depths in the eastern part of China in 2022. We also examined the advantages of the MCI in reflecting the variation in the shallow soil moisture content (0–7 cm depth). Subsequently, by incorporating the shallow soil moisture content, we assessed the differences and stability in long-term monitoring capacity between the MCI dataset and the SPI/SPEI datasets produced in this study across different meteorological and geographical regions of the eastern part of China from 1951 to 2022, as well as the variations among areas with distinct climatic characteristics. On the basis of publicly available meteorological drought literature [45,46,47,48], seven years between 1951 and 2022, when key drought events occurred, were selected to evaluate the monitoring capabilities of drought indices in this study.

Finally, to evaluate the actual drought monitoring capability of the MCI dataset produced in this study during the 1951–2022 period, a pixel-by-pixel analysis was conducted. The proportion of days reaching severe drought levels per decade from 1951 to 2020 in China was statistically analysed to examine the variations in drought characteristics across different regions of China, with comparisons and validations against existing literature. Additionally, taking the extreme drought in the mid-lower Yangtze Basin in 2022 as a specific case study, the spatial extent and temporal duration of the drought event in the eastern part of China were analysed via the MCI dataset generated in this study, followed by comparative validation with existing literature.

3. Data Records

This dataset covers the region of China from 1951 to 2022, with a spatial resolution of 0.1, and provides daily SPI, SPEI, and MCI data. The SPI and SPEI data include two time scales of 30 and 90 days. All the data are projected in the geographic lat./lon. projection and stored in GeoTIFF format. The dataset is divided into 3 parts and is openly accessible through the Zenodo platform, available at https://doi.org/10.5281/zenodo.14741907 (accessed on 17 February 2025), https://doi.org/10.5281/zenodo.14760329 (accessed on 17 February 2025), and https://doi.org/10.5281/zenodo.14795177 (accessed on 17 February 2025).

4. Results

4.1. Consistency Analysis of the SPI and SPEI with Existing Datasets

A pixel-by-pixel consistency analysis was conducted for SPI_{HSTR_China}, SPI_Zhang, SPEI_{HSTR_China}, and SPEI_Zhang at 30- and 90-day scales from 1980 to 2018 and for SPEI_{HSTR_China} and SPEI-GD at 30- and 90-day scales from 1983 to 2021 in China. The spatial distribution of the R² values is shown in Figure 3. Figure 4 shows the average R² values, average biases and RMSEs of the fitting for SPI_{HSTR_China} (and SPEI_{HSTR_China}) versus SPI_Zhang (and SPEI_Zhang and SPEI-GD) at various time scales in different meteorological and geographical regions of China. All the specific values in Figure 4 are described in Table S1 in the Supplementary Materials. At both the 30- and 90-day scales, SPI_{HSTR_China} displays high R² values with SPI_Zhang in the eastern part of China. The average R² values exceeding 0.5 in Northeast China, North China, Central China, East China, and South China, with corresponding RMSEs of the fitting below 1. Additionally, the average biases were close to 0 and significantly smaller than the RMSEs of the fitting, indicating minimal systematic deviations. However, in the western regions, especially in Xinjiang and Tibet, the R² values for SPI_{HSTR_China} and SPI_Zhang are lower, with average R² values below 0.3 in some areas and even below 0.1 in parts of these regions, despite the corresponding average biases in these regions also being relatively small. These results reveal that SPI_{HSTR_China} and SPI_Zhang have high correlations and low systematic deviations in the eastern regions of China and are thus more reliably matched; however, in the western regions, particularly in some areas of Xinjiang and Tibet, the consistency is lower, and thus, the uncertainty is greater. These results indicate that the reliability of the ERA5-Land precipitation dataset used for calculating the SPI is greater in the eastern regions than in the western regions of China. The lower R² values (R² < 0.3) in the western part of China, such as Xinjiang and Tibet, may be attributed to the sparse ground meteorological stations and complex terrain, which limits the accuracy of ERA5-Land precipitation estimates in high-altitude arid regions. Similar limitations have been reported in previous studies [42,49], which noted significant biases in reanalysis data over the Tibetan Plateau due to orographic effects. Compared with that at the 90-day scale, the consistency between SPI_{HSTR_China} and SPI_Zhang at the 30-day scale is slightly greater in different regions of China, indicating that SPI_{HSTR_China} at shorter time scales is more reliable than that at longer time scales. The higher consistency observed at shorter time scales, such as 30 days, than that at longer scales, such as 90 days, can be attributed to the dynamic climate response. Short-term drought variations, such as those over 30 days, exhibit lower complexity, and the simulation results from various meteorological reanalysis models are relatively similar overall. In contrast, long-term drought variations, such as those over 90 days, involve complex land–atmosphere feedback mechanisms, including soil moisture–precipitation coupling [50], which leads to greater uncertainty in the model simulations and, consequently, greater discrepancies among the outputs of the different meteorological reanalysis models. The spatial distribution characteristics of the R² values between SPEI_{HSTR_China} and SPEI_Zhang and SPEI-GD are similar to those between SPI_{HSTR_China} and these datasets, and the R² values in different regions are generally higher for the SPEI than those for the SPI, with average R² values reaching approximately 0.6 in Northeast China, North China, Central China, East China, and South China and above 0.3 in Xinjiang and Tibet. Moreover, similar to the analysis results of the SPI datasets, the average biases for SPEI_{HSTR_China} versus SPEI_Zhang (and SPEI-GD) were close to 0 and significantly smaller than the RMSEs of the fitting in most regions, indicating minimal systematic deviations. This finding indicates that SPEI_{HSTR_China}, which considers both precipitation and potential evapotranspiration, is more reliable in different regions of China than the SPI, which considers only precipitation. The potential evapotranspiration data from the ERA5-Land dataset are highly reliable in the region of China and are suitable for calculating drought indices that involve potential evapotranspiration, such as the SPEI and MCI. Like the SPI, the consistency between SPEI_{HSTR_China} and SPEI_Zhang and SPEI-GD at the 30-day scale is slightly greater than that at the 90-day scale of the SPEI in different regions of China, indicating that SPEI_{HSTR_China} is more reliable at shorter time scales than at longer time scales. The R² values between SPEI_{HSTR_China} and SPEI-GD are slightly lower in most regions than those between SPEI_{HSTR_China} and SPEI_Zhang, which is related to the different original spatial resolutions of the two datasets. Given the high reliability of the ERA5-Land precipitation, potential evapotranspiration, SPI_{HSTR_China}, and SPEI_{HSTR_China datasets}, which were calculated on the basis of ERA5-Land datasets for the eastern part of China, the relationships among SPI_{HSTR_China}, SPEI_{HSTR_China}, MCI_{HSTR_China,} daily precipitation, and the soil moisture content at different depths in the eastern part of China are explored in the next section.

4.2. Monitoring Capability of Various Drought Indices for Soil Moisture at Different Depths

On the basis of the drought index dataset generated from this study, daily average processing was performed for the multiscale drought indices (MCI, SPI and SPEI at 30- and 90-day scales), cumulative precipitation, and soil moisture at different depths in the eastern part of China. Figure 5 shows the daily average MCI, SPI and SPEI at the 30- and 90-day scales, as well as the corresponding daily average cumulative precipitation data and soil moisture data at different depths for the eastern part of China from 1 January to 31 December 2022; these values were calculated on the basis of the ERA5-Land dataset. Among the five drought indices studied, the trend of the MCI is most consistent with the characteristics of daily cumulative precipitation changes, followed by those of the SPI and SPEI at the 30-day scale, whose changes slightly lag behind those of the daily cumulative precipitation. The changes in the SPI and SPEI at the 90-day scale significantly lag behind those in the daily cumulative precipitation (as shown in Figure 5a). These results indicate that among the five drought indices, the MCI is most sensitive to daily cumulative precipitation changes and can most accurately reflect the impact of recent precipitation on drought relief in a timely manner. In addition, among the five drought indices, the trend and magnitude of change in the MCI with date are most consistent with the changes in soil moisture at the 0–7 and 7–28 cm depths (as shown in Figure 5b,c), and the R² values between the MCI and soil moisture at the 0–7 and 7–28 cm depths are greater than 0.7, indicating a significant correlation (as shown in Figure 6). All the specific values in Figure 6 are described in Table S2 in the Supplementary Materials. The trends of the SPI and SPEI at the 30-day scale are also consistent with the changes in soil moisture at the 0–7 and 7–28 cm depths, but the degree of fit between the variations in these two drought indices with dates and soil moisture at 0–7 and 7–28 cm is relatively low, with R² values of approximately 0.5, indicating a relatively low correlation. The trends of the SPI and SPEI at the 90-day scale also lag behind the daily variations in soil moisture, which are similar to the findings of previous studies [12,23]; however, compared with those at the 30-day scale, the variations in the SPI and SPEI at the 90-day scale are closer to the variations in soil moisture in the shallow and middle layers, with the R² values for soil moisture at 7–28 cm reaching above 0.6; these results indicate that at the overall level, the SPI and SPEI at the 90-day scale can more accurately reflect the variations in soil moisture in the shallow and middle layers in eastern part of China. Therefore, overall, the variations in soil moisture in the shallow and middle layers in the eastern part of China are affected mainly by short-term droughts. The variation trend of the soil moisture at the 28–100 cm depth is relatively less consistent with the variations in the MCI, SPI, and SPEI at the 30-day scale (as shown in Figure 5d), among which the R² value between the MCI and the soil moisture at the 28–100 cm depth is approximately 0.3, and the R² values between the SPI and SPEI at the 30-day scale and the soil moisture at the 28–100 cm depth are approximately 0.1 (as shown in Figure 6), indicating low correlations. Therefore, the above results indicate that the MCI, SPI, and SPEI at the 30-day scale cannot accurately reflect the changes in soil moisture at the 28–100 cm depth. However, the degree of fit between the soil moisture at the 28–100 cm depth and the changes in the SPI and SPEI at the 90-day scale is high. In addition, the R² values between the soil moisture at the 28–100 cm depth and the SPI at the 90-day scale reach approximately 0.6, indicating a relatively high correlation. Therefore, the SPI at the 90-day scale can accurately reflect the variations in deep soil moisture.

The above results indicate that among the five drought indicators, the MCI can be used to precisely track the daily trends in shallow (0–7 cm) soil moisture changes in approximately real time, followed by those in the moderate-depth layer (7–28 cm). Compared with the SPI and SPEI at the 30-day scale, the MCI has a greater correlation with both shallow and moderate-depth layer soil moisture; thus, the MCI can serve as a quantitative drought indicator that reflects the soil moisture in shallow and medium layers nearly in real time, with significant application potential for drought monitoring of crops with shallow roots, such as grasses and wheat, in the northern part of China. The 90-day-time-scale SPI, on the other hand, can more accurately capture changes in deep soil moisture, making it more applicable for monitoring drought in crops with deep roots (e.g., maize in its later growth stages). Given that in the eastern part of China in 2022, the MCI effectively reflected the daily trends of soil moisture in the shallow layer (0–7 cm) across the entire region in real time, the following analysis delves into the correlation between the MCI and the shallow soil moisture within the five meteorological and geographical regions of eastern part of China and compares it with the SPI and SPEI at 30- and 90-day scales. This approach is used to assess the differences and stability in the multiyear monitoring capabilities of the drought index datasets generated in this study, such as the MCI, between 1951 and 2022, as well as across areas with different climatic characteristics.

4.3. Monitoring Capacity for Shallow Soil Moisture in Different Regions from 1951 to 2022

Via the use of the methodologies outlined in Section 2.9, seven years between 1951 and 2022, when key drought events occurred, were selected to evaluate the monitoring capabilities of drought indices on the basis of publicly available meteorological drought studies. In addition, daily average processing was performed on the MCI, SPI and SPEI data at 30- and 90-day scales in five meteorological and geographical regions in the eastern part of China, as well as soil moisture at 0–7 cm depth, resulting in daily average drought indices and average soil moisture at 0–7 cm for each meteorological and geographical region in the eastern part of China. Linear regression analysis was used to calculate the R² values between each drought index and the 0–7 cm soil moisture in different years, and the results are shown in Figure 7. All the specific values in Figure 7 are described in Table S3 in the Supplementary Materials. In the five meteorological and geographical divisions, compared with the SPI and SPEI, the MCI had higher R² values with 0–7 cm soil moisture in most of the selected years, which is consistent with the results of the correlation analysis between different drought indices and 0–7 cm soil moisture across the eastern part of China in the previous section. These results indicate that, from 1951 to 2022, the MCI dataset generated in this study exhibited greater monitoring capabilities for shallow soil moisture in various meteorological and geographical regions and across different decades than did the SPI and SPEI. In the five meteorological and geographical regions of the eastern part of China, the drought monitoring indices in Central China, East China, and North China presented relatively high R² values with 0–7 cm soil moisture, with the MCIs in these regions having R² values above 0.6 in most years and even reaching above 0.8 in some years. These findings demonstrate that the MCI dataset generated in this study has significant value for monitoring the shallow soil moisture content in Central China, East China, and North China.

Compared with those in the aforementioned three regions, the R² values between the drought monitoring indices and 0–7 cm soil moisture were lower in Northeast China and South China. In these regions, the R² values between the MCI and the 0–7 cm soil moisture in most years are generally less than 0.6. For the SPI and SPEI, in Central China, East China, and Northeast China, the R² values between the 30-day drought index and 0–7 cm soil moisture are significantly greater than those at the 90-day scale in most years, indicating that in these regions, the variation in shallow soil moisture is influenced mainly by short-term droughts. However, in South China and North China, the R² values between the 30-day SPI and SPEI and the 0–7 cm soil moisture are lower than those at the 90-day scale in most years, suggesting that in South China and North China, the variation in shallow soil moisture is influenced by droughts at multiple time scales. Additionally, in Central China, East China, and Northeast China, the R² values between the SPI and 0–7 cm soil moisture in most years before 2011 were higher than those of the SPEI but lower than those of the SPEI in 2011 and 2022. This indicates that before 2011, precipitation had a greater impact on soil moisture, whereas air temperature and potential evapotranspiration had a smaller impact; however, in 2011 and 2022, the impacts of air temperature and potential evapotranspiration on soil moisture increased. These results may be related to the increase in summer temperatures in these regions due to global warming, which in turn led to increased surface evaporation [51,52,53,54].

In South China and North China, the R² values between the SPI and soil moisture at the 0–7 cm depth are generally lower than those of the SPEI in most selected years, indicating that in these two regions, the impacts of temperature and potential evapotranspiration on soil moisture are more significant over the selected years. This may be related to the high temperatures during droughts in these two areas, leading to high potential evapotranspiration. South China is located in southern China, where temperatures and potential evapotranspiration are generally high throughout the year; however, the North China Plain has an open terrain, with sparse rainfall in spring, and temperatures often significantly increase during drought periods, making temperature a key factor influencing droughts [55,56]. The analysis results of the SPI and SPEI for different regions and years reveal that the shallow soil moisture content in each region is influenced by a variety of factors, including precipitation, potential evapotranspiration, and the combined effects of high temperatures and precipitation deficits at different time scales. Therefore, for drought monitoring in eastern part of China, it is particularly necessary to construct a drought index (such as the MCI) that comprehensively considers various drought-influencing factors, such as precipitation and potential evapotranspiration, over multiple time scales.

4.4. Variance in Drought in China Since the 1950s Based on the MCI Dataset

The above analysis indicates that drought monitoring based on the MCI provides significant advantages across different climatic zones and periods, especially in terms of real-time response and high correlation with shallow soil moisture. To further explore the spatiotemporal variation characteristics of drought in China since the 1950s and to verify the applicability of the MCI dataset in long-term drought variation characteristic analysis, the interdecadal spatiotemporal variation characteristics and seasonal patterns of drought events in China are analysed in this section on the basis of the MCI dataset from 1951 to 2022. Through pixel-by-pixel analysis, the proportion of days with severe drought levels or above in China from 1951 to 2020 for each decade was calculated, and the results are shown in Figure 8. Additionally, the daily average MCI values for each meteorological and geographical region in the eastern part of China from 1951 to 2020 were tabulated, and the results are shown in Figure 9. Since 1951, the frequency of severe droughts in China has fluctuated. Prior to the 1970s, severe droughts were concentrated mainly in the mid-lower Yangtze Basin, South China, Xinjiang, and the western part of the Northwest Region, especially in the mid-lower Yangtze Basin and South China regions, where severe droughts often occurred in autumn (October to November) and winter (December to February of the following year). The areas with high frequencies of severe droughts subsequently gradually spread to North China, the eastern part of the Northwest Region, and Northeast China, with increased frequencies in spring (March–April) and early summer (May–June). Related studies have indicated that the southwards shift in the northern boundary of the summer monsoon since the 1980s has led to reduced precipitation in North China [3], whereas the abnormal westwards extension of the subtropical high has shortened the rainy season in the mid-lower Yangtze Basin [57]. Such circulation adjustments have resulted in the frequent occurrence of spring droughts in North China and summer droughts in the Yangtze Basin [45]. In the 21st century, areas with high frequencies of severe droughts in China have significantly expanded compared with those in the 20th century, covering all meteorological and geographical regions in the eastern part of China, and periods of severe droughts have occurred throughout all seasons of the year. Moreover, both the number and intensity of extreme drought events have significantly increased; for example, the mid-lower Yangtze Basin experienced the most severe meteorological drought since the 1960s in 2011 [57,58], whereas the drought event in the eastern region in 2022 surpassed the records from 2011 in terms of both area and intensity [59,60,61]. Among these drought events, the middle–lower Yangtze Basin and its northern areas have been hotspots for frequent severe droughts. Relevant studies have revealed that, owing to the dual stresses of reduced precipitation and increased evapotranspiration, the occurrence frequency of spring droughts in North China has significantly increased compared with that during the 1980s [53]. By 2050, the return period witnessed 2022 compound high-temperature and drought events in the mid-lower Yangtze Basin would likely be reduced by one-third [59]. Our findings conform with emerging research on drought non-stationarity under climate change [17,19,35]. The intensified droughts in North China and the mid-lower Yangtze Basin after 2000 (Figure 8) can be attributed to the combined effect of extreme precipitation deficits and temperature-driven potential evapotranspiration increases.

Figure 9 shows that the drought phenomena in different meteorological and geographical regions of the eastern part of China exhibit a significant seasonal pattern. Spring droughts are concentrated mainly in North China and often extend into early summer. Given that North China is a major planting area for winter wheat nationally, the persistence of spring droughts is likely to significantly lead to a decrease in winter wheat yield [62,63,64]. Summer droughts can be further divided into early summer droughts and mid-summer droughts, with early summer droughts occurring frequently from May to June in various regions of the eastern part of China, whereas midsummer droughts are more common in the Central, East, and South China regions from July to September. During periods of high temperature in summer, crops require a large amount of water, so long periods of mid-summer droughts can have a significant negative impact on the growth and yield of crops such as rice and corn and, in extreme cases, can even lead to crop death [65]. In addition, prolonged midsummer droughts can also cause declines in river, lake, and groundwater levels, thereby affecting industrial production; shipping; socioeconomic activities; and the lives of residents [66]. Autumn and winter droughts affect mainly the Central, East, and South China regions and typically last from October to January or February of the following year. In southern regions, autumn droughts can have a significant effect on the growth of late rice [67]. The above information regarding the spatiotemporal variation characteristics in drought areas in China since the 1950s and the seasonal characteristics of droughts are highly consistent with previous research findings [53,68,69,70,71], thus verifying the reliability of the MCI dataset generated in this study for analysing spatiotemporal drought variation characteristics.

4.5. Analysis of the 2022 Drought in Eastern Part of China

The analysis in the previous section revealed that since the beginning of the 21st century, the frequency of severe drought events in the eastern part of China has significantly increased. In particular, from spring to autumn in 2022, the eastern part of China, which is centred on the mid-lower Yangtze Basin, experienced severe drought. Therefore, the MCI dataset generated in this study was utilized to conduct an in-depth analysis of the spatial extent and temporal duration of drought in the eastern part of China in 2022. Figure 10 shows the distribution of days reaching severe drought status across various meteorological and geographical regions of China throughout 2022, as well as the first and second halves of the year. Except in Northeast China and the southern part of South China, most of the eastern part of China experienced more than 10 days of severe drought or above in 2022. The areas with more days of severe drought or above were mainly concentrated in Central China and East China, where most areas in East China had more than 40 days, and most areas in Central China had more than 70 days, with some areas exceeding 100 days at these drought levels. Figure 11 shows the daily areas of severe drought or above and the average MCI values in the drought regions in the eastern part of China in 2022. Continuous severe drought began in April and lasted until the end of November. The entire process can be divided into two main drought stages. The first stage lasted from early April to early July. According to Figure 10b, the drought affected mainly the southern part of North China, the northern part of Central China, and the northern part of East China during this period. The maximum daily area of severe drought reached 619,479 square kilometres, occurring on May 23rd; the minimum daily average MCI value in the drought regions was −2.24, reaching the level of extreme drought. The second drought stage lasted from the end of July to the end of November. According to Figure 10c, drought affected mainly the central and southern parts of Central China and East China (the mid-lower Yangtze Basin). The maximum daily area of severe drought reached 1,201,793 square kilometres, occurring on September 29th; the minimum daily average MCI value of the drought regions was −2.36. Compared with the previous drought that occurred in northern China, this drought in the mid-lower Yangtze Basin was more severe, wider in impact range, and longer in duration. The above analysis results for drought duration and spatial extent based on the MCI dataset are highly consistent with those published previously [72,73], which verifies the reliability of the MCI dataset generated in this study for analysing spatiotemporal changes in specific drought events.

5. Conclusions and Discussion

Given the vast territory of China, with its diverse terrain and climate types, the factors influencing droughts vary in different regions and seasons, as do the degrees of drought, which are influenced by various factors at different time scales. Therefore, there is an urgent need for a drought index dataset that can comprehensively consider the multiple factors influencing droughts and be applicable to different seasons. However, the existing HSTR gridded meteorological drought index datasets are limited to a single time scale and do not fully consider the impact of droughts on vegetation in different seasons, which restricts their application in precise drought monitoring across various regions of China. In this study, daily cumulative precipitation data and potential evapotranspiration data from ERA5-Land were used to generate daily MCI and 30- and 90-d SPI and SPEI datasets for the Chinese region at a spatial resolution of 0.1 degrees from 1951 to 2022, aiming to reduce the time and cost of drought monitoring in regions of China. Compared with the existing SPI and SPEI datasets, the datasets generated in this study show a higher degree of agreement with those in the eastern part of China (R² > 0.5; average biases were close to 0 and significantly smaller than the RMSEs of the fitting). Furthermore, considering the complexity of meteorological drought in the Chinese region, the MCI dataset displays a stronger correlation with the shallow soil moisture content in various meteorological and geographical divisions in the eastern part of China than does the single-time-scale SPI and SPEI and can more accurately reflect the changes in shallow soil moisture content in the eastern part of China in a timely manner (R² can reach up to 0.7 or greater for the 0–7 cm depth). Therefore, it is more suitable for research on agricultural and grassland drought development and for extended research on the corresponding effects on ecosystems and socioeconomics. However, the lower correlation of MCI with deep soil moisture (28–100 cm, R² ≈ 0.3) indicates its limited applicability for plants with deep root systems (e.g., maize and trees). For such scenarios, integrating longer-scale indices (e.g., SPI and SPEI at a 90-day scale or greater) is recommended, as they can better capture the long-term agricultural and hydrological drought impacts [12,23]. Finally, on the basis of the MCI dataset generated in this study, the regional and seasonal variation characteristics of droughts in China since the 1950s were accurately analysed, and the drought development process, duration, spatial range, and development trend in the eastern part of China in 2022 were accurately identified, maintaining good consistency with the results reported in previous papers. Nevertheless, there is still room for improvement and further research on the MCI dataset generated in this study.

Automated and precise data extraction and case library construction for drought events are crucial for studying the formation and development processes of droughts. In recent years, to improve the accuracy and automation level of drought identification, point clustering methods such as DBSCAN have been widely applied [72,74]. Liu et al. [45] and Liu et al. [72] further developed the 3D DBSCAN algorithm based on the 90-day SPI to automatically identify three-dimensional (latitude–longitude–time) drought events. It was successfully applied for the identification of significant drought events in the mid-lower Yangtze Basin and nationwide in 2022. Therefore, in the future, we will attempt to use algorithms such as 3D-DBSCAN, which is based on the MCI dataset generated in this study, to construct a case library of drought events in the region of China from 1951 to 2022.

Furthermore, although the MCI drought index dataset developed on the basis of ERA5-Land meteorological reanalysis data can accurately reflect the changes in shallow soil moisture content in the eastern part of China in a timely manner, in the western part of China, due to factors such as the scarcity of observation stations, the uncertainty of meteorological reanalysis data is relatively high [42,49]. In the future, additional meteorological observation experiments will be conducted, and additional meteorological observation stations will be deployed in the western part of China. This will enable further accuracy assessment and error correction of the drought index datasets produced in this study for these areas. Additionally, the spatial resolution of current meteorological reanalysis datasets, such as ERA5-Land, remains relatively low, reaching a maximum of approximately 0.1 degrees. Satellite remote sensing monitoring is one of the main methods for current drought monitoring [75,76,77]. Compared with meteorological observation data from ground stations and meteorological reanalysis data, remote sensing drought monitoring results can provide a higher spatial resolution and can achieve the same spatial resolutions and stable, reliable monitoring results in areas with sparse populations and few ground observation stations, such as densely populated areas. Notably, researchers have developed many drought indices based on information from various bands, such as the visible light, near-infrared, shortwave infrared, thermal infrared, and microwave bands, and have successfully applied this information for drought monitoring in different regions worldwide [78,79]. However, compared with drought index datasets and drought event case libraries based on ground station and meteorological reanalysis data, relatively little research has been conducted on large-area (e.g., the entire Chinese region) long-term drought index datasets and drought event identification based on remote sensing data. Moreover, comparisons of remote sensing drought indices and common meteorological drought indices, such as the MCI, are lacking. In the future, with the MCI dataset developed in this study as a benchmark, further development of large-area quantitative drought monitoring methods comparable to the MCI drought index should be explored on the basis of remote sensing data, and a nationwide drought index dataset spanning over 20 years should be established.