Variations in the Thermal Low-Pressure Location Index over the Qinghai–Tibet Plateau and Its Relationship with Summer Precipitation in China

Abstract

The thermal and dynamic effects of the special topography of the Qinghai–Tibet Plateau have a significant impact on rainfall in China. Utilizing NCEP/NCAR monthly reanalysis data alongside precipitation observations from 1936 monitoring stations across China spanning from 1966 to 2022, this study establishes a location index for the thermal low-pressure center situated over the Qinghai–Tibet Plateau. Temporal variations in the location index and summer (July) precipitation patterns in China were studied. Over the past six decades, thermal low-pressure centers have been predominantly positioned near 90° E and 32.5° N within a geopotential height of 4360 gpm, with their distribution extending from east to west rather than from south to north. The longitudinal and latitudinal position indices showed the same linear trend, with a negative trend before the 21st century, and then began to turn positive. Mutation analysis highlights pronounced weakening mutations occurring in 1981 and 1973, with the longitudinal index transitioning from an interannual cycle of approximately 6–8 years, while the latitudinal index displays quasi-cyclic oscillations of 5 and 8 and 12–14 years. Strong negative correlations are evident between the location indices and precipitation along the southeastern edge of the Qinghai–Tibet Plateau and in southern China, contrasting with the positive correlations observed in the central-eastern plateau, northwest, north, and the Huang-Huai region of China. The center of the thermal low is located to the east and north, corresponding to the deeper surface thermal low in most areas east of China, and the stronger transport of warm and wet air from the southwest wind, leading to greater convergence of southwest wind and northwest wind in China’s northern region. The south of the Yangtze River is controlled by the strengthening West Pacific subtropical high and South Asia high, resulting in a significant decrease in precipitation, and the warm and humid air from the southwest on the west side of the West Pacific subtropical high is also transported to the north, increasing the precipitation in most parts of the north.

1. Introduction

Known as the third pole of the earth, the Qinghai–Tibet Plateau is located in the central and southern parts of Asia. Its uplift has changed climate patterns in Asia [1]; its climate changes before the rest of China [2]. In the past 60 years, the Qinghai–Tibet Plateau has been the fastest-warming region in China, more than twice the global warming rate over the same period. The frequency of high temperature extremes and heavy precipitation events has increased significantly. The Qinghai–Tibet Plateau has important impacts on circulations over and around it and influences climate in China and the world [3,4,5,6]. Therefore, the Qinghai–Tibet Plateau has attracted much attention. The thermal low-pressure on the plateau is one of the systems that can affect weather and climate downstream. However, more attention is focused on other large-scale systems such as plateau monsoons [7,8], plateau heat sources [9,10], etc., but less attention is paid to the thermal low pressure on the plateau.

Thermal low-pressure systems are nonfrontal cyclones that appear near the surface layer. They are shallow and less moving, warm, closed, low-pressure systems due to the uneven heating of the air near the surface layer and are mostly present in summer. Among the most extensively studied is the thermal low-pressure system of the Saudi Peninsula, which is primarily driven by boundary layer dynamics and underlying surface influences [11,12,13,14]. Beyond the thermal low-pressure phenomenon of the Qinghai–Tibet Plateau, China also experiences the southwest thermal low-pressure and southern Xinjiang thermal low-pressure systems [15,16]. Observations indicate that at 600 hPa, the Tibetan Plateau exhibits low pressures during summer and high pressures during winter [7]. By employing an intensity index to gauge the strength of the thermal low-pressure system over the Tibetan Plateau, a general weakening trend was observed in terms of the summertime low pressure over the Qinghai–Tibet Plateau spanning the last 62 years [17]. Additionally, longitude and latitude indexes have been utilized to elucidate the east–west and north–south spatial variations in the southwest thermal low-pressure system’s position. These indexes, derived as averages across all stations within the thermal low-pressure study area, have facilitated discussions on the spatiotemporal distribution characteristics of the southwest thermal low-pressure system [18,19].

In China, the occurrence of extreme summer precipitation poses a significant threat to people, property, and the national economy, underscoring the urgent need for in-depth investigations. Many scholars have researched the relationship between thermal low pressure and heavy rainfall [20,21,22]. However, due to the intricate interplay of dynamic and thermodynamic atmospheric processes on the plateau, research on summer rainfall induced by plateau thermal low pressure in China remains limited. Prior studies have suggested that the plateau’s atmospheric heat sources serve as crucial indicators of summer precipitation in regions such as Jianghuai, South China, and North China. Enhanced plateau heat sources correspond to increased precipitation in the upper reaches of the Yangtze and Huaihe River basins and decreased precipitation in southeast and North China. The fluctuation in heat sources over the Qinghai–Tibet Plateau correlates closely with variations in thermal low-pressure intensity [23,24]. Consequently, fluctuations in thermal low pressure over the plateau may lead to variations in summer precipitation across China. Specifically, during summers characterized by abnormally strong thermal low pressure over the Qinghai–Tibet Plateau, precipitation in the upper, middle, and lower reaches of the Yangtze River and the northwest region of the plateau increases abnormally. Conversely, precipitation totals are notably diminished in the western Sichuan Basin, the Hetao area, and North China [7]. The intensification of plateau thermal low pressure typically results in increased precipitation in the Yangtze River Basin and Xinjiang, while precipitation is reduced in northeast China, North China, southwest China, and South China, and vice versa [17].

Currently, there is a scarcity of research focusing on the thermal low-pressure system over the Qinghai–Tibet Plateau. Compounded by the region’s complex terrain and limited observational data, there is a lack of standardized definitions for the circulation index utilized to characterize its active center. Addressing this gap, this paper proposes a method for defining a Qinghai–Tibet Plateau thermal low-pressure location index. Additionally, this study investigates the interannual and interdecadal variations in the thermal low-pressure location index over the Qinghai–Tibet Plateau. By analyzing the spatial and temporal variability of the thermal low-pressure system’s location, this research aims to elucidate its relationship with summer precipitation patterns across China. Ultimately, the findings of this study aim to provide a theoretical framework for enhancing summer precipitation forecasting capabilities in China.

2. Data and Methods

The summertime (July) monthly mean height and wind field from the reanalyzed NCEP/NCAR data were used, and the horizontal grid resolution was 2.5° × 2.5°. Since there was no low-pressure center in the 600 hPa altitude field for many years in the main area of the plateau from 1948 to 1965, we used data from 1966 to 2022. The monthly precipitation data from 1936 stations in China during summer (July) from 1966 to 2022 were also used. The main range of the Qinghai–Tibet Plateau was determined to be 27.5°~40° N and 80°~102.5° E. However, due to the difference between the model terrain used in the assimilation system and the real terrain, there may be certain limitations.

In terms of data selection, Bai Huzhi et al. [7] proved that the 600 hPa height of the NCEP/NCAR reanalysis data can be applied to plateau areas. Some researchers (Tang Maocang [8]; Qi Dongmei et al. [25]) also used these data to conduct plateau research. In addition, since 600 hPa is closer to the surface of the Qinghai–Tibet Plateau in summer, it is affected by surface heating on the Qinghai–Tibet Plateau, and this is a good representative of the thermal low pressure of the plateau. Therefore, the 600 hPa monthly average of the NCEP/NCAR reanalysis data was used to discuss the characteristics of the thermal low pressure on the Qinghai–Tibet Plateau.

The definition of the location index of thermal low pressure is based on the definition of the polar vortex location index. Because the thermal low pressure is shallow and the range is relatively small, it is impossible to uniformly select a closed center line to determine the position of the center of low pressure in a long-term time series. Moreover, due to there being minimal variabilities in the geopotential height of the Qinghai–Tibet Plateau and the resolution of the data used (2.5° × 2.5°), the lowest point of the thermal low-pressure center over the Qinghai–Tibet Plateau will be the same for many consecutive years. Therefore, closed center lines of low pressure with the lowest values were selected for each year. When there were two or more centers with the same center value, the low-pressure center closest to the center of the plateau was selected. Surrounded by a designated closed center line, the mean latitude and longitude of all grid points represent the central location of the plateau thermal low pressure. When the longitude and latitude indexes are high, this indicates that the central locations of the Qinghai–Tibet Plateau thermal low pressure are inclined to the east and north, respectively, and vice versa.

First, we used the Mann–Kendall method [26] and moving t-test [26] to analyze the mutations relating to the thermal low location index and used Morlet wavelet analysis [26] to discuss their periodic variation. Then, we calculated the correlation coefficients between the thermal low-pressure location index and atmospheric circulation to analyze the dynamic mechanisms.

2.1. Mann–Kendall Method

The Mann–Kendall method [26] is a nonparametric statistical test method that can detect not only the change in a sequence but also the turning points in a sequence. For the time series ${\mathrm{x}}_1$, ${\mathrm{x}}_2$, …, ${\mathrm{x}}_{\mathrm{n}}$, construct an order column as follows:

$${\mathrm{S}}_{\mathrm{k}}={\sum }_{\mathrm{i}=1}^{\mathrm{k}}{\mathrm{R}}_{\mathrm{i}}\hspace{1em}(\mathrm{k}=2,3,\dots ,\mathrm{n})$$

(1)

$${\mathrm{R}}_{\mathrm{i}}=\left\{\begin{array}{cc}+1& {\mathrm{x}}_{\mathrm{i}}>{\mathrm{x}}_{\mathrm{j}}\\ 0& {\mathrm{x}}_{\mathrm{i}}\le {\mathrm{x}}_{\mathrm{j}}\end{array}\right.\hspace{1em}(\mathrm{j}=1,2,\dots ,\mathrm{i})$$

(2)

S_k represents the cumulative count of sample x_i greater than x_j (1 ≤ j ≤ i). Under the assumption of the independence of a random time series, we define UF_k as follows:

$${\mathrm{UF}}_{\mathrm{k}}={\displaystyle \frac{{\mathrm{s}}_{\mathrm{k}}-\mathrm{E}\left({\mathrm{s}}_{\mathrm{k}}\right)}{\sqrt{\mathrm{v}\mathrm{a}\mathrm{r}\left({\mathrm{s}}_{\mathrm{k}}\right)}}}\hspace{1em}\mathrm{k}=1,2,\dots ,\mathrm{n}$$

(3)

Here, k = 1, UF₁ = 0, and E(S_k), var(S_k) are the mean and variance of the cumulative counts, respectively; they are calculated as follows:

$$\mathrm{E}({\mathrm{S}}_{\mathrm{k}})={\displaystyle \frac{\mathrm{n}(\mathrm{n}-1)}{4}}$$

(4)

$$\mathrm{var}({\mathrm{S}}_{\mathrm{k}})={\displaystyle \frac{\mathrm{n}(\mathrm{n}-1)(2\mathrm{n}+5)}{72}}$$

(5)

Given the significance level α, Uα represents a normal distribution. If |UF_i| > Uα, this indicates that there is an obvious change in the sequence.

In the reverse order of time series x, x_n, x_n−1, …, x₁, we repeat the above process, while ensuring that UB_k = −UF_k (k = n, n − 1, …, 1), UB₁ = 0.

(1)

Compute sequential time series for order columns, and UF_k

(2)

Calculate the order columns of the reverse time series, and UB_k

(3)

Given the significance level, such as a = 0.05, then the critical value u_0.05 = ±1.96, the two statistical sequence curves of UF_k and UB_k and the two lines of ±1.96 are drawn on the same graph; if the value of UF_k or UB_k is greater than 0, it indicates that the sequence has an upward trend; otherwise, it is decreasing. When they pass the critical value line, it indicates that the upward or downward trend is significant. If the two curves intersect, and the intersection point is between the critical value, then the moment corresponding to the intersection point is the time when the mutation begins.

2.2. Moving t-Test

The moving t-test [26] is designed to test whether the mean values of two segments in a climate sequence are significantly different. If the mean difference between the two sequences exceeds a certain significance level, it can be considered that the mean has undergone qualitative change and a trend change has occurred.

For time series x with n samples, a certain point is artificially set as the reference point, and the samples of the two subsequences before and after the reference point are n₁ and n₂, the mean value of the two subsequences is x₁ and x₂, and the variance is s₁ and s₂, respectively. We define the statistics as follows:

$$\mathrm{T}=(\overline{{\mathrm{x}}_1}-\overline{{\mathrm{x}}_2})/(\mathrm{S}\sqrt{({\displaystyle \frac{1}{{\mathrm{n}}_1}})+({\displaystyle \frac{1}{{\mathrm{n}}_2}})})$$

(6)

$$\mathrm{S}=\sqrt{({\mathrm{n}}_1{\mathrm{s}}_1^2+{\mathrm{n}}_2{\mathrm{s}}_2^2)/({\mathrm{n}}_1+{\mathrm{n}}_2-2)}$$

(7)

The expression follows the t distribution of freedom v = n₁ + n₂ − 2.

2.3. Morlet Wavelet

Wavelet analysis [26] is also known as multiresolution analysis, which is considered to be a breakthrough in terms of Fourier analysis methods. By decomposing the time series into the time-frequency domain, wavelet transform can obtain the significant wave pattern of the time series, namely the periodic change dynamic and the time pattern of the periodic change dynamic.

Wavelet transform is divided into continuous wavelet transform and discrete wavelet transform.

Morlet wavelet is not only nonorthogonal, but also an exponential complex wavelet regulated by a Gaussian function.

$${\mathsf{\Psi }}_0(\mathrm{t})={\mathsf{\pi }}^{{\displaystyle \frac{-1}{4}}}{{\mathrm{e}}^{\mathrm{i}{\mathrm{t}\mathsf{\omega }}_0}\mathrm{e}}^{{\displaystyle \frac{-{\mathrm{t}}^2}{2}}}-\infty <\mathrm{t}<\infty$$

(8)

where t represents time and ${\mathsf{\omega }}_0$ represents dimensionless frequency, and when ${\mathsf{\omega }}_0$= 6, the wavelet scale s is basically equal to the Fourier period; therefore, the scale term and the period term can be substituted for each other. The Morlet wavelet maintains a fine balance between the localization of time and frequency.

Next, the wavelet power spectrum is time-averaged over a certain period, and the global wavelet spectrum can be obtained as follows:

$${\overline{\mathrm{W}}}^2(\mathrm{s})={\displaystyle \frac{1}{\mathrm{N}}}{\sum }_{\mathrm{n}=0}^{\mathrm{N}-1}{\left|{\mathrm{W}}_{\mathrm{n}}\left(\mathrm{s}\right)\right|}^2$$

(9)

N represents the total time data number of the time series, and s represents the wavelet scale.

$${\mathrm{s}}_{\mathrm{j}}=s_0{2}^{\mathrm{j}{\mathsf{\delta }}_{\mathrm{j}}}\hspace{1em}\mathrm{j}=0,1,\dots ,\mathrm{J}$$

(10)

$$\mathrm{J}={(1/\mathsf{\delta }}_{\mathrm{j}}){\log }_2(\mathrm{N}{\mathsf{\delta }}_{\mathrm{t}}/{\mathrm{s}}_0)$$

(11)

${\mathrm{s}}_0$ is the smallest scale that can be resolved, and J is the largest scale that can be determined. ${\mathrm{s}}_0$ should be chosen appropriately so that the Fourier period is approximately 2${\mathsf{\delta }}_{\mathrm{t}}$, and the choice of a sufficiently small ${\mathsf{\delta }}_{\mathrm{j}}$ depends on the width of the spectral space of the wavelet equation.

3. Results

3.1. The Variation Characteristics of the Thermal Low Location Index over the Qinghai–Tibet Plateau

The Qinghai–Tibet Plateau is mainly located in the west of China, the terrain is high in the west and low in the east (Figure 1a), and the region contains the world’s highest peak, Mount Everest (8844.43 m). By averaging the monthly mean reanalysis data of the summer NCEP/NCAR height field, we found the location of the center of summer thermal low-pressure events during 1966–2022. The centers were mainly located around 90° E and 32.5° N within a geopotential height of 4360 gpm (Figure 1b). This area is considered the abdomen of the Qinghai–Tibet Plateau, about 5000 m above sea level, and the distribution extended more significantly from east to west than from south to north. There was no thermal low-pressure center in the main area of the plateau in 1994 and 1997, and we chose locations with the lowest geopotential height in the main area of the Qinghai–Tibet Plateau.

The longitude index showed an overall linear downward trend, with a trend coefficient of −0.023 (Figure 2a), while the latitude index showed an upward trend, with a trend coefficient of 0.013 (Figure 2b), both of which declined before the 21st century and began to rise in the 21st century; however, the latitude index increased more significantly. This indicates that before the 21st century, the thermal low-pressure center on the Qinghai–Tibet Plateau moved to the southwest, corresponding with research on the plateau heat source described by Feng Song [27], Wei Zhigang et al. [28], and Hu Jun et al. [29]. The warming trend in the southwest is significantly stronger than that in the east. However, since the beginning of the 21st century, these two indices have begun to move east and north, so the low-pressure center is moving northeastward.

In order to further explore the decadal variation in Qinghai–Tibet Plateau thermal low pressure, we calculated this anomaly for each decade (

Table 1. Interdecadal anomalies in the longitude and latitude location indexes over the Qinghai–Tibet Plateau in summer during 1966–2022.

Years	Anomaly (lon)	Anomaly (lat)
1970–1979	1.5748	0.4478
1980–1989	−0.6472	−0.5122
1990–1999	−1.1592	−1.0472
2000–2009	0.3338	0.0068
2010–2019	−0.1022	1.1048

).

$$\mathrm{y}={\mathrm{x}}_1-{\mathrm{x}}_2$$

(12)

The time anomaly is y, ${\mathrm{x}}_1$ is the mean value of the longitude/latitude in each decade, and ${\mathrm{x}}_2$ is the mean value of the longitude/latitude during 1966–2022.

The longitude and latitude location indexes of Qinghai–Tibet Plateau thermal low pressure were positive in the 1970s and 2010s, indicating that they were high (northeast) during these periods, but negative in the 1980s and 1990s, indicating that the low-pressure location indexes were low (southwest) during these periods. The longitude and latitude indexes had the largest absolute negative values in the 1990s, indicating that thermal low pressure was positioned more southwesterly in this era. The longitude index reached its maximum positive value in the 1970s, while the latitude index reached its maximum positive value in the 2010s, indicating that Qinghai–Tibet Plateau thermal low pressure was positioned significantly more north (east) than it had been during the rest of this period. The same conclusion can be drawn from the polynomial fitting curve shown in Figure 2. It can be seen that from the late 1970s to the early 1980s, the decadal variation in the longitude and latitude indexes of the Qinghai–Tibet Plateau thermal low pressure suddenly changed from high to low, leading to the mutation in terms of time discussed below.

The negative curves of the longitude index (Figure 3a) from the 1970s to the early 1980s and in the late 2010s were greater than 0, and the latitude index (Figure 3b) from the late 1960s to the early 1970s indicates an exponential upward trend. There was a significant downward trend before the 21st century, and the trend from the mid-1990s to the early 21st century exceeded the 95% confidence level. At the beginning of the 21st century, there was a significant upward trend again, which was consistent with the analysis of the time series anomaly. The positive and negative series of the longitude index intersected in 1981, 2015, 2017, 2020 and 2021, while the positive and negative curves of the latitude index intersected in 1973, 2021 and 2022. However, moving t-test analysis confirmed that the 1981 intersection of the longitude index represents climate change, while the 1973 intersection of the latitude index represents climate change.

Morlet wavelet transform further shows the temporal change in longitude and latitude location indexes of the Qinghai–Tibet Plateau thermal low pressure in the last 60 years. The cycle of the longitude index (Figure 4a) grew slightly before the 1990s and then stabilized at about 8 years. The longitude index had an interdecadal cycle of about 18 years, but it has gradually increased since the early 21st century. The 6–8-year cycle passed the 90% confidence test in the late 1970s through to the early 21st century. However, the latitude index (Figure 4b) had two interannual cycles of 5 and 8 years; the former was mainly assigned to the late 1990s, and it has a gradually shortened trend. The latter increased to about 12 years in the 1990s and then basically stayed the same at about 14 years. The 5-year cycle passed the 90% confidence test from the 1970s to the late 1990s, and the 8–14-year cycle passed the 90% confidence test from the late 1970s to the early 21st century.

3.2. The Relationship between Qinghai–Tibet Thermal Low Pressure and Summer Precipitation and Atmospheric Circulation in China

We analyzed the correlation between the longitude and latitude location indexes of Qinghai–Tibet Plateau thermal low pressure and the monthly precipitation data of 1936 stations in China, alongside the U and V wind fields of the 850 hPa, 1000 hPa, 500 hPa, and 100 hPa height fields from 1966 to 2022. Then, the years with high and low Qinghai–Tibet Plateau thermal low-pressure location indexes were selected for analysis. When the longitude/latitude index is greater than or less than 1° above the multi-year average, it is selected as a high/low-value year. The years in which the longitude index is higher are 1968, 1969, 1971, 1976, 1979, 1982, 1992, 2006, 2012, and 2013. The years featuring a low longitude index are 1978, 1986, 1994, 2010, 2011, and 2014. The years with a high latitude index are 1966, 1976, 2013, 2016, 2018, 2019, and 2022, while the years with a low value are 1977, 1978, 1986, 1994, and 1997. The summer precipitation value, 850 hPa wind field, and the 1000 hPa, 500 hPa, and 100 hPa height fields in the years with a high/low-pressure index were, respectively, synthesized to create difference plots (values in all the high-index years minus those of the low-index years) (figure omitted); these were then compared with the correlation coefficient plot.

The longitude index and precipitation (Figure 5a) were negatively correlated in the northwest of Xinjiang, the southeastern edge of the Qinghai–Tibet Plateau, the southern part of southwest China, the central part of South China, and most of the northeast regions. Among them, the southeastern edge of the Qinghai–Tibet Plateau and the southern and eastern parts of South China are the most significant, and the absolute value of the t-test was above 0.21, passing the 90% reliability test. There was a positive correlation between most of the northwest region and most of the Qinghai–Tibet Plateau region and east of the Huanghuai coastal region. The correlation between the Qinghai–Tibet Plateau and northwest region passed the 90% reliability test.

The correlation between the latitude index and precipitation (Figure 5b) is almost the same as that of the longitude index, except that most of the southwest region has a negative correlation, and the relationship is as significant as that in Central China. The north of the northeast region has a minor positive correlation with the northwest area, while the central and eastern parts of the Qinghai–Tibet Plateau and the Huanghuai area have larger correlation values. These results show that when the thermal low pressure is located in the east or north (west or south) of the Qinghai–Tibet Plateau, the level of precipitation over the southeastern edge of the Qinghai–Tibet Plateau, South China, Central China, and the south and northeast parts of China is significantly lower (higher). The level of precipitation is higher (lower) in most parts of northwest China, most of the plateau, and to the east of the Huang-Huai coastal areas.

The map showing the difference In years with high and low location Indexes Is basically consistent with the correlation coefficient, indicating that less precipitation occurred in the years with high longitude and latitude indexes than in the years with low values from the southeastern edge of the Qinghai–Tibet Plateau to South China. In the eastern and northern parts of the Qinghai–Tibet Plateau, more precipitation occurred in the years with high longitude and latitude indexes than in the years with low values. This shows that the high and low Qinghai–Tibet Plateau thermal low position indexes in this paper are reliable.

3.3. Geopotential Height at 1000 hPa

Due to thermal differences between the ocean and land, changing sea-level pressure is very important for predicting summer precipitation. Therefore, the changes in the longitude and latitude indexes of Qinghai–Tibet Plateau thermal low pressure and sea-level pressure were compared and analyzed. The distribution of the correlation coefficients of the longitude and latitude indexes of Qinghai–Tibet Plateau thermal low pressure at the 1000 hPa height are very similar (Figure 6a). Except for the eastern margin of northeast China, the Indian Ocean, the South China Sea, and the West Pacific region, most other regions are negatively correlated. The central and eastern parts of northwest China, the central and eastern parts of the Qinghai–Tibet Plateau, and the northern parts of southwest China pass the 90% reliability test. Thermal low pressure is located further north and east. The lower the air pressure in most of the country, the stronger the heat, the easier it is to condense after the air flow rises, and the more conducive this process to rainfall.

When the longitude index of thermal low pressure on the Qinghai–Tibet Plateau was high (Figure 6b), most of the country was controlled by low pressure. The Mongolia low-pressure system was located at the junction of southern Mongolia and Inner Mongolia, with a central intensity of 20 gpm, and two centers were located in Central China and Xinjiang, respectively. The Mongolian low-pressure system was located in only two east–west centers in Xinjiang and Inner Mongolia. However, in the low-value years relating to the two areas (figure omitted), the Mongolian low-pressure system was weak. The chart showing the difference in summertime sea-level pressure in the high- and low-value years is basically consistent with the correlation coefficient (figure omitted) and is also consistent with the abovementioned analysis results concerning precipitation.

3.4. The 850 hPa Wind Field

The longitude (Figure 7a) and latitude indexes (Figure omitted) are basically consistent with the correlation coefficient diagram of the 850 hPa wind field distribution in the summer. There is a strong cyclone in central Mongolia; a southeasterly air flow in the southern part of the Pacific anticyclone passes over the South China Sea and is divided into two branches over the Indo-China Peninsula, one of which moves northward from South China, along with western air flow in the northern part of the Bay of Bengal. In the eastern part of northwest China, it meets the northwesterly air flow from the Arctic Ocean; this meets the water vapor condition for more precipitation in northwest China and the eastern part of the Qinghai–Tibet Plateau, while there is less rain in the coastal area of South China. This is consistent with the conclusion obtained by He Jinhai et al. [10]. The other flow joins the Bay of Bengal air flow. The 850 hPa wind field in most parts of China is positively correlated with the latitude and longitude index, especially the central and eastern regions, and the northern regions pass the 90% confidence test. It shows that the more east and north the longitude and latitude indexes, the central and eastern regions have a stronger southwest air conveyor belt, so the northwest region has a stronger convergence of cold and warm air, which is more conducive to rainfall.

In years with high longitude and latitude indexes (Figure 7b), the southern Arabian Sea is controlled by a strong westerly airflow, which passes over the Indian Peninsula and the Bay of Bengal before arriving at South China and the South China Sea; a part of the flow continues eastward through the Philippines to join with the southeasterly airflow south of the West Pacific subtropical high-pressure system, and a part of this flows directly northward from South China and arrives at the central and eastern parts of China. A cyclone forms in west-central Mongolia, while the anticyclone ridge in the Pacific Ocean is at 40° N, with a strong intensity. In years with low latitude and longitude indexes, the northbound airflow over South China is markedly weaker; there is only one cyclone in central Mongolia, and the Pacific anticyclone air flow is also markedly weaker. The plot showing the difference in the years with high and low latitude and longitude indexes is very similar to the correlation coefficient plot.

3.5. The 500 hPa Height Field

Changes in the movement and intensity of various weather systems at 500 hPa, such as the western Pacific subtropical high, upper trough, and blocking high-pressure systems at the middle and high latitudes, have important effects on summer precipitation in China. As can be seen from the correlation coefficient chart (Figure 8a), there is a positive correlation in South China and a negative correlation in most of the other regions. Between 30° N and 50° N, the main positive correlation centers are located in Xinjiang, the Qinghai–Tibet Plateau, and Japan, and the correlation between the longitude index and the 500 hPa height of this region passed the 90% significance level. These results show that the geopotential height in the middle and high latitudes decreases when the longitude and latitude indexes are high, while the western Pacific subtropical high becomes stronger in the southern region, and water vapor Is transported northward from the northwest side of the subtropical high, which is conducive to precipitation in the northern region, while the southern region is arid and has a high temperature.

The ridge at 588 dagpm of the western Pacific subtropical high-pressure system is around 30° N, and the western ridge point does not exceed 120° E in the years with a high longitude index (Figure 8b) or 130° E in the years with a high latitude index. The Indian low-pressure system has only a large center of 584 dagpm in the years with a high longitude index, while no closed center formed in the years with a high latitude index. In the years with low longitude and latitude location indexes (figure omitted), the western Pacific subtropical high-pressure system is markedly weaker; the Yangtze River Basin represents a trough area of low pressure, the Indian low-pressure system has a smaller range, and the center is located more easterly. The map showing the difference (figure omitted) is similar to the correlation coefficient map, indicating that in the years with high longitude and latitude location indexes, the position of the western Pacific subtropical high-pressure ridge is located more to the west of the north than usual, and a summer monsoon mainly affects the area north of the Yangtze River, resulting in more precipitation there.

3.6. The 100 hPa Height Field

The timing and intensity of the South Asian high-pressure system ascending toward the Qinghai–Tibet Plateau, as well as its east–west oscillation and southward and northward movements [30,31,32], have an important impact on summer precipitation in China. The formation of the South Asian high-pressure system is closely related to Qinghai–Tibet Plateau heating [33,34]. The correlation coefficient (Figure 9a) diagram shows that the latitude and longitude indexes are negatively correlated with the geopotential height in Xinjiang, most of the Qinghai–Tibet Plateau, northwest China, northeast China, and Japan, but are positively correlated with central and southern China.

In years with high longitude (Figure 9b) and latitude indexes, the center of the South Asian high-pressure system is located in the Iranian Plateau, with a central intensity of 1686 dagpm, which extends more eastward to the central part of the Qinghai–Tibet Plateau compared with that during the low-value years. The feature line at 1680 dagpm ranges from 24° N to 40° N, and the eastern ridge point is near 110° E. The results show that the higher the latitude and longitude index, the more the South Asian high expands eastward and moves southward, resulting in increased summer precipitation in the north of the Yangtze River Basin and less precipitation in the south of the Yangtze River Basin. This is consistent with the work of Hu Jinggao et al. [35] (pp. 132–133), who pointed out that “the east ridge of the South Asia High pressure is eastward, the Jianghuai River Basin has positive precipitation anomalies, and the southeast coast and South China have negative precipitation anomalies”.

4. Conclusions

This study investigates the characteristics of the changes to the location of the Tibetan–Qinghai Plateau thermal low-pressure system and its impact on summer precipitation patterns in China. The key findings are summarized as follows:

Over the past six decades, the thermal low-pressure centers over the Qinghai–Tibet Plateau have predominantly been positioned near 90° E and 32.5° N within a geopotential height of 4360 gpm. Notably, the distribution area has exhibited more significant expansion from east to west compared to south to north.

The latitude and longitude indices both declined before the 21st century (southwest shift) and then began to rise (northeast shift). In the 1970s and early 2000s, the outliers were positive, but negative the rest of the time. Abrupt changes occurred in 1973 and 1981, transitioning from higher to lower values. The longitude index exhibits 6–8-year quasi-periodic oscillations, while the latitude index displayed 5-year and 8-year interannual oscillations before the 1990s, and gradually shifted to 12–14-year interdecadal oscillations.

The indexes exhibit inverse correlations with precipitation from the southeastern edge of the Qinghai–Tibet Plateau to the southern Yangtze River and northeast region. However, they are significantly positively correlated with precipitation from the central-eastern region of the plateau to the northwest, North China, and the Huang-Huai region.

The higher the index (the location of the center of the thermal low is east to north), the deeper the low pressure on the surface. Most of China to the east of the plateau is the conveyor belt of southwest airflow, while the northwest area is the confluence of northwest airflow and southwest airflow. The West Pacific subtropical high and South Asian high are stronger in the south of the Yangtze River Basin, so the rainfall in the south of the Yangtze River is significantly lower, while most of the north area experiences significantly more rainfall.

Wang Xin et al. [36] (pp. 67–68) found that “there are three main moving paths of the low vortex on the plateau: northeast, southeast and east, among which the number of low vortex moving to northeast is the largest”, therefore, we can speculate whether the movement of the plateau thermal low-pressure center to the northeast was more promoted it. What is the development mechanism between them? Using actual examples to demonstrate the corresponding views in the paper, we can ask how changes in thermal low-pressure systems inform climate prediction models or guide regional water resource management strategies? Further analyzing these physical mechanisms will be the focus of future efforts.