ABSTRACT
The paper analyses the spatial and temporal characteristics of extreme precipitation in the upper reaches of the Yangtze River through extreme precipitation indicators based on the trend method, the Mann–Kendall trend test, and the rescaled extreme deviation extreme deviation using daily precipitation data from 1961 to 2021. The following conclusions were obtained: The overall precipitation in the upper reaches of the Yangtze River is reduced, and the number of rainy days is reduced. The frequency of extreme precipitation is generally reduced, but the spatial difference in the intensity of extreme precipitation is greater, which makes the occurrence of extreme precipitation more concentrated and more destructive. Extreme precipitation indicators showed relatively large fluctuations after 2000, especially in terms of extreme precipitation intensity. The frequency of extreme precipitation in the upper reaches of the Yangtze River is the highest in the main stream of the Yangtze River Basin and the Wujiang River Basin, the intensity of extreme precipitation is in the Jialing River and the Wujiang River Basin, and the accumulation of extreme precipitation is the highest in the Jialing River and the Wujiang River Basin, whereas the maximum value of the station extreme precipitation intensity and frequency is in the Minjiang River Basin.
HIGHLIGHTS
The overall extreme precipitation characteristics of the upper reaches of the Yangtze River under climate change were analysed.
Differences in extreme precipitation characteristics of the first-tier tributaries of the upper Yangtze River were analysed.
Changes in water resources in the upper reaches of the Yangtze River have a major impact on China's economy, food security, and energy resources.
INTRODUCTION
Global warming will lead to an intensification of the global hydrological cycle and an increase in global average precipitation and evaporation. At the same time, precipitation variability may change, which has a direct impact on precipitation, evaporation, runoff, and soil moisture, resulting in large amounts of precipitation over a short period of time, usually leading to natural disasters such as floods and mudslides. The Sixth Assessment Report of the IPCC (Intergovernmental Panel on Climate Change) shows that the frequency and intensity of extreme precipitation events have been increasing from the 1950s to the present (Diffenbaugh et al. 2017; Du et al. 2020; IPCC 2021). The study shows that the total precipitation of Bangladesh shows a significant decrease, but the number of days of heavy rainfall shows a significant increase (Imran et al. 2022). Thailand shows a significant increase in the intensity of extreme precipitation, but at the same time the number of days of precipitation also shows an enhanced trend (Samuels et al. 2017). Total annual precipitation declines in Australia, but the intensity of the most extreme precipitation events is enhanced over the moderate extremes (Osburn et al. 2021). The total amount of extreme precipitation in the UK shows an exponential increase. The different spatial and temporal characteristics of precipitation are highly correlated with regional climate, atmospheric circulation, water cycle, and energy balance. Therefore, different regional characteristics are exhibited in precipitation characteristics, especially extreme precipitation characteristics (Cotterill et al. 2021).
In the past few decades, the trend of total precipitation changes in China is not significant, and its changes in each region and seasonal changes are different (Zhai et al. 2005). From the average condition, the amount of precipitation mainly depends on the number of rainy days, and the annual precipitation and seasonal precipitation are consistent with this law (Gao et al. 2020).
The upper reaches of the Yangtze River are located at the eastern edge of the subtropical Eurasian continent. As the largest river in China, the Yangtze River, in which the Yangtze River Economic Belt is located, has an important strategic position in China's economic development (Guo et al. 2012). The upper reaches of the Yangtze River have a subtropical climate in East Asia (Zhang et al. 2007), with an average annual precipitation of 1,100 mm and summer floods and winter droughts (Chen et al. 2020). Influenced by the monsoon climate, the upper reaches of the Yangtze River are at the same time a vulnerable area to climate change, with frequent occurrence of droughts and floods and other extreme hydrometeorological events (Zhang et al. 2007). In recent years, under the combined influence of climate change and human activities such as hydropower development, the hydrological situation in the upper reaches of the Yangtze River has undergone significant changes, which are mainly manifested in a significant increase in average temperature and evapotranspiration (Zhang et al. 2007) and a significant decrease in precipitation and runoff (Liu et al. 2020). Between 1901 and 2020, mean annual temperatures show a significant upward trend, especially in the last 40 years, while regional variations show heterogeneity (Huang et al. 2012).
Extreme precipitation events are small probability events that exceed the climatic mean of precipitation values in a certain region within a certain time frame and reach or exceed a certain threshold that can be monitored and counted, a small probability event that can be monitored and counted that exceeds a certain threshold (Alexander et al. 2006). The IPCC states that global warming has led to an increase in the frequency and intensity of extreme precipitation events globally every year (Zhang et al. 2013). Atmospheric and ocean warming, glacier melting, sea level rise, and the greenhouse effect have increased significantly (Zhou 2021), exacerbating climate extremes in all regions of the globe (Zarekarizi et al. 2017). In recent decades, extreme precipitation events have occurred frequently, with increased intensity in Southwest China (Xu et al. 2021), further triggering geological disasters such as flash floods and mudslides, which have caused serious loss of life and property. The region has a large water fall and concentrates the major hydro energy resources of the Yangtze River Basin (Jialan et al. 2012). Analysing the changes of precipitation in the upper reaches of the Yangtze River, especially the characteristics of extreme precipitation, is of practical significance for serving the integrated operation of the Three Gorges Reservoir and mitigating the disasters caused by floods. Currently, there have been more studies on the characteristics of precipitation changes in the Yangtze River Basin (Wang et al. 2010; Zhang 2022), but most of the study areas focus on the whole Yangtze River Basin or focus on the middle and lower reaches of the Yangtze River, and the time period of the study is mostly in summer. However, the upper reaches of the Yangtze River are characterised by complex topography, anomalous persistence of large-scale atmospheric circulation (Wu et al. 2015), and large differences in regional climate characteristics.
Based on current scholars' research, the paper analyses the spatial and temporal characteristics of extreme precipitation in the upper reaches of the Yangtze River using daily precipitation data from 1961 to 2021, through the extreme precipitation indicator, based on the trend method, the Mann–Kendall (MK) trend test, and the rescaled extreme deviation (R/S) polarity, and analyses the design precipitation value of extreme precipitation using the regional linear distance method. Heavy rainfall and flooding disasters are frequent in the upper Yangtze River basin, and disaster prevention and mitigation are difficult in some mountainous basins due to the complexity of the terrain and the insufficient deployment of hydrological stations. Understanding the characteristics of extreme precipitation under climate change can provide data support for disaster prevention and mitigation.
MATERIALS AND METHODS
Study area
The Yangtze River is the upper reaches of the river from its source to Yichang in Hubei province, which is located in southwestern China, between 97.37–110.18°E and 21.13–34.33°N, with an area of 100 × 104 km2. The topography of the study area is high in the west and low in the east; the western part belongs to the Hengduan Mountains and the Tibetan Plateau, with an average altitude of 3,000–5,000 m, and the southeastern part is mostly below 500 m. The headwaters of the upper reaches of the Yangtze River come from the side slopes of the eastern edge of the Tibetan Plateau, where the river section has a large drop, deep canyons, and turbulent currents. The climate in the study area is complex, with the western plateau and mountains having a frigid climate with long winters and no summers, while most of the rest of the area is influenced by the southeast and southwest monsoon, with a humid subtropical monsoon climate, abundant rainfall, and heat.
Temporal trends in extreme precipitation, 1960–2020. (the blue fold is the basin yearly average of extreme precipitation information, and the orange one is the 5-year moving average as well as the 5-year moving standard deviation. Box plots show the 10-year standard deviation of extreme precipitation).
Temporal trends in extreme precipitation, 1960–2020. (the blue fold is the basin yearly average of extreme precipitation information, and the orange one is the 5-year moving average as well as the 5-year moving standard deviation. Box plots show the 10-year standard deviation of extreme precipitation).
Characteristics of the spatial distribution of extreme precipitation indicators.
DATA
Rain gauge data
The paper used daily precipitation data from 94 national stations on the upper Yangtze River for the period 1961–2020 from the China Meteorological Data Network (http://data.cma.cn).
Definition of extreme temperature and extreme precipitation indicators
In this paper, the Extreme Precipitation Index (http://etccdi.pacificclimate.org/) jointly developed by the World Meteorological Organisation (WMO) and the Research Programme on Climate Variability and Predictability (RPCVP) was selected, and its specific name and definition are shown in Table 1. These indices are widely used in the study of extreme precipitation (IPCC 2021) and can reflect the intensity, frequency, and duration of extreme precipitation events.
Definition of extreme precipitation indices
ETIs | |||
CWD | Consecutive wet days | Maximum number of consecutive days with RR ≥ 1 mm | Days |
R10 mm | Number of heavy precipitation days | Annual count of days when RR ≥ 10 mm | Days |
R25 mm | Number of very heavy precipitation days | Annual count of days when RR ≥ 25 mm | Days |
RX1DAY | Maximum 1-day precipitation | Annual maximum 1-day precipitation | Mm |
RX5DAY | Maximum 5-day precipitation | Annual maximum 5-day precipitation | Mm |
R95T | Very wet day precipitation | Annual total precipitation when RR ≥ 95th percentile of 1961–2016 daily precipitation | Mm |
R99T | Extremely wet day precipitation | Annual total precipitation when RR ≥ 99th percentile of 1961–2016 daily precipitation | Mm |
PRCPTOT | Annual total wet day precipitation | Annual total precipitation in wet days (RR ≥ 1 mm) | Mm |
SDII | Simple daily intensity index | Annual total precipitation divided by the number of wet days in the year | mm/day |
SPIs | |||
R95PT | Precipitation fraction due to very wet days | Ratio between R95T and PRCPTOT | % |
R99PT | Precipitation fraction due to extremely wet days | Ratio between R99T and PRCPTOT | % |
WD | Wet days | Annual count of days when RR ≥ 1 mm | Days |
ETIs | |||
CWD | Consecutive wet days | Maximum number of consecutive days with RR ≥ 1 mm | Days |
R10 mm | Number of heavy precipitation days | Annual count of days when RR ≥ 10 mm | Days |
R25 mm | Number of very heavy precipitation days | Annual count of days when RR ≥ 25 mm | Days |
RX1DAY | Maximum 1-day precipitation | Annual maximum 1-day precipitation | Mm |
RX5DAY | Maximum 5-day precipitation | Annual maximum 5-day precipitation | Mm |
R95T | Very wet day precipitation | Annual total precipitation when RR ≥ 95th percentile of 1961–2016 daily precipitation | Mm |
R99T | Extremely wet day precipitation | Annual total precipitation when RR ≥ 99th percentile of 1961–2016 daily precipitation | Mm |
PRCPTOT | Annual total wet day precipitation | Annual total precipitation in wet days (RR ≥ 1 mm) | Mm |
SDII | Simple daily intensity index | Annual total precipitation divided by the number of wet days in the year | mm/day |
SPIs | |||
R95PT | Precipitation fraction due to very wet days | Ratio between R95T and PRCPTOT | % |
R99PT | Precipitation fraction due to extremely wet days | Ratio between R99T and PRCPTOT | % |
WD | Wet days | Annual count of days when RR ≥ 1 mm | Days |
METHODS
The spatial and temporal distribution of rainfall is characterised by fractals and trends. The R/S method can reveal the fractal characteristics of the precipitation time series, indicating that the variability characteristics of the precipitation time series and the past and the future have the same or opposite change characteristics, and the MK method can well reveal the trend characteristics within a certain time series. In this paper, a combination of R/S and MK methods are used to analyse the characteristics of future trends in the time series of extreme precipitation indicators.
Linear trend analysis

MK trend analysis
The MK trend test is a non-parametric test originally proposed by Mann and Kendall, recommended by the WMO and continuously applied by many scholars (Mann 1945; Kendall 1970; Kumar & Jain 2010).
The MK trend test can be used to analyse the trend changes in the time series of precipitation, temperature, runoff, etc. The MK trend test is easy to calculate and is suitable for hydrological, meteorological, and other non-normally distributed data (Burn 2008), where the data are not disturbed by a small number of outliers that do not need to follow a certain distribution.
Rescaled polar deviation (R/S) analysis
Using the R/S analysis, the Hurst index and fractal dimension (D) were calculated and combined with the changes of each extreme precipitation index to determine the future trend of the extreme precipitation index, so as to derive the future fractal characteristics of extreme precipitation events in the study area (Wu et al. 2007). R/S analysis is a fractal structure analysis method for dealing with time series proposed by expert Hurst for the Nile region (Makarava et al. 2014). It is commonly used by scholars to analyse the fractal characteristics of time series. It has been shown that precipitation, temperature, tree annual rings, sunspots, and other results indicate a Hurst effect, and for such sequences, they can be analysed using R/S (Mason 2016; Zhou 2019).
Linear simulation of Equation (9) yields the Hurst exponent (0 < H < 1). For different Hurst indices, it means that the series have different trend changes. When H = 0.5, it shows that the sequence is completely independent, indicating that the sequence is a stochastic process. When 0 < H < 0.5, it means that the future state of change is opposite to the past and it is anti-persistence, and the smaller H means the stronger the anti-persistence. When H > 0.5, it means that the future state of change is consistent with the past, and a larger H indicates greater persistence.
RESULTS AND DISCUSSION
Temporal variability of extreme precipitation
From the analysis of the trend of extreme precipitation from 1960 to 2020, it can be seen that the total annual precipitation in the upper reaches of the Yangtze River shows a significant downward trend, with a downward trend of −1.44 mm/decade. The mean precipitation on precipitation days (SDII) showed a significant upward trend of 0.01 mm/decade. The number of precipitation days (WD) showed a decreasing trend of −0.16 days/decade. This suggests that the overall decrease in annual precipitation in the upper reaches of the Yangtze River is due to a decrease in the number of precipitation days and a higher concentration of precipitation, which may increase the frequency of droughts in the upper reaches of the Yangtze River.
R95T and R99T showed a significant downward trend of −3.55 and −3.36 mm/decade. RX1DAY and RX5DAY show a non-significant upward trend of 0.03 and 0.08 mm/decade, respectively. R10 mm showed a significant downward trend of −0.03 days/decade. R25 mm showed a non-significant downward trend of −0.01 days/decade. It suggests that extreme precipitation is more concentrated, more extreme, and more damaging, and, at the same time, may make droughts more frequent.
R/S analysis can determine whether the time series follow a random or biased random process. Therefore, the R/S analysis combined with the trend changes can determine the future abrupt changes and persistence of extreme precipitation in the upper Yangtze River. It can qualitatively analyse whether there are the same or opposite characteristics of past and future changes in the series. If H > 0.5 and D < 1.5, the sequence maintains long-term memory, future increments are similar to past increments, and the likelihood of continuing the current trend is strong; if H < 0.5 and D > 1.5, it is likely that the memory is weakening, and that the trend is ending and the reversal has begun; and if H = 0.5 and D = 1.5, the sequence is close to a random wandering without directional movement. With the exception of the number of consecutive precipitation days (CWD), H > 0.5 and D < 1.5 for the other indicators indicate a high likelihood that the existing trend in their characteristics will continue (Table 2).
Trends in extreme precipitation indicators, 1960–2020
Index . | WD . | PRCPTOT . | SDII . | CWD . | R99T . | R95T . | R25 mm . | R10 mm . | RX5DAY . | RX1DAY . |
---|---|---|---|---|---|---|---|---|---|---|
MK | −2.57 | −1.96 | 2.10 | −3.07 | −1.76 | −1.83 | −1.30 | −1.85 | 1.12 | 0.60 |
T | −0.16 | −1.44 | 0.01 | −0.02 | −3.36 | −3.55 | −0.01 | −0.03 | 0.08 | 0.03 |
H | 0.77 | 0.67 | 0.84 | 0.74 | 0.72 | 0.57 | 0.73 | 0.53 | 0.70 | |
D | 1.23 | 1.33 | 1.61 | 1.26 | 1.28 | 1.43 | 1.27 | 1.47 | 1.30 |
Index . | WD . | PRCPTOT . | SDII . | CWD . | R99T . | R95T . | R25 mm . | R10 mm . | RX5DAY . | RX1DAY . |
---|---|---|---|---|---|---|---|---|---|---|
MK | −2.57 | −1.96 | 2.10 | −3.07 | −1.76 | −1.83 | −1.30 | −1.85 | 1.12 | 0.60 |
T | −0.16 | −1.44 | 0.01 | −0.02 | −3.36 | −3.55 | −0.01 | −0.03 | 0.08 | 0.03 |
H | 0.77 | 0.67 | 0.84 | 0.74 | 0.72 | 0.57 | 0.73 | 0.53 | 0.70 | |
D | 1.23 | 1.33 | 1.61 | 1.26 | 1.28 | 1.43 | 1.27 | 1.47 | 1.30 |
The temporal trend of the extreme precipitation indicator shows that the change in the extreme precipitation indicator is relatively smooth before 2000, while it shows relatively large fluctuations after 2000: a downward trend from 2000 to 2010, peaking around 2010, an upward trend from 2010 to 2015, peaking around 2015, and then a downward trend.
From the box plot analysis (Figure 2), PRCPTOT changed most drastically in the 1990 and 2000s, SDII changed most drastically in the 2010s, R10 and R25 mm changed most drastically in the 2000s, R99T and R95T changed most drastically in the 2000s and 2010s, and RX1DAY changed most drastically in the 1990s and 2010s, whereas RX5DAY changed most drastically in the 2000s and 2010s and showed a more pronounced downward trend in the 2010s. It can be seen that the extreme precipitation indicators took a turn for the worse in the 2000s, and there were large fluctuations in RX1DAY, RX5DAY, R99T, and R95T in the 2010s, which made flooding frequent in the upper reaches of the Yangtze River during this time period.
Spatial distribution of extreme precipitation
Statistical results show that the average annual precipitation of the upper reaches of the Yangtze River from 1960 to 2020 is 763 mm (Figure 3). The average annual precipitation of the Jinsha River, MinTuo River, Jialing River, Wujiang River, and the mainstream area from 1960 to 2020 is 588.5, 844, 796.5, 915.3, and 873 mm, respectively. The overall layout of precipitation is larger than that in the northwestern part of the country. The Jinsha River receives the least amount of precipitation and the Wu River receives the most.
For the 94 meteorological stations in the upper reaches of the Yangtze River, the annual precipitation shows the distribution characteristics of large precipitation in the southeast and small precipitation in the northwest. The precipitation in the Jinsha River is significantly smaller than that in other sub-regions, ranging from 191.4 to 924.6 mm, and is spatially characterised by smaller precipitation in the narrow northwestern region and larger precipitation in the southern region. The precipitation varies greatly among the regions of the MinTuo River, ranging from 507.6 to 1,385.1 mm, with a clear distribution of small in the north and large in the south. The extreme precipitation point is located at Emeishan station, which may be related to the large difference in altitude in the MinTuo River basin. Precipitation in most areas of the Jialing River is evenly distributed, slightly greater in the east than in the west, and slightly greater in the south than in the north, with a range of 386–973.5 mm. Precipitation in the upper mainstream intervals of the Wu River and Yangtze River was generally high and more evenly distributed, with precipitation variations ranging from 741.7 to 1,103.6 mm and 790.7–1,023.9 mm, respectively.
The average RX1DAY was 48.4 mm in the Jinsha River basin, 80.4 mm in the Minjiang River basin, 91.8 mm in the Jialing River basin, 85.2 mm in the Wujiang River basin, and 83.5 mm in the main stream of the Yangtze River. The spatial distribution of RX1DAY in the Jinsha and Minjiang river basins is the most heterogeneous, with a minimum of 20.9 mm and a maximum of 79.8 mm in the Jinsha basin and a minimum of 31.8 mm and a maximum of 129.8 mm in the Minjiang basin. The spatial distribution of RX1DAY in the Jialing River Basin, Wujiang River Basin, and Yangtze River mainstream is more uniform, ranging from 70 to 110 mm. The average RX5DAY was 89.9 mm in the Jinsha River basin, 137.4 mm in the Minjiang River basin, 138.5 mm in the Jialing River basin, 136.4 mm in the Wujiang River basin, and 128.3 mm in the mainstem of the Yangtze River, with a large spatial difference in the distribution between the Jinsha River basin and the Minjiang River basin. The maximum value in the upper reaches of the Yangtze River is 219.8 mm at the Ya'an Station in the Minjiang River Basin, and the minimum value is 41.5 mm in the Jinsha River Basin.
R10 and R25 mm were shown to be the smallest on average in the Jinsha River basin and the largest in the Wu River basin and the Yangtze River main stream. The largest stations are located at the Emeishan Station, with 2,237 and 817 days, respectively. R99T and R95T showed that the Jinsha River basin average was the smallest, the Jialing River basin average had the largest R95T, and the largest R99T basin average was located in the Wu River basin. The CWD is the longest on average for the Jinsha and Minjiang river basins, at 8.2 and 8.3 days, respectively, with the Jialing, Wu and Yangtze river basins ranging between 6 and 7 days.
In summary, it can be seen that the extreme precipitation intensity in the upper reaches of the Yangtze River as a whole shows the distribution characteristics of large in the southeast and small in the northwest. The maximum extreme precipitation intensity is located in the Minjiang River Basin, while the Minjiang River Basin and Jinsha River Basin have the greatest differences in spatial distribution. The frequency of extreme precipitation is the highest in the mainstream of the Yangtze River Basin and the Wujiang River Basin, the smallest in the Jinsha River Basin, and the station with the highest frequency of extreme precipitation is located in the Minjiang River Basin. Extreme precipitation accumulations are largest in the Jialing and Wu River basins and smallest in the Jinsha River basin. The number of CWD is greater in the northwestern part of the upper Yangtze River than in the southeastern part, indicating more continuous precipitation in the northwestern part.
Characteristics of spatial variability of extreme precipitation
PRCPTOT showed a decreasing trend at 65% of the stations in the upper Yangtze River, with a maximum decreasing trend of −0.63 mm/decade, and 81% of the stations showed a decreasing trend in the number of days of precipitation, with a maximum decreasing trend of −0.65 days/decade. The SDII showed a comparable number of sites with upward and downward days, with upward trends averaging 0.20 mm/decade and downward trends of −0.17 mm/decade, respectively.
Characteristics of spatial variability of extreme precipitation indicators.
It can be seen that the annual total precipitation shows a decreasing trend, mainly due to a decrease in the number of precipitation days, whereas at the same time the number of CWD also decreases significantly. Extreme precipitation intensity in the upper reaches of the Yangtze River rises and falls at a comparable number of stations, but the rising trend is more pronounced. The frequency of extreme precipitation shows a decreasing trend in the upper reaches of the Yangtze River. In summary, overall precipitation in the upper reaches of the Yangtze River has decreased, with fewer rainy days, and at the same time, there is greater spatial variability in extreme precipitation.
DISCUSSION
(1) In terms of the overall characteristics of precipitation, the Yangtze River Basin has experienced a slight increase in precipitation and an overall decrease in precipitation frequency over the past 120 years, while the frequency of light to moderate rainfall has increased, with decreasing and increasing trends occurring during the periods 1901–1960 and 1961–2020, respectively (Zhang et al. 2023). The overall annual precipitation in the upper reaches of the Yangtze River shows a decreasing trend, which is more consistent with the findings of the paper. As for the middle and lower reaches of the Yangtze River, the study shows that the extreme daily precipitation is generally increasing during 1961–2012, but the intensity and frequency of extreme daily precipitation events have large spatial differences (Pei et al. 2017). The risk of extreme daily precipitation events in the middle and lower reaches of the Yangtze River is generally increasing, but there is a lack of uniform trends in different aspects such as frequency and intensity.
(2) In terms of the chronological characteristics of precipitation in the Yangtze River Basin, from 1961 to 2020, the extreme precipitation indices show a continuous upward trend, and most of the extreme precipitation indices have larger growth rates, which mainly occur in the middle and lower reaches of the Yangtze River. Compared with the extreme temperature, the trend of extreme precipitation is more ambiguous in the last 60 years (Yuan et al. 2021), and the change of extreme precipitation is simultaneously related to the influence of rapid urbanisation around the observatory (Zhang et al. 2023). The upper reaches of the Yangtze River experienced frequent extreme rainy events in the 1960s, the 1970s were dominated by extreme low rainfall in the whole basin, the centre of extreme rainy events shifted to the middle and lower reaches of the Yangtze River in the 1980s, the upper reaches of the Yangtze River experienced more extreme low rainfall events in the 1990s, and the Yangtze River basin is in the process of transforming from low rainfall to high rainfall since 2000 (Zhang et al. 2019).
(3) In terms of the temporal changes of extreme precipitation indices in the Yangtze River Basin, all precipitation indices in the middle and lower reaches of the Yangtze River show an increasing trend from 1979 to 2015, except for CDD and SDII (Guan et al. 2017; Yang et al. 2023). Among them, CWD and RX1day have an increasing trend, which is similar to the characteristics of the upper reaches of the Yangtze River studied in the paper. While most of the changes do not show significant changes, the intensity, frequency, and duration of extreme precipitation events have increased over time, which is more likely to lead to more extreme precipitation events over time. Most of the extreme precipitation indices changed abruptly during the 1980s and 1990s. CDD, R95p, R99p, RX1day, and SDII show statistically significant increasing trends. R10, R20, R25, PRCPTOT, and RX5DAY have statistically non-significant increasing trends. Extreme precipitation indices mostly have multi-scale periodic oscillations, but most of them tend to have high time–frequency scales of 2–3 years or less. This means that from 1970 to 2018, each index of extreme precipitation has undergone frequent ‘increases and decreases’. The major cycle for CDD was 4.9 years. The first major cycle for CWD, R95p, R99p, RX1 day, RX5 day, and SDII was 3.5 years. The first major cycle for R10, R20, R25, and PRCPTOT was 2.3 years, and the second major cycle was 3.5 years (Wu et al. 2021).
(4) The spatial variation of extreme precipitation indices in the Yangtze River Basin suggests that the flood risk in the region will increase (Wu et al. 2021), but the trend and variation of precipitation indices are extremely uneven in time and space, and it is difficult to detect the relationship between latitude and precipitation trends in the Yangtze River Basin (Guan et al. 2017). SDII, R95PT, R99PT, RX1DAY, RX5DAY, and CDD increased significantly throughout the basin, especially in the middle and lower reaches of the Yangtze River basin. In addition, PRCPTOT, R95PT, R99PT, and SDII also showed significant increases in the eastern part of the Tibetan Plateau. It is worth noting that in the west-central, south-central, and east-central part of the Yangtze River Basin, except for R20 mm in the central part of the Yangtze River Basin and CDD in the west-central part of the Yangtze River Basin, PRCPTOT, R10 mm, R20 mm, and CDD show a significant decreasing trend (Guan et al. 2017), and the opposite trend of extreme precipitation exacerbates the unevenness of precipitation distribution (Yang et al. 2023). For extreme river flow, Hankou and Datong stations in the middle and lower basins have significant increasing trends, while Cuntan and Yichang stations in the upper basin have opposite trends (Fang et al. 2018).
CONCLUSIONS
The characteristics of climate change in the upper reaches of the Yangtze River were analysed through the temporal and spatial changes of extreme precipitation indicators, and the following conclusions were obtained:
(1) Overall annual precipitation in the upper reaches of the Yangtze River has shown a decreasing trend, mainly due to a decrease in the number of days of precipitation, which has led to a greater concentration of precipitation, and at the same time exacerbated the risk of drought. Extreme precipitation in the upper reaches of the Yangtze River is characterised by a decrease in the total amount of extreme precipitation, an increase in the frequency of extreme precipitation, and an increase in the intensity of continuous extreme precipitation. Extreme precipitation is more damaging and drought frequency has increased, while changing characteristics are likely to continue in the future.
(2) Changes in extreme precipitation indicators in the upper Yangtze River were relatively smooth before 2000, but showed relatively large fluctuations after 2000, with a downward trend from 2000 to 2010, peaking around 2010, and an upward trend from 2010 to 2015, peaking around 2015, especially for the indicators of extreme precipitation intensity and extreme precipitation, which corresponded to the frequent occurrence of floods in the upper Yangtze River in this time period.
(3) The maximum values of extreme precipitation intensity and extreme precipitation frequency in the upper reaches of the Yangtze River are in the Minjiang River Basin, while the Minjiang River Basin and the Jinsha River Basin are the basins with the largest differences in spatial distribution in the upper reaches of the Yangtze River. From the average of the upper Yangtze River basin, the frequency of extreme precipitation is shown to be the largest in the mainstream of the Yangtze River basin and the Wujiang River basin, and the smallest in the Jinsha River basin. Extreme precipitation intensity was greatest in the Jialing and Wu River basins, and extreme precipitation accumulation was greatest in the Jialing and Wu River basins and least in the Jinsha River basin. The number of days of continuous precipitation was greater in the northwestern part of the basin than in the southeastern part, indicating more continuous precipitation in the northwestern part.
(4) The spatial variation of extreme precipitation indicators in the upper reaches of the Yangtze River is characterised by a decreasing trend in total annual precipitation, mainly due to a decrease in the number of precipitation days, while at the same time the number of CWD has also decreased significantly. Extreme precipitation intensity rises and falls at a comparable number of stations across the basin, but the rising trend is more pronounced. Precipitation frequency is predominantly decreasing in the upper reaches of the Yangtze River. Overall, the upper reaches of the Yangtze River have seen a decrease in overall precipitation and fewer rainy days, and at the same time, there has been a decrease in the frequency of extreme precipitation but a greater spatial variation in the intensity of extreme precipitation, which has resulted in a more concentrated occurrence of extreme precipitation and greater destructive power.
ACKNOWLEDGEMENTS
This research was financially supported by Chengdu University of Information Engineering Scientific Research Fund Grant Results (no. KYTZ202129) and Heavy Rain and Drought Flood Disasters in Plateau and Basin Key Laboratory of Sichuan Province (no. SZKT202201).
DATA AVAILABILITY STATEMENT
Data cannot be made publicly available; readers should contact the corresponding author for details.
CONFLICT OF INTEREST
The authors declare there is no conflict.