EN中文

Journal of Resources and Ecology >

2021 , Vol. 12 >Issue 5: 600 - 608

DOI: https://doi.org/10.5814/j.issn.1674-764x.2021.05.003

Human Activities and Ecological Security

Estimation of Critical Rainfall for Flood Disasters in the Qinghai-Tibet Plateau

  • MA Weidong , 1 ,
  • LIU Fenggui , 1, 2, * ,
  • ZHOU Qiang 1 ,
  • CHEN Qiong 1 ,
  • ZHANG Cungui 1 ,
  • LIU Fei 1 ,
  • LI Yanyan 1 ,
  • ZHAO Pei 3
Expand
  • 1. School of Geographic Science, Qinghai Normal University, Xining 810008, China
  • 2. Academy of Plateau Science and Sustainability, Xining 810008, China
  • 3. Dongguan Guangliang Holding Group Co., Ltd, Dongguan, Guangdong 523080, China
*LIU Fenggui, E-mail:

MA Weidong, E-mail:

Received date: 2020-12-16

  Accepted date: 2021-03-15

  Online published: 2021-11-22

Supported by

The Key Research and Development Projects of the Ministry of Science and Technology of China(2019YFA0606900)

The Second Qinghai-Tibet Plateau Scientific Expedition and Research Program(2019QZKK0906)

Fold

Abstract

According to the results of The Second Comprehensive Scientific Expedition on the Qinghai-Tibet Plateau, the balance of solid and liquid water on the Qinghai-Tibet Plateau is disturbed, and a large amount of solid water, such as glaciers and perpetual snow, is transformed into liquid water, which aggravates the risk of flood disasters in the Plateau. Based on the historical flood disaster records of the Qinghai-Tibet Plateau, this paper analyzed the temporal and spatial distribution characteristics of the flood disasters in the Plateau, and estimated the critical rainfall for the flood disasters combined with precipitation data from the meteorological stations in each basin of the Qinghai-Tibet Plateau. The results show that most of the flood disaster events in the Plateau are caused by precipitation, and the average annual occurrence of flood disasters is more than 30 cases and their frequency is on the rise. The high frequency areas of flood disasters in the Qinghai-Tibet Plateau are mainly in the Hehuang Valley and the Hengduan Mountains area; the secondary high frequency areas are located in the valley area of South Tibet and the peripheral area of the Hehuang valley. Finally, we found that the highest critical rainfall value of flood disasters in the Qinghai-Tibet Plateau is in the southern area of the plateau, followed by the eastern and southeastern parts of the plateau, and the lowest values are in the central, western and northern parts of the Plateau.

Key words： Qinghai-Tibet Plateau; flood disaster; critical rainfall

Cite this article

MA Weidong , LIU Fenggui , ZHOU Qiang , CHEN Qiong , ZHANG Cungui , LIU Fei , LI Yanyan , ZHAO Pei . Estimation of Critical Rainfall for Flood Disasters in the Qinghai-Tibet Plateau[J]. Journal of Resources and Ecology, 2021 , 12(5) : 600 -608 . DOI: 10.5814/j.issn.1674-764x.2021.05.003

1 Introduction

The increase in disaster risk caused by global climate change has become a major challenge affecting global security and development (Fang et al., 2014). According to the fifth assessment report of the Intergovernmental Panel on Climate Change, abnormal changes in extreme weather events around the world may lead to frequent meteorological disasters and increase the risks of various disasters in the context of global warming (IPCC, 2013). With the further aggravation of global climate change and the high concentration of social capital, the risk, vulnerability and physical exposure of the disaster causing body are changing rapidly (Shi, 2016). With the in-depth study of global climate change, more and more international institutions and scholars have begun to pay close attention to the impact of climate change on extreme hydrological processes (UNESCO, 2007, 2013).
Regional flood disaster prediction based on critical rainfall is an effective method to respond to flood disasters. At present, many useful explorations have been made in the research of critical rainfall for flood disasters, mainly focusing on the method for estimating the critical rainfall needed for flood disasters and regional empirical research. The most representative achievement is the early warning index series of the United States. This method calculates the early warning index based on the typical links of rainfall, runoff generation, confluence, evolution, and early warning index back stepping, and provides the dynamic change result (Norbiato et al., 2008). In addition, many scholars have estimated the critical rainfall of flood disasters in small watersheds or typical areas, which provides a reference and basis for the early warning and prediction of flood disasters (Duan, 2008; Fu et al., 2018; Li, 2018; Luo et al., 2018; Wang et al., 2018). Due to the abundance of hydrology, rainfall and historical flood disaster records, there are many different types of estimation methods for critical rainfall in a flood disaster event, such as statistical method of measured rainfall (OLGCFFPP, 2003; Chen and Yuan, 2005), inverse method of water level (Ye et al., 2008), critical curve method of rainstorms (Jiang and Shao, 2010), analogy method of disaster conditions (Chen, 2011), hydrological model method and effective rainfall method in the early stages (Liu et al., 2005; Ni et al., 2007; Liu et al., 2010; Song et al., 2012; Li and Guo, 2013; Huang and Chen, 2014; He et al., 2015).
The above empirical analysis and method discussion proves that flood disasters are predictable. Different calculation methods are suitable for various regions, where the geographical environment and climatic conditions are different. Based on abundant hydrological and historical disaster records, the occurrence of flood disasters can be predicted by using the appropriate hydrological model. Besides, the minimum range of the basin and rich hydrological data contribute to the high accuracy of the critical rainfall for a flood disaster event. However, most of the existing studies have only focused on the calculation of critical rainfall on a small-scale, and there is no reference for the study of critical rainfall estimation on a large-scale.
According to the results of The Second Comprehensive Scientific Expedition on Qinghai-Tibet Plateau, the solid- liquid balance there is disturbed, and a large amount of solid water, such as glaciers and perpetual snow, is transformed into liquid water. Under the background of climate change, the evaporation is increasing, which causes an increasing amount of water to enter the regional atmospheric circulation. With the increase of precipitation, extreme precipitation events are much more frequent, which increases the risk of flood disasters in the Qinghai-Tibet Plateau (Ma et al., 2020). Therefore, it is urgent and necessary to understand the characteristics of flood disaster events and their relationship to the precipitation in the Qinghai-Tibet Plateau. Thus, this paper selected the pre-effective rainfall method, which is suitable for the areas where the measured data are not abundant, to estimate the critical rainfall of the flood disasters on the Qinghai-Tibet Plateau, and provides a data reference for the early warning and prediction of regional flood disasters.

2 Survey of research areas

The Qinghai-Tibet Plateau is situated in the south-central part of the Eurasian continent, from the southern slope of the Himalayas in the south, to the north of the Kunlun-Qilian Mountains in the north, to the Pamir Plateau in the west and to the Hengduan Mountains in the east. The area of the Qinghai-Tibet Plateau in China is 2.57×106 km2, accounting for 26.80% of the total land area of China. Located at 26°00′12″-39°46′50″N, 73°18′52″-104°46′59″E, the east-west direction spans 31 longitudes with a length of about 2945 km and the north-south direction spans 13 latitudes with a width of about 1532 km (Zhang et al., 2002).
Fig. 1 Overview of the research area
The annual average temperature of the Qinghai-Tibet Plateau is between -5.75 ℃ and 2.57 ℃. The overall temperature of the plateau is lower than that of the eastern plain area of the same latitude, and the Qinghai-Tibet Plateau has a long winter or even a completely absent summer season. The precipitation of the Qinghai-Tibet Plateau mainly comes from the southwestern monsoon of the Indian Ocean. The dry and wet seasons are distinct, and the range of annual average precipitation varies greatly. In the arid areas such as the Qaidam Basin, the annual average precipitation is less than 100 mm. The southeastern part of the plateau has abundant precipitation, and in some areas the annual average precipitation exceeds 1700 mm. Although the precipitation varies greatly within the plateau, the annual average evaporation of the plateau maintains a consistently high level of 1204-1327 mm. Because of the high altitude, low cloudiness, low air density, and high reflectance of solar radiation from the underlying surface of the snow-covered area, the plateau is in a strong solar radiation environment, and the annual sunshine hours in most areas exceed 2600 hours. Under the control of the high-altitude westerly jet, gale weather occurs often all year round in the plateau. The average wind speed is 2-3 m s-1, and the local maximum wind speed is 34 m s-1 (Yao et al., 2000; Zhang et al., 2003).

3 Data sources and methods

3.1 Sources of data

The flood disaster data of the Qinghai-Tibet Plateau from 1961 to 2010 are derived from the disaster data monographs, including the China Meteorological Disaster Canon, Qinghai Natural Disasters and the climate bulletins of six provinces and autonomous regions (EBNDQP, 2003; Dong and Wen, 2005; Shi and Wen, 2006; Zhan and Wen, 2006; Liu and Wen, 2006; Wang and Wen, 2007; Liu and Wen, 2008). Daily precipitation data of meteorological stations are from China Meteorological Data Network (http://data.cma.cn/). Based on the principles of the longest period of daily precipitation data and the continuity of daily precipitation data, 78 meteorological stations within the Qinghai-Tibet Plateau were selected. The time range spans from January 1, 1961 to December 31, 2010. The data of all stations in the sequence passed the extreme value and consistency tests.

3.2 Estimation method of the critical rainfall for flood disasters

The critical rainfall for flood disasters refers to the possibility of a flood disaster in a region or basin after the 24-hour rainfall reaches a certain threshold. This threshold is called the critical rainfall for a flood disaster event.
Due to the limited monitoring equipment of water regimes and forecasting ability on the Qinghai-Tibet Plateau, the collection of water regime data in time to achieve timely early warning is not likely. For the Qinghai-Tibet Plateau, where water regime data are scarce, historical flood disaster investigation data and rainfall data for that time are used to analyze the contribution rate of early effective rainfall to the flood disaster. Then the critical rainfall for flood disasters can be determined by combining that data with the excitation rainfall on the day of the disaster. The equation for calculating effective rainfall is as follows:
P=kP1+k2P2++knPn
where, P represents the effective rainfall in the n days before the flood, Pn represents the rainfall of the n-th day, and kn represents a decreasing coefficient such that the closer to the day of the flood disaster, the higher the decreasing coefficient, with lower decreasing coefficients for times farther from the day of the flood disaster. The value of k is generally 0.8-0.9. In areas with sufficient precipitation, the value of k is usually about 0.8, and on the Qinghai-Tibet Plateau, the value of k is 0.9 (Pan et al., 2012; He et al., 2015).
According to the time of the flood disaster occurrence, the corresponding precipitation data of the nearest meteorological station were obtained. Then, the data for the effective rainfall of the five days prior to the day of the flood disaster were sorted and summarized to generate the statistical table of rainfall for regional historic flood disasters. Subsequently, formula (1) was used to calculate the effective rainfall of each flood disaster, and a discrete distribution map was obtained by combining it with the rainfall of the day when the flood disaster occurred. Finally, a polynomial formula was fitted to obtain the possible range of the critical rainfall value.
Based on the county scale, a total of 1506 records of flood disasters which occurred on the Qinghai-Tibet Plateau from 1961 to 2010 were collated. The effective flood disasters caused by precipitation were identified based on two conditions. One condition was the elimination of flood disaster events caused by non-precipitation factors, such as a dam break or melting water from snow and ice. Another was to exclude the flood disaster events for which no precipitation was recorded within five days prior to and the day of the disaster. The precipitation data of the station nearest to where the disaster occurred was used to approximately represent the precipitation that caused the flood disaster. Finally, the precipitation on the day of the flood disaster and the preceding five days were identified from the daily precipitation data of the station to calculate the effective precipitation in the early stage.
In the Qinghai-Tibet Plateau, long time series precipitation data are not available, and the distribution of precipitation observation stations is uneven. Therefore, for regions with frequent floods and abundant historical data, the critical rainfall of each station is calculated separately, based on the precipitation data of that station. So the critical rainfall of the region where the station is located is finally obtained. For regions with insufficient historical flood records, all individual flood events and related precipitation records in the region are summarized to calculate the critical rainfall value of the regional flood disaster as a whole.
3.2.1 The calculation of critical rainfall of flood disasters in cases with abundant data
To show the calculation of the critical rainfall value of a single station we will take Bomi as an example. Firstly, we identified the precipitation on the day of the flood disaster event and the preceding five days. Then, we generated the statistical table of historical flood disaster rainfall (Table 1).
Table 1 Statistical table of rainfall for historical flood disasters in Bomi (Unit: mm)
Serial number Date P5 P4 P3 P2 P1 Precipitation of the day
1 1983-07-30 3.1 0.6 0.9 1.0 7.0 19.6
2 1984-08-06 0 2.8 3.6 0 0 16.4
3 1985-05-31 1.2 3.6 0 4.5 8.6 32.6
4 1985-06-17 0 1.1 3.1 18.0 20.7 18.7
5 1992-06-24 4.2 2.9 4.3 23.4 23.0 43.1
6 1999-09-02 11.1 1.0 2.9 2.7 17.9 16.8
7 2000-04-10 12.5 6.2 2.8 8.4 2.7 25.0
8 2006-09-12 7.2 0.1 11.1 17.0 5.2 34.9
9 2010-05-11 0.2 0.3 0.2 3.5 2.8 39.3
10 2010-06-07 0.7 5.1 7.8 16.5 10.6 37.5

Note: Pn represents the rainfall of the n-th day.

After the statistical table of historical flood disaster rainfall was obtained, the early effective rainfall P was calculated by Formula (1), and the critical rainfall value of flood disasters in Bomi was finally fitted by combining the early effective rainfall and the same day rainfall. The fitting results for this example are shown in Fig. 2.
Fig. 2 Fitting results of critical rainfall for Bomi flood disasters
It can be seen that flood disasters may occur when the daily rainfall reaches 29.71 mm, even without any previous precipitation in the five days before the flood. The critical rainfall values of other stations are calculated by the same method, and the regional critical rainfall is finally obtained. For regions like the Yarlung Zangbo-Ganges River Basin, which has frequent flood disasters and relatively abundant records, the critical rainfall of each station in the region is calculated individually, and then the critical rainfall of the whole region is obtained.
3.2.2 The calculation of the critical rainfall of flood disasters for cases with limited data
Because the division of the region is based on the principle of the regional natural geographical environment integrity and similar precipitation, the regions such as the Qiangtang plateau internal flow area and the Qaidam-Tarim Basin internal flow area, yield flood disaster records with less site distribution density and less abundant data. The critical rainfall of flood disasters in these regions can be directly calculated by integrating the historical records of flood disasters in the whole region and matching it with the precipitation data for each disaster.
Taking the Qiangtang Plateau as an example, because the data of a single station could not meet the requirement for calculating critical rainfall, the critical rainfall of flood disasters can be calculated by integrating the flood historical records of the whole region. Firstly, the precipitation levels on the day of flood and the preceding five days were identified, and the statistical tables of regional historical flood rainfall were generated (Table 2). After obtaining the statistical tables of historical flood rainfall, Formula (1) was used to calculate the effective rainfall P in the previous period. The critical rainfall value was finally fitted by combining the effective rainfall in the earlier period and the rainfall on the same day as the flooding disaster. The fitting results are shown in Fig. 3.
There are six meteorological stations in the inner flow area of the Qiangtang Plateau, including Gerze, Xainza, Bangoin, Burang, Shiquanhe and Amdo. The average annual precipitation in the area is 248.04 mm. The historical flood disaster data of these six stations could not meet the requirements for calculating the critical rainfall of single station flood disasters. Therefore, the historical flood disaster records of the six stations are combined to calculate the critical rainfall. Finally, the critical rainfall value of the flood disasters in the flow area of the Qiangtang Plateau is 15.56 mm.
Table 2 Key parameters of rainfall for historical flood disasters in the inland flow area of the Qiangtang Plateau (Unit: mm)
Serial number County Date P5 P4 P3 P2 P1 Precipitation of the day
1 Burang 1990-08-08 0 0 3.1 0 0 15.3
2 Shiquanhe 1999-08-09 5.3 0 1.4 16.0 0 18.3
3 Gerze 1999-08-11 0 2.1 0.1 0 0 11.9
4 Shiquanhe 2000-07-28 7.0 0 0 0.5 0 16.1
5 Bangoin 2002-06-08 1.3 0 1.3 12.6 0.4 10.8
6 Xainza 2002-07-19 0 0 0.9 0 4.8 12.1
7 Amdo 2002-07-28 2.9 12.7 8.9 1.8 0.9 21.0
8 Xainza 2003-07-06 2.0 4.2 2.6 0.2 0.3 17.9
9 Xainza 2003-09-06 3.3 2.6 5.0 1.6 0 12.5
10 Burang 2006-07-31 4.7 1.8 1.5 0.2 0 12.4
11 Gerze 2010-08-12 0 0 0.3 3.9 16.9 17.5

Note: Pn represents the rainfall of the n-th day.

Fig. 3 Fitting results of critical rainfall for flood disasters in the inland flow area of the Qiangtang Plateau

3.3 Basin division of the Qinghai-Tibet Plateau

The Qinghai-Tibet Plateau has diverse natural landscape types, large differences in the nature of the underlying surface of the region and large altitude differences, all of which contribute to the spatial heterogeneity. The Qinghai-Tibet Plateau has many different physical geographical units that influence each other and are relatively independent. Since there are few monitoring stations for long-term precipitation data and they have uneven spatial distribution in the Qinghai-Tibet Plateau, the record richness of flood disaster events in each region in the historical period are variable. Hence, it is necessary to estimate the critical rainfall of flood disasters on the Qinghai-Tibet Plateau for the different regions.
Considering that the climatic characteristics of the basin are relatively uniform and the landform types are relatively close, based on the zoning results of Lawrence Crissman and Lex Berman, the Qinghai-Tibet Plateau is divided into eight basins (Zhang, 2019), as shown in Fig. 4.
Fig. 4 The eight divisions of the Qinghai-Tibet Plateau Basin

4 Results and analysis

4.1 Characteristic analysis of flood disasters

Based on the county scale, 1506 records of flood disasters in the Qinghai-Tibet Plateau from 1961 to 2010 were collected. The statistics of these flood events indicated that there were three kinds of inducing factors for the flood disasters on the Qinghai-Tibet Plateau. Most of the flood events were caused by precipitation, with only a small number of flood events caused by dams, rivers, glaciers and snow melting water. Among the 1506 floods, precipitation triggered 1490 of them, six were caused by the breaking of reservoirs and rivers, and ten were caused by melting snow from glaciers.
The average number of floods which occurred on the Qinghai-Tibet Plateau from 1961 to 2010 was 30.12 per year. The frequency of floods in 19 out of 50 years was above the average, while that in 31 years was below the average. Among the years, the largest number of flood events was recorded in 1998, with 108 records in total; the smallest number of flood events was only one recorded in 1968. In terms of annual variation, the frequency of floods increased by 10.10-fold per decade on average. From the perspective of the chronological changes of flood disaster frequency, the numbers of floods in 1961-1970, 1971-1980, 1981-1990, 1991-2000 and 2001-2010 were 83, 167, 298, 604 and 354, respectively. The decade at the end of the last century (1991-2000) was the decade with the highest frequency of floods. In the preceding 40 years of 1961-2000, the frequency of floods showed a trend of continuously increasing. In the 10 years after 2000, the frequency of floods decreased.
According to the temporal distribution of the floods, 81.20% of the floods occurred from June to August, 18.80% in March, April, May, September and October, and there were no flood disasters during November to February. Thus, the floods in the Qinghai-Tibet Plateau mainly occurred in summer, with relatively few in spring and autumn.
Based on the county scale, the frequencies of flood disasters in 213 counties of the Qinghai-Tibet Plateau during 1961-2010 were counted. Among them, the frequency of flood disasters was calculated as the average annual frequency of flood disasters in the counties. The areas with an average of annual flood disasters of 0.44 events or more on the Qinghai-Tibet Plateau were defined as the high frequency areas; the areas with 0.32-0.43 flood disasters per year were defined as the secondary high frequency areas; the areas with 0.21-0.31 flood disasters were defined as the medium frequency areas; the areas with 0.09-0.20 were defined as the sub-low frequency areas, and the areas with 0.08 events and below were defined as the low frequency areas.
Fig. 5 Interannual variation of flood disaster frequency in the Qinghai-Tibet Plateau from 1961 to 2010
From the spatial distribution, the high frequency areas were mainly in the eastern and southern parts of the plateau, such as the Hehuang Valley and Hengduan Mountains (Fig. 6). Among them, the flood frequencies of most counties in Hehuang Valley were above 0.44 events per year. The secondary high frequency area was located in the southern Tibetan Valley and the periphery of the Hehuang Valley. The middle frequency area was located in the Qaidam Basin, the northern of Kunlun Mountains, the southern valley of Tibet and the Hengduan Mountains. The north of Qilian Mountains and the periphery of the southern Tibetan Valley were the sub-low frequency areas of floods. In addition, the entirety of the Qiangtang Plateau, the southern Qinghai Plateau and most of the Hengduan Mountains were low frequency areas of floods.
Fig. 6 Spatial distribution of flood disaster frequencies in the Qinghai-Tibet Plateau

4.2 Estimation of critical rainfall of flood disasters in the Qinghai-Tibet Plateau

4.2.1 Estimation of critical rainfall of flood disasters in each area
Based on the estimation method of critical rainfall of flood disasters, the critical rainfall levels of eight basins on the Qinghai-Tibet Plateau were calculated. There were 14 meteorological stations in the Yellow River Basin, such as Xining, Menyuan and Guide, with an average annual precipitation of 489.62 mm. The flood disaster events of 11 of these stations with insufficient historical flood disaster data, such as Guide, Xinghai, and Guinan, were combined to calculate a single critical rainfall value. The other three stations had sufficient data to calculate their critical rainfall values individually, so ultimately four critical rainfall values were obtained. The critical rainfall values of each station are shown in Table 3.
Table 3 Critical rainfall values of flood disasters at stations in the Yellow River Basin (Unit: mm)
Serial number County Critical rainfall value
of flood disasters
1 Xining 36.12
2 Tongren 28.06
3 Qacuk 24.49
4 The other eleven stations 24.46
The average of the above four critical rainfall values (28.28 mm) was taken as the critical rainfall value of the whole Yellow River Basin.
There were four stations in the Hexi Corridor-Alashan inner flow area, including Wushaoling, Qilian, Yeniugou and Tole. The average annual precipitation of that area was 385.57mm. Because the data conditions of four stations could not meet the requirements of calculating critical rainfall, the historical records of flood disaster of these four stations were combined to calculate the critical rainfall. Finally, the critical rainfall of this area was 27.60 mm.
There were 14 stations in the Yarlung Zangbo-Ganges River Basin, including Lhasa, Shigatse, Bomi, Linzhi and Nyalam, with an average annual precipitation of 473.81 mm. The critical rainfall values were calculated by combining the flood disaster events corresponding to eight stations, such as Nyalam, Dingri, Parry and Dangxiong, which had insufficient historical flood disaster data. The other six stations with sufficient data had their own critical rainfall values calculated individually. Thus, seven critical rainfall values were finally obtained. The critical rainfall values of each station are shown in Table 4.
Table 4 Critical rainfall values of flood disasters at stations in the Yarlung Zangbo-Ganges River Basin (Unit: mm)
Serial number County Critical rainfall value
of flood disasters
1 Lazi 34.11
2 Lhasa 37.92
3 Shigatse 31.15
4 Jiangzi 36.54
5 Zedang 31.72
6 Longzi 26.09
7 The other eight stations 24.63
The average of the above seven critical rainfall values (31.74 mm) was taken as the critical rainfall value of the whole Yarlung Zangbo-Ganges River Basin.
There were four meteorological stations in Lancang River Basin, namely Zaduo, Nangqian, Changdu and Deqin. The average annual precipitation was 547.33 mm. The critical rainfall values were calculated by combining the flood disaster events corresponding to the three stations (Zaduo, Baoqian and Changdu) with insufficient historical flood disaster data. The critical rainfall value of Deqin was calculated separately, so two critical rainfall values were finally obtained. After calculation, the critical rainfall of flood disasters in Deqin was 33.01 mm, and the critical rainfall of other three stations was 26.71 mm. Therefore, the average value of 29.87 mm was taken as the critical rainfall of flood disasters in the internal flow area of the Lancang River Basin.
There were four meteorological stations in the Nujiang River Basin, namely Naqu, Suo, Dingqing and Gongshan. The annual precipitation of the other three stations was 559.74 mm, while that of Gongshan was much higher, at 1718.9 mm. One critical rainfall value was calculated by combining the flood events of Naqu, Suo and Dingqing stations, which were lacking in historical flood disaster data. The critical rainfall value for Gongshan was calculated separately, to yield two critical rainfall values. Based on the calculations, the critical rainfall of flood disasters in Gongshan was 44.14 mm, and the critical rainfall of other three stations was 32.89 mm. Therefore, the average value of 38.52 mm was taken as the critical rainfall of flood disasters in the Nujiang River Basin.
There were 22 meteorological stations in the Yangtze River Basin, including Wudaoliang, Qumalai, Yushu, Batang, etc., with an average annual precipitation of 618.21 mm. Although there were 22 stations in the Yangtze River Basin, the historical flood disaster data of all 22 stations could not meet the requirements of the calculation. So the critical rainfall was calculated by combining the historical flood disaster records of all 22 stations. Finally, the single critical rainfall value for the Yangtze River Basin was 28.12 mm.
There were 10 meteorological stations in the internal flow area of the Qaidam-Tarim Basin, such as Tashkurgan, Golmud and Dulan. The average annual precipitation was only 113.71 mm, which was the lowest on the Qinghai-Tibet Plateau. Only one station (Dulan) had enough data to calculate the critical rainfall of flood disasters in this area. Therefore, the critical rainfall of flood disasters in Dulan was calculated separately, while the data of the other nine stations were combined to calculate the critical rainfall. The final critical rainfall of flood disasters in this area was the average of these two values. Based on the calculations, the critical rainfall of flood disasters in Dulan was 22.16 mm, and the critical rainfall of flood disasters in the other nine stations was 22.67 mm. So the average value of these two was 22.42 mm, which was the critical rainfall value of flood disasters in the internal flow area of the Qaidam-Tarim Basin.
4.2.2 Estimation of critical rainfall of flood disasters on the Qinghai-Tibet Plateau
Based on the estimation method of critical rainfall of flood disasters, the critical rainfall values of the eight river basins on the Qinghai-Tibet Plateau were calculated, as shown in Fig. 7.
The map in Fig. 7 shows that the high value area of the critical rainfall value of flood disasters on the Qinghai-Tibet Plateau is the Nujiang River Basin, with a critical rainfall value of 38.52 mm. The sub-high critical rainfall areas were the Yarlung Zangbo-Ganges River Basin and the Lancang River Basin, with critical rainfall values of 31.74 mm and 29.87 mm, respectively. The median critical rainfall areas were the Hexi Corridor-Alashan inner flow area, the Yellow River Basin and the Yangtze River Basin, with critical rainfall values of 27.60 mm, 28.28 mm and 28.12 mm, respectively. The sub-low critical rainfall area was the inland flow area of the Qaidam-Tarim Basin, with a critical rainfall of 22.42 mm. Finally, the inland flow area of the Qiangtang Plateau was the low critical rainfall area of flood disaster, with a critical rainfall of 15.56 mm. Overall, the critical rainfall value of flood disasters in the southern part of the plateau was the highest, followed by the eastern and southeastern parts of the plateau, and the critical rainfall values in the central, western and northern parts of the plateau were the lowest.
Fig. 7 Spatial distribution of critical rainfall values of flood disasters in the Qinghai-Tibet Plateau

5 Discussion and conclusions

5.1 Discussion

In this paper, based on the basin division of the Qinghai-Tibet Plateau, the values of critical rainfall of flood disasters were calculated for different regions. Although there were enough meteorological stations in the frequently flooded areas to calculate the critical rainfall, the distribution of meteorological stations in some basins was less dense on the whole, which decreased the reference significance of the flood disaster critical rainfall comparisons between different regions. However, in the region of the Qinghai-Tibet Plateau, where there was a lack of stations, the measured data of runoff was also lacking. The historical records were not enough, but the estimation results based on the combination of historical flood disaster data and existing meteorological data of stations still provided very important references for coping with flood disasters.
The results show that the critical rainfall value of flood disasters is also higher in areas with abundant rainfall, and the spatial distribution pattern of values and the spatial pattern of precipitation both show a trend of decreasing from southeast to northwest. The proportion of extreme precipitation in the total precipitation in the Qinghai-Tibet Plateau has not changed significantly, and the frequency of extreme precipitation is increasing, which indicate that the impact of extreme precipitation events on the Qinghai-Tibet Plateau will be further intensified in the future (Ma et al., 2021). The critical rainfall value of flood disasters is estimated based on historical flood data and meteorological data, which contain the background information of the natural geographical environment and climate characteristics of the region. Therefore, it is of some significance to take the critical rainfall value of flood disasters as the reference for early warning information (Ma, 2019). In order to improve the accuracy of critical rainfall estimates, many procedures, such as the accuracy of the meteorological and hydrological monitoring data and the calculation model in line with the regional geographical environment, are needed to improve the accuracy in defining the critical rainfall for a flood disaster event.

5.2 Conclusions

A total of 1506 flood events were recorded on the Qinghai-Tibet Plateau from 1961 to 2010. The vast majority of these flood events were caused by precipitation. Only a small number of flood events were caused by dams, rivers, glaciers and snow melting water. From 1961 to 2010, there were 30.12 floods per year on average on the Qinghai-Tibet Plateau, and the frequency of floods increased by 10.10-fold per decade on average.
In terms of spatial distribution, the flood-prone areas on the Qinghai-Tibet Plateau are mainly in the Hehuang Valley and Hengduan Mountains. The sub-high incidence areas are located in the valley area of southern Tibet and the periphery area of the Hehuang Valley. The middle areas are located in the Qaidam Basin, the northern wing of Kunlun Mountains, the southern valley of Tibet and the Hengduan Mountains. The northern part of the Qilian Mountains and the periphery of the southern Tibetan valley are the sub-low-incidence areas. In addition, the Qiangtang Plateau, the southern Qinghai Plateau and most of the Hengduan Mountains are low-incidence areas.
According to the spatial distribution of flood critical rainfall on the Qinghai-Tibet Plateau, the flood critical rainfall in the southern part of the plateau is the highest, followed by the eastern and southeastern parts of the plateau, while the lowest values are in the central, western and northern parts of the plateau. The high value area of critical rainfall is the Nujiang River Basin, where the critical rainfall value for flood disasters is 38.52 mm. The sub-high critical rainfall areas are Yarlung Zangbo-Ganges River Basin and Lancang River Basin. The critical rainfall values of these two basins are 31.74 mm and 29.87 mm, respectively. The median critical rainfall areas are the Hexi Corridor-Alashan inner flow area, the Yellow River Basin and the Yangtze River Basin. The critical rainfall values of these basins are 27.60 mm, 28.28 mm and 28.12 mm, respectively. The sub-low critical rainfall area is the inland flow area of the Qaidam-Tarim basin, where the critical rainfall is 22.42 mm. The inward flow area of the Qiangtang Plateau is the low value area of critical rainfall, where the critical rainfall is 15.56 mm.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Chen G Y, Yuan Y M. 2005. Research on critical rainfall analysis and calculation methods of mountain flood. People’s Yangtze River, 36(12):40-43. (in Chinese)

[2]
Chen W H. 2011. Determination of early warning indicators in non-engineering measures for flash flood prevention and control—Taking Tangchang County, Longnan City, Gansu Province as an example. Guangdong Water Resources and Hydropower, 12(3):65-67. (in Chinese)

[3]
Dong A X, Wen K G. 2005. China meteorological disaster canon (Gansu volume). Beijing, China: Meteorology Press: 220-234. (in Chinese)

[4]
Duan S R. 2008. Calculation and analysis of critical rainfall in mountainous flood disasters in typical small watersheds: A case study of the Datong River Tributary in the Yellow River Basin. China Rural Water and Hydropower, 12(8):63-68. (in Chinese)

[5]
EBNDQP(Editorial Board of Natural Disasters in Qinghai Province). 2003. Natural disasters in Qinghai. Qinghai, China: Qinghai People’s Publishing House: 26-29. (in Chinese)

[6]
Fang J, Du J, Xu W, et al. 2014. Research progress on impact of climate change on flood disasters. Advanced Earth Sciences, 29(9):1085-1093. (in Chinese)

[7]
Fu C X, Huang J L, Zhang J L, et al. 2018. Analysis of flash flood warning indicators in typical rainy mountainous areas of south China. Guangdong Water Resources and Hydropower, 12(6):26-30. (in Chinese)

[8]
He N, Chen N S, Zeng M, et al. 2015. Study on critical rainfall of debris flow near field of dam site of Baihetan Hydropower Station. Journal of Hydraulic Engineering, 46(2):239-247. (in Chinese)

[9]
Huang G R, Chen Z L. 2014. Discussion on critical rainfall of mountain flood disaster based on early effective rainfall. Journal of Natural Disasters, 23(5):192-197. (in Chinese)

[10]
IPCC. 2013. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change The Physical Science Basis, Summary for Policymakers, Climate Change 2013.https://www.ipcc.ch/pdf/assessment-report/ar5/wg1/WGIAR5_SPM_brochure_en.pdf.

[11]
Jiang J H, Shao L P. 2010. Flash flood warning standards based on rainfall observations. Journal of Hydraulic Engineering, 41(4):458-463. (in Chinese)

[12]
Li C Z, Guo L. 2013, A review of critical rainfall determination methods. Flood and Drought Relief in China, 23(6):23-28. (in Chinese)

[13]
Li L. 2018. Research on critical rainfall of small flood basin based on GIS. Journal of Water Resources and Architectural Engineering, 16(3):248-252. (in Chinese)

[14]
Liu G X, Wen K G. 2008. China meteorological disaster canon (Tibet volume). Beijing, China: Meteorology Press: 108-150. (in Chinese)

[15]
Liu J H, Wen K G. 2006. China meteorological disaster canon (Yunnan volume). Beijing, China: Meteorology Press: 172-316. (in Chinese)

[16]
Liu Z Y, Yang D W, Hu J W. 2010. Flash flood warning method for small and medium rivers based on dynamic critical rainfall and its application. Journal of Beijing Normal University(Natural Science), 46(3):317-321. (in Chinese)

[17]
Liu Z, Martina M L V, Todini E. 2005. Flood forecasting using a fully distributed model: Application of the TOPKAPI model to the Upper Xixian Catchment. Hydrology and Earth System Sciences, 9(4):347-364.

[18]
Luo R H, Huang J L, Wang L H. 2018. Analysis of critical rainfall in sanya river basin. Guangdong Water Resources and Hydropower, 12(4):21-23. (in Chinese)

[19]
Ma W D, Liu F G, Zhou Q, et al. 2020. Characteristics of extreme precipitation over the Qinghai-Tibet Plateau from 1961 to 2017. Journal of Natural Resources, 35(12):3039-3050. (in Chinese)

[20]
Ma W D, Liu F G, Zhou Q, et al. 2021. Extreme precipitation dataset of Qinghai-Tibet Plateau (1961-2017). Journal of Global Change Data & Discovery, 5(1):1-7.

[21]
Ma W D. 2019. Characteristics of extreme precipitation in the Qinghai-Tibet Plateau and estimation of critical rainfall for flood disasters. Diss., Xining, China: Qinghai Normal University. (in Chinese)

[22]
Ni H Y, Li Z L, Ba R J, et al. 2007. Preliminary study on critical rainfall value of debris flow outbreak in Moxi River Basin on the east slope of Gongga. Journal of Mountain Science, 25(6):721-728. (in Chinese)

[23]
Norbiato D, Borga M, Degli E, et al. 2008. Flash flood warning based on rainfall thresholds and soil moisture conditions: An assessment for gauged and ungauged basins. Journal of Hydrology, 362(3-4):274-290.

[24]
OLGCFFPP(Office of the Leading Group for China’s Flash Flood Prevention and Control Plan). 2003. Critical rainfall analysis and calculation rules for flash flood. Beijing, China: China State Environmental Protection Administration. (in Chinese)

[25]
Pan H L, Ou G Q, Huang J C, et al. 2012. Study on precipitation rainfall threshold of debris flow in areas without data. Rock and Soil Mechanics, 33(7):2122-2126. (in Chinese)

[26]
Shi P J, Ying Z R. 2016. Impact of meteorological disasters on macroeconomic growth in China. Journal of Beijing Normal University (Natural Science), 52(6):747-753. (in Chinese)

[27]
Shi Y G, Wen K G. 2006. China meteorological disaster canon (Xinjiang volume). Beijing, China: Meteorology Press: 88-124. (in Chinese)

[28]
Song X, Chen T, Sun Y S. 2012. Study on calculation method of critical rainfall in small watershed. Shandong Water Conservancy, 12(9):47-49. (in Chinese)

[29]
UNESCO. 2007. Seventh phase of IHP planning. Paris, France: Meeting of the Intergovernmental Council of the International Hydrological Program.

[30]
UNESCO. 2013. Eighth phase of IHP planning. Paris, France: Meeting of the Intergovernmental Council of the International Hydrological Program.

[31]
Wang J, Zhao R F, Zhang S N, et al. 2018. Critical rainfall analysis of mountain flood disasters in Ji’an City. Water Resources Planning and Dsign, 12(9):56-59. (in Chinese)

[32]
Wang Z, Wen K G. 2007. China meteorological disaster canon (Qinghai volume). Beijing, China: Meteorology Press: 32-121. (in Chinese)

[33]
Yao T D, Liu X D, Wang N L, et al. 2000. Amplitude of climatic changes in Qinghai-Tibet Plateau. Chinese Science Bulletin, 45(13):1236-1243.

[34]
Ye Y, Wang Z Y, Fan B Q. 2008. Analysis of critical rainfall determination methods for mountain flood disasters in Zhejiang Province. Hydrology, 28(1):56-58. (in Chinese)

[35]
Zhan Z Y, Wen K G. 2006. China meteorological disaster canon (Sichuan volume). Beijing, China: Meteorology Press: 64-207. (in Chinese)

[36]
Zhang G Q. 2019. Dataset of river basins map over the TP (2016). Beijing, China: National Tibetan Plateau Data Center.

[37]
Zhang Y L, Li B, Zhen D. 2002. On the scope and area of the Qinghai-Tibet Plateau. Geographical Research, 21(1):1-8. (in Chinese)

[38]
Zhang Y S, Ohata T, Kadota T. 2003. Land-surface hydrological processes in the permafrost region of the eastern Tibetan Plateau. Journal of Hydrology, 283(1-4):41-56.

Options
Outlines

/

〈 〉