Abstract
The water bodies existing along highways in the high altitude areas of the Qinghai–Tibet Plateau (QTP) will aggravate subgrade settlement and road damage, and affect the long-term stability of road networks. Based on remote sensing and geographic information system (GIS) techniques, this study analyzed the changes in the number and area of water bodies along the G109 and G219 highways in the QTP in the past 20 years. The results showed that between 2000 and 2019, the number of water bodies along the two highways increased by 24 and 19%, respectively, and their area increased by 26 and 19%, respectively. The area and the number of water bodies >1 km2 in the permafrost area and those <1 km2 in the seasonal permafrost area both changed significantly. The change in the number of water bodies in the permafrost area was positively correlated with annual average temperature, while that in the seasonal permafrost area was significantly positively correlated with annual precipitation. This study provided basic data that could be used in studies on the interrelationship between engineering and water bodies within the context of climate change and will contribute to revealing the mechanisms through which engineering projects affect frozen soils.
HIGHLIGHTS
Variation of water bodies along the G219 and G109 highways in China was studied with the consideration of climate change.
Characteristics of variation in the number and area of water bodies along G219 and G109 were quantitatively revealed from 2000 to 2019.
Difference of correlation between the variation of water bodies and meteorology factors (air temperature and precipitation) in seasonally frozen soil and permafrost regions was examined.
Graphical Abstract
INTRODUCTION
As an important component of the global water cycle, the temporal and spatial variations of natural surface water bodies are essential for predicting and assessing regional food availability, energy production, human and ecosystem health, stability of terrestrial aquatic habitats, and social stability (Rodell et al. 2018). The Qinghai–Tibet Plateau (QTP) has the largest area, the highest average altitude, and the largest number of plateau lakes in the world, accounting for 49.5% of the total lake area in China, and is the source of many major rivers in China and neighboring countries. The characteristics of water bodies along highways in the QTP region are different from those of natural surfaces due to disturbances associated with engineering projects. The water bodies present on both sides of highways affect the hydrothermal balance of the road project and produce a certain degree of erosion, while water seepage causes frost heave, affecting the subgrade's stability (Chen et al. 2020). Therefore, studying the variation of water bodies along the highways using long-term series can reflect the interaction between engineering operations and the environment, which is of great practical significance for the evaluation and prediction of the stability of engineering projects.
The QTP has the highest and largest permafrost area of the middle and low latitudes in the world, accounting for 70.6% of the total in the high-Asia cryosphere. The hydrogeological characteristics of permafrost areas are significantly different from those of ordinary areas and are influenced by the weak permeable layers of permafrost, the weak transformation relationship between surface water and groundwater, and the extremely sensitive response of surface water to climate change (Mao et al. 2013). Global temperatures are increasing each year, and the response to the increase in high altitude regions such as the QTP is particularly sensitive. In the past 50 years, the glaciers on the QTP and surrounding regions have experienced an overall mass loss, with glacier melting and area decreasing by 20 and 18%, respectively (Liu et al. 2015). Surface water and groundwater have a continuing increase, and the number and area of thermokarst lakes increase annually as well (Niu et al. 2011). Therefore, the trend of decreasing ice and snow reserves and increasing liquid water reserves on the QTP not only will cause various water-related engineering problems to occur more frequently but will also affect the distribution characteristics and utilization patterns of water resources in the QTP region and the whole of China.
Based on remote sensing images, various studies have been conducted on single lakes, such as Nam Co (Zhu et al. 2010), Yamzho Yumco (Chu et al. 2012), Qinghai Lake (Liu et al. 2013), Mapam Yumco (La et al. 2012), and Selin Co (Duo et al. 2010). Overall, an increasing trend in the lake area has been detected. In Nam Co, the increasing rate was 2.37 km2/a from 1971 to 2004, and in Selin Co, it was as high as 41 km2/a from 1998 to 2008. In contrast, a decreasing trend was observed in the area of Mapam Yumco, and Yamzho Yumco also showed a decreasing rate of 8.59 km2/a from 2004 to 2010. The total area of lakes in the QTP experienced a rapid expansion at first and then a slow growth for 15 years after 2003. From 1960 to 2006, 30 new lakes >1 km2 formed on the QTP, and the original five lakes >1 km2 disappeared (Wan et al. 2014). At present, researchers are more concerned about the relationship between the changes in the lake area on the QTP as a whole (or in a certain area) and the climatic environment and pay less attention to the changes in the area and the number of water bodies related to specific road projects. In this paper, the Qinghai–Tibet Highway (G109) and the Xinzang Highway (G219) in the QTP were selected as the research object. The changes in water bodies located in the seasonally frozen ground and permafrost regions along the two highways were calculated. This study examined the variation in surface water bodies located along engineered roads, which is of great significance to identify the interrelationship between these water bodies and meteorological factors and determine the impact of engineering projects on them.
STUDY AREA
The G109 highway is the main channel into Tibet used for the circulation of goods in the region. It runs from Golmud in Qinghai Province in the north to Lhasa in the Tibet Autonomous Region in the south, over a total length of 1158 km. The G109's average elevation exceeds 4000 m, and the highest point is the Tanggula Pass, which can reach 5231 m. The length of the entire route through the permafrost zone is 632 km, of which 550 km is continuous permafrost and 82 km is discontinuous permafrost (Yin et al. 2014). The G219 highway starts from Yecheng County in Kashgar (Xinjiang Uygur Autonomous Region) in the north, connects with the G315, and continues south toward Lhaze County in Xigaze (Tibet Autonomous Region), where it connects with G318, for a total length of 2143 km. The average elevation of the whole road is more than 4500 m, and the highest point is in the Hongtu Daban in Ritu County, which can reach 5380 m. From north to south, the G219 crosses eight rivers, including the Yarkand River, Karakhash River, and Brahmaputra River. The highway is also the main passage from southwestern Xinjiang to northwestern and southwestern Tibet, as well as a national defense trunk road to the southwestern and northwestern borders of China (Yang et al. 2002).
DATA AND METHODS
Meteorological data
Meteorological data were derived from the temperature and precipitation grid data of the QTP and surrounding areas between 1998 and 2017, which were available from the National Tibetan Plateau Science Data Center (https://dx.doi.org/10.11888/Meteoro.tpdc.270239) (Ding 2019). From 2000 to 2019, the data of 12 meteorological stations, six located along the G109 (in Xidatan, Wudaoliang, Tuotuo River, Nagqu, Damxung, and Lhasa) and six on the G219 (in Kudi, Shahidulla, Tserang, Rutog, Shiquanhe, and Lhaze), were collected, and the annual average air temperature and annual precipitation data of each station were obtained.
Extraction of water body information
Relationship between water bodies and meteorological factors
The Mann–Kendall (M–K) analysis was used to examine the trend of changes in the number and area of water bodies along the G109 and G219 from 2000 to 2019. As a type of time series trend analysis, the M–K test is a nonparametric test method recommended by the World Meteorological Organization and has been widely employed. It can quantitatively calculate the variation of trends in time series, does not require samples to follow a certain distribution, and is immune to a few outliers. The test is suitable for data with a non-normal distribution, such as hydrological and meteorological data. It is easy to calculate and has been widely applied to time series datasets of parameters such as precipitation, temperature, runoff, and water quality (Hamed 2008). The correlation between the number and area of water bodies and meteorological factors was evaluated by Pearson's correlation coefficient, and the quantitative relationship was determined using the stepwise regression method.
RESULT
Trends in water body variations
Comparison of shapes of typical rivers in 2000 and 2019: (a) expansion of the Golmud River; (b) contraction of lakes in the Buqu River; (c) expansion of the Yixiantian reservoir; and (d) expansion of the Naijin River.
Comparison of shapes of typical rivers in 2000 and 2019: (a) expansion of the Golmud River; (b) contraction of lakes in the Buqu River; (c) expansion of the Yixiantian reservoir; and (d) expansion of the Naijin River.
The Yixiantian secondary reservoir is located 74 km from Golmud on the Nagin River, a tributary of the Golmud River, and the Yixiantian primary reservoir is located 4 km to the east. Based on Figure 3(c), both reservoirs experienced a significant increase in the river area, while the river flow between the further eastern region and Nachitai decreased, and the river area located 1 km away from Nachitai increased significantly. The area of the Naijin River, the east of Nachitai, increased except along the G109 (94°25′–94°22′E, 35°53′–35°52′N) and the rest of the section did not change significantly (Figure 3(d)).
Contraction of the river channel in 2000 and 2019: (a) Yarkand River; and (b) Qaraqash River.
Contraction of the river channel in 2000 and 2019: (a) Yarkand River; and (b) Qaraqash River.
Expansion of the river channel in 2000 and 2019: (a) Qaraqash River; (b) Longmu Co; (c) Angrengin Co; and (d) Lang Co.
Expansion of the river channel in 2000 and 2019: (a) Qaraqash River; (b) Longmu Co; (c) Angrengin Co; and (d) Lang Co.
Change characteristics of water body variations
Variation in the number and area of water bodies along the G109 from 2000 to 2019: (a) number of water bodies and (b) area of water bodies.
Variation in the number and area of water bodies along the G109 from 2000 to 2019: (a) number of water bodies and (b) area of water bodies.
Variation in the number and area of water bodies along the G219 from 2000 to 2019: (a) number of water bodies and (b) area of water bodies.
Variation in the number and area of water bodies along the G219 from 2000 to 2019: (a) number of water bodies and (b) area of water bodies.
Results of the M–K analysis along G109 (α = 0.05): (a) number and area of water bodies in the 0.1–1 km2 range and (b) the number of water bodies in 1–10 km2 range.
Results of the M–K analysis along G109 (α = 0.05): (a) number and area of water bodies in the 0.1–1 km2 range and (b) the number of water bodies in 1–10 km2 range.
Results of the M–K analysis along G219 (α = 0.05): (a) number of water bodies ≤0.001 km2 and in the 0.1–1 km2 range and (b) area of water bodies in 1–10 and the 10–100 km2 range.
Results of the M–K analysis along G219 (α = 0.05): (a) number of water bodies ≤0.001 km2 and in the 0.1–1 km2 range and (b) area of water bodies in 1–10 and the 10–100 km2 range.
Water body changes by frozen soil type
Distribution of permafrost on the Qinghai–Tibet Plateau (0 – seasonally frozen ground, 1 – permafrost, and 2 – non-frozen soil).
Distribution of permafrost on the Qinghai–Tibet Plateau (0 – seasonally frozen ground, 1 – permafrost, and 2 – non-frozen soil).
Variation in the number and area of water bodies along the G109 in the seasonally frozen ground region from 2000 to 2019.
Variation in the number and area of water bodies along the G109 in the seasonally frozen ground region from 2000 to 2019.
Variation in the number and area of water bodies along the G219 in the seasonally frozen region from 2000 to 2019.
Variation in the number and area of water bodies along the G219 in the seasonally frozen region from 2000 to 2019.
Variation in the number and area of water bodies along the G109 in the permafrost region from 2000 to 2019.
Variation in the number and area of water bodies along the G109 in the permafrost region from 2000 to 2019.
Variation in the number and area of water bodies along the G219 in the permafrost region from 2000 to 2019.
Variation in the number and area of water bodies along the G219 in the permafrost region from 2000 to 2019.
Relationship between water body, air temperature, and precipitation
Mean annual air temperature and annual precipitation of 12 meteorological stations along G109 and G219.
Mean annual air temperature and annual precipitation of 12 meteorological stations along G109 and G219.
Correlation between the number and area of water bodies and air temperature in permafrost regions.
Correlation between the number and area of water bodies and air temperature in permafrost regions.
Correlation between the number and area of water bodies and precipitation in seasonally frozen ground regions.
Correlation between the number and area of water bodies and precipitation in seasonally frozen ground regions.
DISCUSSION
The results show the trend in water body variation, increasing rate in the number and area of water bodies, and change characteristics in different soil types. The variation of water bodies along the G109 and G219 is different from that in other regions. The driving factors of water body variation mainly include precipitation, temperature, and engineering disturbance.
Influence of meteorological factors on water body changes
From 2000 to 2019, 1058 (24%) and 1219 (19%) additional water bodies were detected along the G109 and G219, respectively, and the water area increased by 10.85 km2 (26%) and 25.91 km2 (19%), respectively, which is consistent with the results reported in a previous study on the variation of lake number on the QTP in the same period (Zhang et al. 2021). It is believed that the increase in air temperature and precipitation, the increase in melted water of glacier and ground ice, and the degradation of permafrost are responsible for the increase in water bodies on the QTP (Lei et al. 2013; Kuang & Jiao 2016).
Since the 1960s, the QTP has experienced continuous warming and the weakening of monsoons, which, in turn, has weakened the external heat transfer of the plateau, aggravating its heating (Yang et al. 2011). While the reduction in potential evaporation and glacier mass loss may be one of the reasons for the lakes’ growth, a significant lake water excess is mainly due to the increased regional precipitation, as shown by the continuous rise in annual precipitation on the QTP, recorded at a rate of 8.06 mm/10a from 1961 to 2017 (Zhang et al. 2021). The continuous increase in air temperature over the past 20 years has accelerated the degradation of permafrost, leading to the collapse of the shores of large lakes and the formation of small ponds, ultimately resulting in an increase in the number of water bodies (Luo et al. 2015). In 2007, there were ∼69 thermokarst lakes and 234 water pits on both sides of the G109 connecting the Xidatan Basin to the Hoh Xil Hill region, over an area of ∼0.025 km2; the thermokarst lakes underwent a long-term slumping and collapse, and the water area increased to 0.35 km2 by 2009 (Niu et al. 2011). Other sources of water along the highways include seasonal surface runoff within permafrost regions (which are blocked by topography and sub-permafrost), rainfall, glaciers, and snowmelt, which allow surface water and groundwater to pool at the foot of subgrade slopes (Ding et al. 2020).
The results in Figure 15 show that the evolution of the water bodies located along the highways in the permafrost regions was mainly affected by air temperature. Most of the permafrost along the G109 is classified as high-temperature permafrost and it is obviously affected by changes in external temperature. Under the action of external forces, such as climate warming and human activities, the high temperature and ice-rich permafrost thawed, leading to the formation of thermokarst lakes. Due to the low reflectivity and high heat storage capacity of the ponded area, solar radiation will accelerate the temperature increase in it and the underlying permafrost. Therefore, once the thermokarst lake is formed, its thermal convection process will cause the underlying permafrost to continue to thaw and form a thaw zone (Fang et al. 2022), which performs an expansion of water body and the rise of water level on the ground surface.
Influence of engineering factors on water body changes
The overlapping and intersection of the highway network and the river system inevitably affect, or even change, the natural attributes of the water system in the region and the hydrological relationships within the basin. The construction of highways blocked the flow of surface water and groundwater to a certain extent and also led to changes in the hydrological characteristics of the underlying surface, consequently affecting the surface runoff and runoff paths. During the construction of highways, the engineering measures adopted to excavate mountains destroy the natural path of underground runoff and affect the groundwater level. Occasionally, during construction operations, groundwater is discharged and the groundwater level is lowered, resulting in the evaporation of surface water or its diversion. In addition, surface excavation coupled with heavy precipitation causes the groundwater level of the lower permeable layer to rise, replenishing the original dry surface with water (Medovar et al. 2018).
The Qinghai–Tibet railway (QTR) is generally built in parallel with the QTH and the average distance between the two lines is less than 10 km. The average distance from Golmud to Buqu is even less than 2 km. The continuous permafrost regions located along the Qinghai–Tibet Engineering Corridor (QTEC) mostly consist of permafrost with high temperature and high ice content, and the response to external disturbances is particularly severe. In addition, the water resistance of permafrost hindered the conversion of surface water and groundwater, thereby affecting the distribution characteristics of surface water. The construction and operation of projects in this area will inevitably lead to changes in permafrost and subsequent environmental consequences (De Grandpré et al. 2012; Chen et al. 2020). The construction of highways, railways, and oil pipelines has modified the original ground surface type and altered the heat exchange between the surface and the atmosphere, leading to an increase in the ground temperature of the underlying permafrost, thawing of the ice-rich permafrost, and ultimately, to the formation of thermokarst ponds and water pits in the plain or basin area. In addition, the pits left by construction operations are replenished by external water sources and, after the long-term collapse, they also develop into thermokarst ponds. These thermokarst ponds and water pits are widely distributed in the Chumarhe High Plateau, Wudaoliang Basin (a part of the Hoh Xil Hill region), Beiluhe Basin, and other areas (Lin et al. 2011). The results of this study showed that after the second phase of the QTR construction between 2001 and 2006, and the restoration and reconstruction of the QTH in 2008, the number and area of water bodies along the QTH have increased each year. This is consistent with the results reported in Niu et al. (2011), which indicated that, due to the disturbance caused by the reconstruction project along the QTH in 2008, the area of thermokarst ponds and water pits along the highway from the Xidatan Basin to the Beiluhe Basin increased by ∼0.33 km2. Therefore, the construction operations in the QTEC were partly responsible for the increase in the number and area of water bodies along the QTH.
CONCLUSION
In the present study, satellite images were used to extract data related to the number and area of surface water bodies located along the G109 and G219 highways within a range of 2 km, and their variation was examined from 2000 to 2019. In addition, this study analyzed the correlation between the number and area of water bodies, and air temperature and precipitation, and attempted to explain the reasons for the observed variations by considering both climatic and engineering factors. The following conclusions were drawn:
- (1)
From 2000 to 2019, the number of water bodies increased to 5517 (∼24%) with a total area of 52.76 km2 (∼26%) along the G109 and to 7659 (∼19%) with a total area of 161.56 km2 (∼19%) along the G219. The number of water bodies with small-area ranges accounted for 91% of the total, and water bodies with large-area ranges accounted for 83% of the total. Therefore, the increase in water bodies is mainly contributed by the increase in the number of water bodies <1 km2 and the area of water bodies >1 km2.
- (2)
From 2000 to 2019, the water area along the G109 and G219 increased significantly by ∼87 and ∼30%, respectively, in the seasonally frozen ground region. In the permafrost region, the water area along the G109 and G219 increased significantly by ∼25 and ∼7%, respectively. The variation of water bodies in the seasonally frozen ground region was considerably greater than that in the permafrost region, indicating that the two highways are significantly affected by the increase in temperature and humidity reported over the entire QTP.
- (3)
The observed changes in the number and area of water bodies in permafrost regions along the two highways were significantly positively correlated with air temperature whereas, in seasonally frozen ground regions, they were significantly positively correlated with precipitation. In addition, the increase of water bodies along the highways in the QTP was controlled by engineering projects, permafrost, and other factors. Both the seasonal permafrost region and the permafrost region show the most significant water body variation in the 0.1–1 km2 range.
- (4)
However, there are some limitations in this paper. The water body variation in our manuscript focuses on a width of 2 km on both sides along the roadway, which mainly reflects the engineering disturbance on water bodies. The trend of water bodies and their driving factors were preliminarily analyzed based on the obtained data of the past 20 years. In the next step, a longer time series of water body data should be collected, and the quantitative relationship between water body variation and driving factors will be obtained.
ACKNOWLEDGEMENTS
This work was supported by the National Natural Science Foundation of China (Grant no. 42001065); the Open Fund of the State Key Laboratory of Frozen Soil Engineering (Grant no. SKLFSE202106); and the First-Class Discipline Construction Project of Ningxia University, China (Grant no. NXYLXK2021A03).
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.