Abstract
Arid and semi-arid regions are the first to be affected by hydro-climatic changes. The Great Lakes Depression Basin in western Mongolia is the most notable example of such a region. Therefore, analyzing hydro-climatic changes in the Great Lakes Depression region is essential for future climate, hydrological, eco-hydrological processes, and ecosystem studies in similar areas and basins. In this study, Mann–Kendall (MK), innovative trend analysis method (ITAM), and Sen's slope estimator test (SSET) were used to determine the interrelationship between climate and river discharge changes and lake water level changes through statistical analysis. During the last 30 years, the air temperature has increased by 1.2 °C (Z = 1.16). Total annual precipitation decreased by 23.44 mm, resulting in 134.16 mm (Z = −0.79). The river discharge of the major rivers, such as Khovd River (Z = −3.51) and Zavkhan River (Z = −6.01), has significantly decreased. In Uvs (Z = 0.30) and Khyargas (Z = 2.03) lakes, the water level has also dropped. This study confirms that the increase in air temperature in the depression area of the Great Lakes reduces the amount of precipitation, and the decrease in precipitation affects the decrease in river discharge, which further affects the water level of the inflowing lakes.
HIGHLIGHTS
Arid and semi-arid regions are the first to be affected by hydro-climatic changes.
One of the regions most affected by climate change is the Great Lakes Depression Region of Mongolia.
Temperatures have increased in the Great Lakes Depression Region of Mongolia, and precipitation has decreased in most areas.
The river discharge in the Great Lakes Depression region of Mongolia has decreased, and the water level of the lakes has also decreased.
INTRODUCTION
The effects of global climate change are manifesting differently and have different impacts (Payne et al. 2020). Numerous studies have confirmed that global climate change is occurring rapidly under the influence of natural and human factors (Cavicchioli et al. 2019; Mikhaylov et al. 2020). As a result of the combined effect of these factors, global warming has become more intense in recent years (Lawrence et al. 2020). Studying changes in critical variables and understanding their relationships is essential to determine climate change. In particular, determining the parameter most susceptible to change and then modifying another parameter based on this parameter may be required for mitigating possible risks and spotting and monitoring any changes (Ford et al. 2019). These changes further affect ecosystems, socioeconomics, and human health (Desrousseaux et al. 2022; Roy et al. 2022). Therefore, climate change has different magnitudes, scopes, exposure sequences, and patterns (Yu et al. 2021).
Sensitive areas affected by climate change are classified by direct and indirect exposure and can be divided into local, national, regional, and global levels (Froese & Schilling 2019). In addition, depending on the region's characteristics, the amount of exposure may differ. The most sensitive areas affected by climate change are the arid and semi-arid regions of Central Asia (Luo et al. 2019, 2020). Numerous studies have shown that temperature changes in these regions are twice the global average (Dorjsuren et al. 2018; Yu et al. 2020). It further degrades the aquatic ecosystems in those regions and creates changes. Due to this, droughts, floods, desertification, frequency of natural disasters, degradation of farmers and agriculture, socio-economic decline, and poverty strongly affect these regions (Ren et al. 2022). Due to climate change, there can be considerable changes in the natural components. In arid and semi-arid regions, the first natural component to change may be alterations in the hydrological environment (Goulden et al. 2016). Therefore, climate change is closely related to changes in the hydrological system.
The surface waters of the Central Asian plateau are divided into several hydrological systems. This region encompasses various climate zones, resulting in a unique structure with distinct hydro-climate zones. Several studies on water climate change in Mongolia have revealed a significant increase in air temperature, a slight decrease in precipitation, and a reduced river discharge (Fernández-Giménez et al. 2017; Dorjsuren et al. 2022). The Mongolian part of Central Asia is divided into the Arctic Ocean Basin, the Pacific Ocean Basin, and the Central Asian Internal Drainage Basin (Chuluunbat et al. 2022). From these basins, the hydrological system of the Central Asian Internal Drainage Basin plays a unique role in the ecosystem of this region. Here 19.5% of Mongolia's total area is in the Arctic Ocean Basin, 11.5% in the Pacific Ocean Basin, and 69% in the Central Asian Internal Drainage Basin (Tugjamba 2021). Among the above basins, the driest, most sensitive, and the most subject to change is the Central Asian Internal Drainage Basin (Erdenee et al. 2021). However, it is vast and occupies a large area starting from the western part of Mongolia, passing through the central and southern parts of the country, and reaching the eastern end (Ganguli 2020). The most water-sensitive area of the Central Asian Internal Drainage Basin is the ‘Great Lakes Depression’ region in the western part of Mongolia, a unique ecosystem with diverse natural patterns (Alonso et al. 2019). This region has many different ecosystems, including high mountains, low mountains, steppes, dry steppes, large lakes, and river ecosystems (Лебедева et al. 2020). This region may be the most suitable area for monitoring the effects of hydro-climate change.
Several studies on changes in hydro-climatic variables in the Great Lakes Depression region of Mongolia have indicated a rapid warming trend and an increasing frequency of droughts over the past 50 years (Surenkhorloo et al. 2021; Mukhanova et al.). Water and climate change may occur strongly if these trends continue, and surface and groundwater sources may decrease. Therefore, it is crucial to precisely analyze and research how any impacts of climate change influence various types of ecosystems and natural settings confined in a small area (Schewe et al. 2019).
By studying the water climate change in the Great Lakes Depression, it is essential to understand the main parameters' interrelationship and the change process and correctly calculate their impact on the time and distance characteristics, ecosystem, and human life (Enkhbold et al. 2022). These parameters include air temperature, evaporation, precipitation, and river discharge, such as the amount of water discharge (Sumiya et al. 2020). Due to the increase in air temperature, the rate of evaporation increases. Increased evaporation causes changes in precipitation (Konapala et al. 2020). Changes in precipitation significantly impact the surface water supply and circulation of the hydrological system (Hao et al. 2019). Alteration of surface water is one of the most severe problems in arid and semi-arid regions (Ouhamdouch et al. 2019; Nakayama et al. 2021). To correctly solve the problem of water change, a hydro-climate study will significantly impact the regulation and management of water resources and natural processes related to water (Helmecke et al. 2020). Therefore, this study may be of great importance in understanding the impact of climate-dependent hydrological processes in Central Asia's arid and semi-arid regions.
This study aims to study the trend of hydro-climate change in the Great Lakes Depression region in western Mongolia. To accomplish the objectives of this study, the following goals were established: (i) to identify the trends of climate change at the chosen stations; (ii) to identify the spatial and temporal changes in river discharge; and (iii) to identify the correlation between the primary climatic and hydrological variables.
MATERIALS AND METHODS
Study area
Data sources
Several different data sets were collected to study climate and hydrological interrelationships. The data of air temperature, precipitation, and hydrologic data in the Great Lakes Depression region of Mongolia were taken from Information and Research Institute of Meteorology, Hydrology, and Environment (IRIMHE) hosts (http://irimhe.namem.gov.mn/) and were accessed to verify these data hydro-climatic data measured at meteorological station locations were downloaded and checked from NASA's power database (https://power.larc.nasa.gov/). Global SRTM 90 m resolution digital elevation model (DEM) data were obtained from the CGIAR-CSI GeoPortal (https://cmr.earthdata.nasa.gov/). Daily climatic and river flow data from the research were combined to create seasonal and annual data. Using data spanning 31 years, from 1990 to 2020, annual variations in the water climate were analyzed. The most readily available stations throughout the basin were chosen to gather and analyze the data for the research. These stations are typical of the basin. The location of the seven climate and water gauge stations used in this study is shown in Figure 1. For the selection of meteorological and water gauge stations in the Great Lakes Depression region of Mongolia, the following factors were taken into consideration: (1) spatial distribution of stations, (2) station capacity and data reliability, (3) availability of a water gauge station near the meteorological station, and (4) representation of large sub-basins (Dorjsuren et al. 2018; Liu et al. 2021).
Methods
This paper used the Mann–Kendall (MK) test method to detect the trends in climate and river discharge time series data. To evaluate the reliability of MK, the results were compared with the innovative trend analysis method (ITAM) and Sen's slope estimator test (SSET). In addition, annual and seasonal precipitation variability time series data were investigated by statistical analysis. Significance levels at 10, 5, and 1% were taken to assess the climate and river discharge time series data by MK, ITAM, and SSET methods.
MK trend test
The MK test method also shows upward and downward trends with statistical significance. The trend's strength depends on the data series's magnitude, sample size, and variations. The trends in the MK test are not significantly affected by the outliers in the data series since the MK test statistic depends on positive or negative signs (Dorjsuren et al. 2018; Gonfa et al. 2022). Annual data series are used for trend analysis in this study. The trends of yearly precipitation, air temperature, and river discharge have also been analyzed separately.
Finally, and are drawn as and curve. If there is an intersection between the two curves, the intersection is the beginning of the mutation.
Innovative trend analysis method
ITAM has been used in many studies to detect hydrometeorological observations, and its accuracy was compared with the results of the MK method (Gedefaw et al. 2018; Dorjsuren et al. 2022). The ITAM divides a time series into two equal parts and sorts both subseries in ascending order. Then after, the two halves are placed on a coordinate system on the X-axis and Y-axis. If the time series data on a scattered plot are collected on the 1:1 (45°) straight line, it indicates no trend. However, the trend increases when data points accumulate above the 1:1 straight line and decreasing trend when data points accumulate below the 1:1 straight line.
A positive value of indicates an increasing trend. However, a negative value of indicates a decreasing trend. However, when the scatter points closest around the 1:1 straight line, it implies the non-existence of a significant trend.
Sen's slope estimator test
The sign of shows whether the trend is increasing or decreasing.
Statistical analysis
Here, r is the correlation coefficient, x is the variable, is the standard deviation of x variable, y is the variable, and is the standard deviation of y variable.
RESULTS
Analysis of air temperature
The annual trend analysis of air temperature in all stations using the MK test, ITAM, and SSET results is presented in Table 1. The trend in the MK test shows an increasing trend in all stations. ITAM (ф) results show an increasing trend in air temperature everywhere except Uvs and Uliastai. It can also be seen that the results of SSET (β) have a growing trend at all stations. Therefore, it is possible to use the results of the MK trend analysis.
Station no. . | Name of station . | Z . | ϕ . | β . |
---|---|---|---|---|
1 | Ulgii | 1.81* | 2.63** | 0.02 |
2 | Khovd-Myangad | 0.76 | 0.92 | 0.01 |
3 | Uvs | 1.00* | −0.60 | 0.02 |
4 | Khyargas | 1.67* | 4.20*** | 0.03 |
5 | Durvuljin | 1.36* | 3.77*** | 0.03 |
6 | Uliastai | 2.63** | −3.18*** | 0.04 |
7 | Bayantes | 0.27 | 1.86* | 0.01 |
8 | Average | 1.16* | 3.24*** | 0.01 |
Station no. . | Name of station . | Z . | ϕ . | β . |
---|---|---|---|---|
1 | Ulgii | 1.81* | 2.63** | 0.02 |
2 | Khovd-Myangad | 0.76 | 0.92 | 0.01 |
3 | Uvs | 1.00* | −0.60 | 0.02 |
4 | Khyargas | 1.67* | 4.20*** | 0.03 |
5 | Durvuljin | 1.36* | 3.77*** | 0.03 |
6 | Uliastai | 2.63** | −3.18*** | 0.04 |
7 | Bayantes | 0.27 | 1.86* | 0.01 |
8 | Average | 1.16* | 3.24*** | 0.01 |
*Trends at 0.1 significance level.
**Trends at 0.05 significance level.
***Trends at 0.01 significance level.
Air temperature changes show an increasing trend in all stations of the study area, and the areas with the most significant changes (extreme climate processes) were in the lowlands and near lakes. This suggests that considerable ecological changes are expected to occur in big river valleys and lakes as a result of temperature changes. The most significant increase in air temperature occurred at the climate stations Uliastai (Z = 2.63) and Ulgii (Z = 1.18) which are close to the source of the Khovd and Zavkhan Rivers, the main sources of the hydrological network in this region, which originates from the high-mountain permafrost and glaciers. This can lead to the rapid melting of high-mountain snow and glaciers.
Analysis of precipitation
Table 2 displays the yearly trend analysis of annual precipitation at all stations using the MK test, ITAM, and SSET findings. The trend in the MK test shows a decreasing trend in all stations except the Bayantes station. ITAM (ϕ) results show a decreasing trend in precipitation everywhere except Durvuljin and Bayantes stations. It can also be seen that the results of SSET (β) have a decreasing trend at all stations.
Station no. . | Name of station . | Z . | ϕ . | β . |
---|---|---|---|---|
1 | Ulgii | −2.10** | −1.56* | −0.82 |
2 | Khovd-Myangad | −0.24 | −0.38 | −0.06 |
3 | Uvs | −1.06* | −2.82** | −0.57 |
4 | Khyargas | −4.52*** | −5.40*** | −2.44** |
5 | Durvuljin | −0.73 | 0.61 | −0.31 |
6 | Uliastai | −0.37 | −0.54 | 0.00 |
7 | Bayantes | 2.16** | 2.38** | 2.26** |
8 | Averagae | −0.79 | 0.07 | 0.59 |
Station no. . | Name of station . | Z . | ϕ . | β . |
---|---|---|---|---|
1 | Ulgii | −2.10** | −1.56* | −0.82 |
2 | Khovd-Myangad | −0.24 | −0.38 | −0.06 |
3 | Uvs | −1.06* | −2.82** | −0.57 |
4 | Khyargas | −4.52*** | −5.40*** | −2.44** |
5 | Durvuljin | −0.73 | 0.61 | −0.31 |
6 | Uliastai | −0.37 | −0.54 | 0.00 |
7 | Bayantes | 2.16** | 2.38** | 2.26** |
8 | Averagae | −0.79 | 0.07 | 0.59 |
*Trends at 0.1 significance level.
**Trends at 0.05 significance level.
***Trends at 0.01 significance level.
Analysis of river discharge and lake water level
The annual trend analysis of annual river discharge and lake water level in all stations using the MK test, ITAM, and SSET results are presented in Tables 3 and 4. The trend in the MK test shows a decreasing trend in all stations except the Bayantes station. ITAM (ϕ) results show a decreasing trend in river discharge and lake water level everywhere except Bayantes stations. It can also be seen that the results of SSET (β) have a decreasing trend at all stations. The MK, ITAM, and SSET trends all indicate a downward tendency for all stations.
Station no. . | Name of station . | Z . | ϕ . | β . |
---|---|---|---|---|
1 | Ulgii-Khovd River | −1.60* | −3.83*** | −0.60 |
2 | Myangad-Khovd River | −3.51*** | −2.66** | −1.36* |
3 | Durvuljin-Zavkhan River | −6.01*** | −9.07*** | −0.84 |
4 | Uliastai-Bogd River | −2.26** | −8.87*** | −0.13 |
5 | Bayantes-Tes River | 0.31 | 2.17** | 0.00 |
6 | Average conditions | −3.92*** | −3.91*** | −0.89 |
Station no. . | Name of station . | Z . | ϕ . | β . |
---|---|---|---|---|
1 | Ulgii-Khovd River | −1.60* | −3.83*** | −0.60 |
2 | Myangad-Khovd River | −3.51*** | −2.66** | −1.36* |
3 | Durvuljin-Zavkhan River | −6.01*** | −9.07*** | −0.84 |
4 | Uliastai-Bogd River | −2.26** | −8.87*** | −0.13 |
5 | Bayantes-Tes River | 0.31 | 2.17** | 0.00 |
6 | Average conditions | −3.92*** | −3.91*** | −0.89 |
*Trends at 0.1 significance level.
**Trends at 0.05 significance level.
***Trends at 0.01 significance level.
Station no. . | Name of station . | Z . | ϕ . | Β . |
---|---|---|---|---|
1 | Uvs Lake level | −0.30 | −3.02*** | −0.64 |
2 | Khyargas Lake level | −2.03** | −2.05** | −3.39*** |
3 | Average lake-level conditions | −2.51** | −2.00** | −2.14** |
Station no. . | Name of station . | Z . | ϕ . | Β . |
---|---|---|---|---|
1 | Uvs Lake level | −0.30 | −3.02*** | −0.64 |
2 | Khyargas Lake level | −2.03** | −2.05** | −3.39*** |
3 | Average lake-level conditions | −2.51** | −2.00** | −2.14** |
*Trends at 0.1 significance level.
**Trends at 0.05 significance level.
***Trends at 0.01 significance level.
Analysis of interrelationships of hydro-climatic variables
During the study period, the average air temperature of the basin increased by 1.2 °C from −0.5 to 0.7 °C (y = 0.023x − 0.4195) and (Z = 1.16). The most significant increase was observed in 2015 in Khovd station, where the average annual temperature was 3.48 °C. Regarding stations, the highest temperatures have been observed in Durvuljin, Zavkhan, Tes, and Khyargas Lakes, which are low elevations. Also, the maximum warming and the maximum cooling of the average air temperature are taking place in the areas mentioned earlier, which indicates that the possibility of extreme weather events is increasing in these regions.
Precipitation is relatively low, and the average annual precipitation for the basin is 134.16 mm. The total annual precipitation decreased by 23.44 mm, from 157.6 to 134.16 mm. During the study period, the year with the highest rainfall was 1990, with relatively high precipitation in mountainous regions. During the research period, the sum of precipitation showed a downward trend (y = −1.4698x + 157.68) and (Z = −0.79).
River discharge showed a decreasing trend at all stations, and the area average river discharge decreased from 39 to 24 m3/s (y = −0.5033x + 39.236) and (Z = −0.47). Fluctuations in river discharge were high. This indicates a significant factor influencing river discharge in the basin.
The water level of Khyargas Lake constantly decreased from 1995 to the end of 2020 (y = −10.262x + 735.56). It is conceivable that the hydroelectric power plant that was built alongside Khyargas Lake's main tributary, the Zavkhan River (in connection with the creation of the artificial lake ‘Gegeen Lake’), caused the lake's water level to decrease.
DISCUSSION
The decade trend of air temperature change in the study area increased between 1990 and 2000 and decreased between 2000 and 2010. However, it has increased in the last decade, from 2010 to 2020. However, everywhere in the study area, the precipitation decreased from 1990 to 2000. Between 2000 and 2010, precipitation decreased significantly in areas near large lakes and increased slightly in the remaining regions. But in the third decade, between 2010 and 2020, the amount of precipitation also decreased. The areas with the highest trend of air temperature increase are the highest elevations of the Great Lakes Basin and the sensitive areas that are the source of the Hovd and Zavkhan Rivers with permanent snow and glaciers. Changes in air temperature also differed in time, and natural zones differed from one another. January is the coldest month, with average air temperatures of –15 to –35 °C. Broken down by region, it is –30 to –34 °C in the valleys of the Altai and Khangai, –25 to –30 °C in the high mountainous area, –20 to –25 °C in the steppe, and –15 to –20 °C in the arid and semi-arid regions. July is the warmest month. The average air temperature in July is lower than 15 °C in the Altai and Khangai, 15 − 20 °C in the valleys of mountainous areas, and 20 − 25 °C in the arid and semi-arid regions. This difference in air temperature varies depending on the basin's natural zones, belts, altitude differences, migration of high-pressure areas, and environmental influences (Batima et al. 2005).
There was a slight downward trend in precipitation during the study period. But in the last 4 years, there has been a slight increase. Some scientists who have researched this region confirm that precipitation has decreased, for example, found a significant increasing trend in precipitation in the southern Altai from 1966 to 2015, while precipitation decreased in the northern Altai (Oyunmunkh et al. 2019). In general, this region's precipitation may continue to decline due to increased air temperature and evaporation. According to the research of Shi et al. (2019), it is believed that the precipitation around the Altai Mountains is mainly affected by the dryness of Central Asia and is expected to decrease in the future. Therefore, these results are consistent with the decreasing precipitation trend in the study area. Autumn and spring are fairly dry seasons. From November on, there is usually a steady snow cover into May. The summer months see around 70% of the annual precipitation. The amount of total precipitation is considerable in the high mountains, where the snow cover lasts for a long period. Conversely, the snow cover remains for a short while in the arid and semi-arid lowland areas, and there is minimal total precipitation (Yembuu 2021).
Permafrost melting plays a unique role in the water flow regime of this region because significant rivers, such as the Khovd River, Zavkhan River, Kharkhiraa, Turgen, and Buyant originate from the high-mountain's permanent snow (Tugjamba 2021). According to some researchers, the melting of the snow and glaciers into these rivers has decreased in recent years, so it is believed that the flow of the rivers will decrease. Water scarcity will probably increase across western Mongolia in the near future. From the 1980s to 2010, 63 lakes (>1 km2) disappeared, and about 683 rivers dried up, with many of these water bodies in the Altai's foothills (Pan et al. 2019). During the research period, the river discharge decreased significantly and sometimes increased and again decreased. Still, the decrease is substantially more significant at each station during the entire period, similar to the above results.
When measuring the water level of the two largest lakes in the study area, Uvs and Khyargas, the water level of both lakes showed a decreasing trend. Khyargas Lake's water level decreased most noticeably. This lake directly depends on the water supply from the Zavkhan River until it is fed. During this time, the ‘Gegeen Lake’ was constructed by storing water from the Zavkhan River. Since the hydroelectric power plant drew water from the artificial lake, the lake's water level decreased as a result (Enkhbold et al. 2021). Due to the smoothing accumulation of water in Gegeen Lake since 2016, the lake's water level has been slightly increasing. The precipitation has also increased somewhat during this period, which coincides with improving the protection of the lake water. As a result, it illustrates both the direct and indirect impacts of human activity on the basin's water and climate.
Considering the general trend of the hydro-climate during the research period, it was found that the increase in air temperature and decrease in precipitation mainly affected river discharge. Also, the reduction in the water level of large lakes such as Khyargas due to human activities may be related to the creation of Gegeen Lake and the changes in water regulation in the Zavkhan River (Erdenejargal et al. 2021). Therefore, hydro-climate parameters in this region are closely related.
CONCLUSIONS
In this study, the trend analysis of hydro-climatic variables in the Great Lakes Depression region in the western part of Mongolia was determined between 1990 and 2022. In the research, the interrelationships between climate and river discharge changes and lake water level changes were analyzed using Mann–Kendall (MK), ITAM, SSET, and statistic analysis.
Considering the effects of climate change in the basin of the Great Lakes Depression over the past 30 years, the average annual air temperature has increased by 1.2 °C (Z = 1.16), and the total yearly precipitation decreased from 157.6 to 134.16 mm (Z = −0.79). The most significant rivers in the depression of the Great Lakes, such as the Khovd River (Z = −3.51) and the Zavkhan River (Z = −6.01), have significantly decreased river discharge. In addition, the lake's water level has dropped dramatically in Uvs (Z = −0.30) and Khyargas (Z = −2.03) lakes.
This study confirms that an increase in air temperature in the Great Lakes Basin decreases the amount of precipitation, and a decrease in precipitation affects the decrease in river discharge, which further affects the water level of inflowing lakes. According to the analysis of hydro-climate changes in the basin, it is seen that the changes in the level of rivers and lakes are closely related to the factors of climatic. In the study area, as the air temperature increases, the precipitation decreases. While the temperature decreases, the precipitation increases. However, the trends have been different for three decades and the study region. Shortly, it is necessary to consider another factor affecting the Great Lakes Depression region of Mongolia.
AUTHOR CONTRIBUTIONS
B.D. and D.Y. conceptualized the whole article; B.D. developed the methodology; N.B. arranged the software; O.D., O.Y., and E.B. validated the data; B.G. rendered support in formal analysis; V.Z. and Y.Y. investigated the work; O.D., S.W., and X.L. arranged the resources; B.G. and E.B. rendered support in data curation; B.D. and A.E. wrote the original draft; B.D. and O.N. wrote the review and edited the article; O.Y. and X.L. visualized the data; V.Z. supervised the work; B.D. rendered support in project administration; B.D. and H.Z. supported in funding acquisition. All authors have read and agreed to the published version of the manuscript.
FUNDING
This research was funded by The Mongolian Science and Technology Foundation (grant CHN-2022/274) and also, was funded by The National Key Research and Development Project of China (grant 2022YFE0119400). Also, supported by the Ministerial Innovation Scholarship for Postdoctoral Research (grant: 19ХХ04DI208).
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.