Abstract:
The inland lakes in the Tibetan Plateau (TP), with closed catchments and minimal human disturbance, are important indicators to climate change. However, examination of the spatiotemporal patterns of the Tibetan lake changes, especially for water level variation, was usually limited by inadequate measurement accessibility. This obstacle has been remedied by the developing satellite altimetry observations. The more recent studies revealed the growth tendency of lakes in the central TP had been decelerated or reversed during the period 2010-2016. It has not been systematically investigated whether the deceleration or hiatus would last in the following years thus far. This study aims to combine the traditional and recently-advanced altimetry observations to update our understanding of Tibetan lake changes in recent years. The results reveal that water level changes of the 22 examined lakes showed abrupt rises during the period 2016-2018, but the onsets and magnitudes of the rises varied among the lakes. During the study period, the water level of the nine lakes in the northern TP displayed a drastic rising trend with an average rate of 0.82 m/a. In the central TP, the lake level changes were generally divided into two categories. The water levels for the lakes in the western CTP rose rapidly, while in the eastern CTP, the lake water levels rose slowly with an average rising rate less than 0.40 m/a. The water levels for lakes in the northeastern TP and northwestern TP kept a stably rising tendency. According to the results of climate analysis, the spatial differences of the lake level rise rates were primarily caused by the spatial and temporal changes of precipitation over the TP, which may be related to the large-scale atmospheric circulation affected by the El Niño and La Niña events.
Key words : Tibetan Plateau, lake, water level, satellite altimetry, climate change, precipitation

1 Introduction

Lakes are an important part of global hydrological and biogeochemical cycles, as well as an essential resource for human societies (Lehner and Döll 2004; Palmer et al. 2015; Sheng et al. 2016). These inland waterbodies are quite sensitive to climate change and human activities. The last several decades have witnessed the widespread and tremendous changes in lakes in the context of global changes at regional and global scales (Shi et al. 2014; Smith et al. 2005; Song et al. 2013; Vörösmarty 2002; Wang et al. 2014). Lake changes have profound effects on regional water balance, ecosystems, biogeochemical cycles, exchange of energy and tracing gases with the atmosphere, and human water consumption. Therefore, investigation of lake variability is critical for a wide range of socioeconomic, political, and scientific interests. Several recent efforts such as Pekel et al. (2016), Donchyts et al. (2016), Yao et al. (2019), and Busker et al. (2019) elucidate lake dynamics at global scale using multi-temporal satellite optical and altimetry observations. The prominent characteristics revealed in these studies suggest that many large lakes, such as Caspian Sea, Aral Sea, Urmia Lake in endorheic basins of Eurasian hinterland, shrank remarkably in the near decades(Wang et al. 2018). On the contrast, widespread lake growths are observed on Earth’s highest plateau, the Tibetan Plateau (TP).
The TP, known as “Asian Water Tower”, accommodates vast distributions of glaciers and lakes. There are more than 1200 lakes larger than 1 km2, covering an inundation area of ~43,000 km2 (recorded around 2000) (Zhang et al. 2014). The lack of historical in situ data in the remote and sparsely populated TP has greatly limited the investigation of the hydrological and climate responses of lakes to the changing climate. However, this obstacle has been reduced by the rapidly emerging satellite observations from various sensors. Many prior studies employed optical satellite images to examine lake changes across space and time in the TP (Huang et al. 2011; Lei et al. 2014; Lu et al. 2017; Mao et al. 2018; Qiao et al. 2019; Song et al. 2014a; Song et al. 2014b; Wan et al. 2016; Zhang et al. 2019; Zhu et al. 2010). However, the acquisition of high-quality optical images is usually impeded by unfavorable weather conditions, and the derived lake area variations cannot directly indicate the water budget changes. Therefore, more and more satellite altimetry measurements have been synergized for the investigation of the lake dynamics across the TP.
Phan et al. (2012), Song et al. (2014b), Wang et al. (2013), and Zhang et al. (2011) have examined the changes in the lake water levels in the TP and found that most Tibetan lakes experienced rapid water level rises during 2003–2009 by using the Ice, Cloud and land Elevation Satellite (ICESat) laser altimetry data. However, the long revisit cycle and sparse sampling of the ICESat mission is insufficient for tracing long-term or abrupt changes (e.g. sudden shift) in water level, resulting in an incomplete misunderstanding of the trends and heterogeneous characteristics of lake changes in the TP. Favorably, the height measurements of satellite radar altimeters that have been launched continuously since the early 1990s can extend lake level observations to a longer temporal coverage. The advantages are applications that are not affected by weather and a reasonable high frequency of revisit from monthly to ten-day time scales, so that we can better solve the spatial and temporal patterns of water level changes and driving mechanisms in the TP. The combination of multi-radar altimetry missions, including TOPEX/POSEIDON, ERS-1/-2, Envisat, Cryosat-2 and Jason-1/-2/-3, at the height accuracy ranging from decimeters to centimeters, have been used to investigate the decadal water level changes of lakes in the TP (Gao et al. 2013; Kleinherenbrink et al. 2015; Song et al. 2015a). However, these radar altimeters are only suitable for large water bodies and because of their large-size footprint and along-track/cross-track spacing, many lakes are only accessible by one or two satellites.
The recent advances in satellite altimeters, e.g. CryoSat-2 (2010−), SARAL/AltiKa (2013−), Jason-3 (2016−) and Sentinel-3 (2016−), provide alternative options of measuring relatively smaller water bodies. The more recent studies benefiting from the longer and finer observations revealed the growth tendency of lakes in the central TP had been decelerated or reversed during the period 2010-2016 (Hwang et al. 2019; Jiang et al. 2017; Lei et al. 2019; Song et al. 2015b). However, whether the deceleration or hiatus would last in the following years remains unclear. It has not been systematically investigated thus far. Thus, this study aims to combine the traditional and recently-advanced radar altimetry measurements to update our understanding of the changing characteristics of Tibetan lakes in the near past several years. Furthermore, we will explore the potential climate driving mechanism of the recent varying tendency of lake changes in the TP.

2 Materials and methods

2.1 Study region

The TP is located in central Eurasia with a geographic area about 3 million km² (Song et al. 2014b). It has a mean altitude of approximately 4,000 m. The climate in the TP is marked by low temperature and strong solar radiation (Piao et al. 2011). The precipitation in the TP is characterized by strong seasonality. Most of the precipitation (60% -90% of annual total precipitation) occurs between June and September, while the rainfall from November to February is less than 10%. (Xu et al. 2008). The TP and its surroundings are sometimes known as “Earth’s Third Pole” and “Water Tower of Asia” because of the the rich storage of water resources. The geographical environment of this area is complex, and the water system is developed. It is the birthplace of many large rivers (Yangtze River, Yellow River, Mekong River, etc.) in China and throughout Asia. In addition, there are many lakes in the TP, which is one of the most densely distributed regions in China, and these lakes are currently undergoing significant changes due to the impact of climate changes (Ma et al. 2010). In this study, a total of 22 lakes larger than 100 km² in the TP were analyzed by combining the traditional and recently-advanced radar altimetry measurements. The characteristics of the studied 22 TP lakes covered by satellite altimetry are presented in Fig. 1 and Table 2.