Rapid rise of water level for Tibetan lakes: an analysis of the relation with climate

The Tibetan Plateau (TP) has a large number of alpine lakes, which are sensitive indicators of climate variability due to minimal disturbances from human activities. Although earlier work has examined lake area and water level changes on the TP in the past several decades, so far, the climate-driven mechanism of lake variations is still not clear. In particular, it is uncertain which climatic factor (increased glacial meltwater caused by climate warming, or precipitation changes, etc.) induced the acceleration of lake growth since mid-1990s. This study examines water level changes of lakes during1990s~2011 by combining satellite Laser altimetry (covering small lakes due to finer footprints, but only during 2003~2009) and Radar altimetry (since 1990s, but only for a few large lakes due to coarse footprints). The precipitation and evaporation changes are also analyzed based on the GPCP data and station observations, which reveal that precipitation on the inner and northeast TP has experienced a significant increase of 2~8 mm/yr since mid-1990s and evaporation of most stations has showed an upward tendency. Two main findings of analyses on the relation of lake expansion and climate variability are summarized as follows: (1) The ICESat altimetry data during 2003~2009 shows that there is no significant difference between the change rates of water level of the 56 glacier-fed lakes and other 40 lakes without glacial meltwater supply, which implies that glacier melting induced by climate warming is probably not the dominating factor of rapid lake expansion. Six pairs of adjacent lakes with and without glacier supply (each pair is assumed under similar climate conditions) in different geographical regions (near the Nyainqêntanglha Mts., east Gangdise Mts., southeast Karakorum Mts., the Kunlun Mts., and the HolXil) were selected to further examine the impact of the glacier melting on lake expansions. Results show that some lakes without glacier supply even have higher growth rates than lakes with larger supply coefficients, which confirms that the rapid lake growth was more related with the precipitation increase rather than the glacial melting. (2) All 14 super-large lakes located in four different climate sub-zones showed a sharp water-level increase during 1995 ~ 2011 from the LEGOS (multiple radar altimetry data sets), but the timing of accelerated growth for lakes in different sub-zones is spatially heterogeneous. The abrupt change points of water level time series match very well with the years of more precipitation. For example, the water level of Lake Qinghai and Ngoring Co on the northeast TP declined in late-1990s and early-2000s, and had a sudden rise in 2004/2005 before keeping a slight increasing trend, which is in good agreement with the precipitation change in this region. The lakes in central Tibet, including Namco and Silingco, showed the earliest accelerated water level growth (since 1996/1997) due to more advanced increasing tendency of precipitation than other regions. In the contrast, temperature showed rapid rising trend in late 1980s and early 1990s which do not coincide with the timing of lake expansion. Thus, the glacier melting is probably not the primary factor of accelerated lake growth on the TP.