of LAC reached maximum values in September 2019 (81.44 × 103 km3), August 2020 (7.66 × <sup>10</sup><sup>3</sup> km3), and August 2021 (32.43 × 103 km3).

**Figure 9.** Monthly relative change of lake water volume and precipitation from 2019–2021, (**a**,**b**) lake water volume change and precipitation of BMC, (**c**,**d**) lake water volume change and precipitation of LAC, (**e**,**f**) lake water volume change and precipitation of LMC.

Abnormally high rainfall in the westerly region (LAC and LMC) in 2019 meant that the precipitation was significantly higher compared with other years. This high precipitation shows a good correlation with the obvious increase in lake water volume in 2019. In addition, there is also a good correlation between the peak of precipitation and the increase in lake water volume, indicating that precipitation is an important factor affecting lake change.

### *4.4. Response of Lake Change to Climate Change*

### 4.4.1. Interdecadal Variation of Meteorological Elements

From 1979 to 2018, the temperature showed an overall increasing trend with different rates: BMC-PC basin in the monsoon region is 0.68 ◦C/10a; LMC-SMXC basin in the westerly region is 0.24 ◦C/10a; LAC-MPYC basin in the westerly–monsoon interaction region is 0.064 ◦C/10a (Figure 10), the annual mean temperature of the three basins are −1.96, −4.26 and −8.48 ◦C, respectively. Among them, we can see that the BMC-PC basin showed a rapid warming trend, while the LAC-MPYC basin was relatively stable, and it is worth noting that the LMC-SMXC basin was stable before 2000, but it sharply decreased in 2000 then increased significantly.

The annual cumulative precipitation in different climate regions showed an overall increasing trend with a ratio of 36.31, 12.40, and 119.12 mm/10a, respectively, from 1979 to 2018. Among them, the cumulative precipitation of the BMC-PC basin was relatively stable before 2000 but fluctuated significantly after 2000, that of the LAC-MPYC basin increased significantly after 2000, and that of the LMC-SMXC basin was stable before 1998 and then increased rapidly. The increasing trend of the cumulative precipitation rate was mostly

bounded by the year 2000, and the cumulative precipitation increased significantly after 2000, which corresponded to the obvious increase in the lake area after 2000 (Figure 11).

**Figure 10.** Annual mean temperature of three basins in different climate region, the green lines mean one linear fitting equation, (**a**) the BMC-PC basin, (**b**) the LAC-MPYC basin, (**c**) the LMC-SMXC basin.

The annual mean wind speed from 1979 to 2018 showed different trends in different climate regions: the wind speed of the BMC-PC basin decreased slightly, with the rate of −0.16 m/(s·10a), and the wind speed fluctuated significantly from 1990 to 2000, then stabilized; the wind speed of LAC-MPYC basin decreased with the rate of −0.20 m/(s·10a), the wind speed increased in 2000 then decreased rapidly; the wind speed of LMC-SMXC basin decreased slightly before 1997 then increased rapidly with the rate of 0.39 m/(s·10a) (Figure 12).

The annual average specific humidity is generally stable from 1979 to 2018, but there are abnormally large or small around 2000: the specific humidity of the BMC-PC basin in the monsoon region and the LMC-SMXC basin increased significantly from 1995 to 2000 then decreased gradually while that of the LAC-MPYC basin showed a contrary tendency (Figure 13).

The variation trends of these meteorological elements changed in 2000, and the variation trends of the lake also changed in the same period. Therefore, the response mechanism of lakes to climate change will be analyzed in different periods, and the responses will be discussed over the entire time, before and after 2000.

**Figure 11.** Annual cumulative precipitation of three basins in different climate region, the green lines mean one linear fitting equation, (**a**) the BMC-PC basin, (**b**) the LAC-MPYC basin, (**c**) the LMC-SMXC basin.

**Figure 12.** *Cont*.

**Figure 12.** Annual mean wind speed of three basins in different climate region, the green lines mean one linear fitting equation, (**a**) the BMC-PC basin, (**b**) the LAC-MPYC basin, (**c**) the LMC-SMXC basin.

**Figure 13.** Annual Specific Humidity of three basins in different climate region, the green lines mean one linear fitting equation, (**a**) the BMC-PC basin, (**b**) the LAC-MPYC basin, (**c**) the LMC-SMXC basin.

4.4.2. Response of Lake Change to Climate

In this study, we analyzed the correlation between the lake area and the annual mean temperature, wind speed, specific humidity, and annual cumulative precipitation. Analysis results from 1979 to 2918 show that the main factors affecting the lake change are different from the change in climate zone. For the BMC and PC in the monsoon region, the correlation between temperature and lake area was the highest, with a correlation coefficient of 0.90 (*p* < 0.01), followed by specific humidity, with a correlation coefficient of −0.37 (*p* = 0.23), indicating a strong correlation; for the LMC and SMXC, the correlation between cumulative precipitation and lake area was the highest, with a correlation coefficient

of 0.87 (*p* < 0.01) and 0.88 (*p* < 0.01), followed by specific humidity, with a correlation coefficient of −0.36 (*p* = 0.24), indicating a strong correlation; for the LAC and MPYC, the correlation between wind speed, cumulative precipitation, and lake area were both high, with a correlation coefficient of 0.78 (*p* < 0.01) and −0.71 (*p* < 0.01), respectively. In addition, the dominant meteorological factors also changed in different periods. Before 2000, the dominant factors in the monsoon zone, the westerly zone, and the monsoon– westerly interaction zone were temperature, precipitation, and wind speed, while after 2000, which is the period of acceleration of warming on the TP, the dominant factors changed to temperature, precipitation, temperature and wind speed, specific humidity, respectively (Table 4). The wind speed and specific humidity (which can represent evaporation) are the dominant meteorological elements in the LAC-MPYC basin and show a negative correlation; this explains the contraction trend of lake area changes in the LAC and MPYC.


**Table 4.** Correlation analysis between lake area and meteorological factors.

In conclusion, although the dominant meteorological factors change with different climate regions, temperature and precipitation are always the main factors affecting the change of the lake. With the rapid warming of the TP, the influence of temperature on the expansion of the lake area is becoming more significant. The warming of the lake catchment area leads to the melting of glaciers and snow, which increases runoff and leads to the expansion of lakes. Moreover, with the warming and wetting of the TP, the precipitation gradually increases, which also promotes the expansion of lakes.

### **5. Discussion**

In recent years, the TP has become warmer and wetter. This trend involves a decreasing surface sensible heat and increasing latent heat from the northwest to the southeast, resulting in a significant increase in precipitation in the southeast and a decrease in the northwest [41,42]. In addition to being affected by precipitation, lakes in the TP are also closely related to glacial meltwater and permafrost degradation in the basin. Lake changes in the TP are significantly influenced by water and heat exchange between the lakes and the atmosphere, which in turn affects the regional water cycle. However, quantifying the dominant factors affecting lakes in different regions and how these factors will change under climate change conditions is key to accurately understanding the mechanistic role that lakes play in the water cycle of "Asian Water Tower".

Therefore, in each typical area of the TP, we should strengthen research on lake water balance and its response to climate change. This will involve collecting observations of the spatial and seasonal distribution of precipitation, runoff (including precipitation, meltwater, and underground runoff), and evaporation, and using methods such as a total water balance and isotope segmentation to study the response of lake water balance to changes in different supply sources. Such research would provide valuable information about the processes and mechanisms of how climate change will impact lakes in the future.

The changes in the area, water level, and water volume, as well as the monthly water volume variations of three pairwise lakes from different climate zones, were analyzed using remote sensing data and in situ observations. For the response of lake change to meteorological factors, we mainly focused on qualitative research but lacked quantitative research on the contribution of each meteorological element. In addition, there is still a part of the research that has not been carried out, such as the reason for lake water decline after 2020 due to lack of meteorological data. Therefore, quantifying the contribution of meteorological elements to lake change and using high-frequency meteorological data and eddy covariance data to analyze the reduction of lake level after 2020 will be the focus of the next stage of research. This will be helpful in accurately understanding the mechanism of climate change affecting lakes. Moreover, with the development of tourism in Tibet, the TP has been affected by more and more human activities; consideration of direct human impacts on the TP water supply remains poorly articulated but potentially important to the lake change research [43,44].

### **6. Conclusions**

There is a lack of research on the annual water volume changes and spatial differences of typical lakes in the TP due to the remote location and the lack of observational data. In the study, we have found there is obvious spatial heterogeneity in the seasonal changes of lake water volume in different climate regions. By using Sentine−2 remote sensing images, multisource altimetry, and observational water level data, the following conclusions are drawn.

(1) Inter-annual variations of lakes in different climatic zones are markedly different. From the 1970s to 2021, lakes in the monsoon (BMC and PC) and westerly (LMC and SMXC) regions show an overall expansion trend, while lakes in the westerly–monsoon interaction region (LAC and MPYC) showed an overall shrinking trend [45]. In the westerly–monsoon interaction region, the lake area shows a rapidly shrinking trend from the 1970s to 2005, after which the reduction slows or stabilizes [46]. The three groups of lakes have similar trends.

(2) Monthly variations of the lakes during the year in different climatic zones generally show similar trends. The changes are highly correlated with increases and decreases in monthly rainfall. This correlation is especially strong in 2019, which was a year of abnormal fluctuations in the westerly belt, with increased precipitation and significantly increased monthly changes in lake water volume. In addition, there is a good correlation between the peak of precipitation and lake water volume increase. These findings indicate that precipitation is a dominant factor affecting lake changes in the TP.

(3) The paper focuses on the effects of climate change on lakes from 1979 to 2018. The meteorological factors that dominate lake variation are temperature, precipitation, specific humidity, and wind speed (where specific humidity and wind speed can represent evaporation). Increases in temperature (which promotes melting of glaciers and snow) and precipitation promote the lake expansion, while increases in evaporation cause the lake shrinkage. For lakes in different climate regions, the main impact of meteorological elements is different, but with the accelerated warming on the TP, temperature plays an increasingly important role in accelerating lake expansion, while in the LAC-MPYC basin, evaporation is the leading factor that has caused the lake to shrink over the past decade.

**Author Contributions:** Conceptualization, W.M. (Weiyao Ma), L.B. and W.M. (Weiqiang Ma); methodology, W.M. (Weiyao Ma) and L.B.; validation, L.B., Z.X., W.H., R.S. and B.W.; formal analysis, W.M. (Weiyao Ma), W.H. and Z.X.; investigation, L.B.; resources, W.M. (Weiyao Ma), L.B., W.H. and R.S.; data curation, W.M. (Weiyao Ma) and L.B.; writing—original draft preparation, W.M. (Weiyao Ma) and L.B.; writing—review and editing, W.M. (Weiqiang Ma), Z.X., W.H., R.S., B.W., and Y.M.; visualization, W.M. (Weiyao Ma); supervision, W.M. (Weiqiang Ma); All authors have read and agreed to the published version of the manuscript.

**Funding:** This research has been funded by the National Natural Science Foundation of China (Grant No. 41830650); the Second Tibetan Plateau Scientific Expedition and Research Program (STEP) (Grant No. 2019QZKK0103) and the Strategic Priority Research Program of Chinese Academy of Sciences (Grant No. XDA20060101).

**Data Availability Statement:** The ICESat−2 data can be availability in the National Snow and ICE Data Center (https://nsidc.org/data/atl13/versions/5); The CryoSat−2 data can be availability in European Space Agency Earth Online (https://earth.esa.int/eogateway/missions/cryosat); The Sentinel 3B data can be availability in the Copernicus Open Access Hub (https://scihub.copernicus. eu/); The Jason−2 data can be availability in NOAA NCEI (https://www.ncei.noaa.gov/products/ jason-satellite-products).

**Acknowledgments:** The authors would like to acknowledge the National Aeronautics and Space Administration (NASA) and European Space Agency (ESA) for providing remote-sensing and multisource altimetry data, the Institute of Tibetan Plateau Research, Chinese Academy of Science for providing the lake water level and precipitation data from the Bamu Co, Langa Co and Longmu Co, and the students who went to Tibet to collect data.

**Conflicts of Interest:** The authors declare no conflict of interest.
