Search

Abstract Lake is an important water resources in Mongolia, which has undergone a large variation in past decades. However, it is still challenging to monitor long-term changes in lake water storage (LWS) due to the lack of lake level monitoring and long-term satellite altimetry data for Mongolian lakes. According to the Advanced Land Observing Satellite Digital Elevation Model (ALOS DEM) and the Joint Research Centre (JRC) dataset, we estimated the LWS changes of 55 Mongolian lakes (> 10 km2) from 1991 to 2020. The results showed that the LWS increased by 40.24 km3 from 1991 to 1997, especially for northwestern Mongolia with 31.47 km3. However, the LWS decreased by 32.44 km3 from 1998 to 2010, and the lakes in the northwestern and southern decreased by 20.24 km3 (62%) and 7.38 km3 (23%), respectively, and then the LWS continued to decrease by 10.22 km3 from 2011 to 2020. The precipitation was the primary cause of lake change, which could explain 45.13% of LWS change based on Generalized Linear Model, followed by temperature (26.33%) and irrigated area (16.72%). Analyzing the changing characteristic and driving mechanisms of LWS can provide a scientific basis for local water resources management and planning.

Abstract The Tibetan Plateau (TP) is suffering from a substantial decline in terrestrial water storage (TWS) in exorheic basins, threatening water resources that are critical for ∼2 billion people downstream. TWS changes are commonly estimated using gravity satellites through observations of the total terrestrial mass storage (TMS) change, with an implicit assumption of a negligible contribution from sediment transport. Through long‐term (2002–2017) sediment flux observations in seven headwater basins on the TP, we reveal that the gravity satellite‐derived TMS has decreased at a rate of 3.85 ± 0.23 Gt yr−1 in the seven basins, of which 0.35 ± 0.04 Gt yr−1 is contributed by sediment transport. Neglecting this contribution leads to an overestimation of the TWS loss by 10.1 ± 1.3%, equivalent to the annual water demand of an additional 0.62 million people in the surrounding nations. Regionally, the overestimation is surprisingly high in the Indus River and Yarkant River basins, reaching up to 50.8%–77.6%.

Accurate lake depth mapping and estimation of changes in water level and water storage are fundamental significance for understanding the lake water resources on the Tibetan Plateau. In this study, combined with satellite images and bathymetric data, we comprehensively evaluate the accuracy of a multi-factor combined linear regression model (MLR) and machine learning models, create a depth distribution map and compare it with the spatial interpolation, and estimate the change of water level and water storage based on the inverted depth. The results indicated that the precision of the random forest (RF) was the highest with a coefficient of determination (R2) value (0.9311) and mean absolute error (MAE) values (1.13 m) in the test dataset and had high reliability in the overall depth distribution. The water level increased by 9.36 m at a rate of 0.47 m/y, and the water storage increased by 1.811 km3 from 1998 to 2018 based on inversion depth. The water level change was consistent with that of the Shuttle Radar Topography Mission (SRTM) method. Our work shows that this method may be employed to study the water depth distribution and its changes by combining with bathymetric data and satellite imagery in shallow lakes.

It is essential to study how to obtain precise bathymetric and mapping in inland lakes only from satellite data. In this study, the Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) photon point cloud was first denoised by combining the manual interaction (MI) and density-based spatial clustering of applications with noise (DBSCAN) algorithm in QiXiang Co (QXC) and Caiduochaka (CK) on the Tibetan Plateau (TP). Second, we proposed to fit the complete profile depth of the along-track profile and the constructed free profile by using a mathematical function. Finally, the random forest (RF) and linear regression (LR) models were used to obtain bathymetric maps, and bathymetric performance and the benefits of free profile for improving accuracy were verified with the measured depth. The results indicated that the combination of the MI and DBSCAN algorithm could obtain signal photons, and the mean absolute error (MAE) between the fitted free profile water depth and the measured profile water depth was less than 0.90 m. Compared with the original tracks (gt1l, gt2l, and gt3l) in QXC (0–28 m), the MAE of the RF and LR models decreased by 0.35 m and 0.46 m for the validation dataset after adding the free profile. Similarly, the MAE of the RF and LR models decreased by 0.11–0.33 m and 0.13–0.15 m in CK, respectively. This study suggested that this method could be used to obtain the bathymetric map and improve accuracy by fitting profile depth based on ICESat-2 and Sentinel-2 data in shallow water.

The variation of lake water storage is an important indicator for studying both climate change and ecological environment changes. Previous studies have mainly focused on the lake storage changes in recent decades, and predicting future lake storage changes on the Tibetan Plateau under climate change scenarios remains a crucial gap. We addressed this gap by establishing prediction models for water storage changes in nine lakes using historical water storage and climate data from the past 29 years and predicting the water storage changes for the next 80 years under three scenarios based on Phase 6 of the Coupled Model Intercomparison Project (CMIP6) data. The Quantile-mapping (QM) method was applied to correct the precipitation data of CMIP6 with assimilated data. The results indicated that the prediction model performed well, with high correlation (R2 > 0.7 for the training set) and low mean absolute error (MSE < 0.1 km3). The results suggest that most lakes will experience a slight increase in water storage until 2050, followed by a rapid rise until 2100 under all three SSP (Shared Socioeconomic Pathways) scenarios, including SSP126, SSP245, and SSP585. By the end of the century, the total projected increase in lake water storage is estimated to be 189.676 ± 16.266 km3, 191.762 ± 10.683 km3, and 186.212 ± 6.441 km3 until 2100, respectively.

Salt lakes on the Tibetan Plateau (TP) are rich in lithium (Li), boron (B) and other mineral resources, and accurate assessment of the mineral content and spatial distribution of the brine in those salt lakes is important to guide the development and utilization of their mineral resources. There are few studies estimating the mineral content of salt lakes on the TP due to the lack of in situ investigation data. This study introduced an intelligent prediction model combining a feature selection algorithm with a machine learning algorithm using Sentinel-2 satellite data to estimate the Li, B, and TDS contents of Bieruoze Co and Guopu Co lakes on the TP. First, to enrich the spectral information, four mathematical transformations (reciprocal, logarithmic, reciprocal of logarithm, and first-order derivative) were applied to the original bands. Then, feature selection was performed using the genetic algorithm (GA) to select the optimal input variables for the model. Finally, prediction models were constructed by partial least squares regression (PLSR), multiple linear regression (MLR), and random forest (RF). The results showed that: 1) The spectral mathematical transformation provided rich spectral information for the mineral content estimation. 2) The performance of the estimation model constructed by the feature optimization method using GA was better than that of the estimation model constructed based on all spectral bands. Based on GA for feature optimization, the MAPE of GA-RF for estimating Li, B and TDS contents on the testing set was reduced by 77.52%, 28.54% and 36.79%, respectively. 3) Compared with the GA-MLR and GA-PLSR models, GA-RF estimated Li (R2=0.99, RMSE=1.15 mg L-1, MAPE=3.00%), B (R2=0.97, RMSE=10.65 mg L-1, MAPE=2.73%), and TDS (R2=0.93, RMSE=0.60 g L-1, MAPE=1.82%) all obtained the optimal performance. This study showed that the combination of the GA-based feature selection method and the RF model has excellent performance and applicability for monitoring the content of multiple minerals using Sentinel-2 imagery in salt lakes on the TP.

High-accuracy lake bathymetry and mapping are crucial for estimating lake water storage on the Tibetan Plateau (TP). In this article, we constructed traditional empirical (TE) models and machine learning (ML) models to compare the prediction accuracy and remote sensing bathymetric mapping performance by using Sentinel-2 satellite imagery and in situ measured water depth from Caiduochaka (CK) and QiXiang Co in the central TP. We analyzed the relationship between the band reflectance and depth and explored the universality of the model in different lakes. The results indicated that when using the TE model, the mean absolute percentage error (MAPE) varied between 14.5% and 26.5% for the test dataset at different study sites. When using the ML models, the MAPE varied between 7.6% and 18.9%, and it was the better choice overall. For the test dataset of the random forest model with the highest accuracy, in the CK with the maximum depth of approximately 16 m, the mean absolute error (MAE) and root-mean-square error (RMSE) were 0.54 and 0.89 m, and the precision was the highest with an MAE of 1.13 m and RMSE of 1.67 m in QiXiang Co with a maximum depth of approximately 28 m, whereas the portability of the model was not satisfactory. Overall, the results indicated that the ML model can obtain bathymetric maps with high accuracy, good visual performance, and reliability, outperforming the TE model. It can be used effectively for deriving accurate and updated high-resolution bathymetric maps for shallow lakes.

The Tibetan Plateau (TP) has the largest number of high-altitude glaciers on Earth. As a source of major rivers in Asia, this region provides fresh water to more than one billion people. Any terrestrial water storage (TWS) changes there have major societal effects in large parts of the continent. Due to the recent acceleration in global warming, part of the water environment in TP has become drastically unbalanced, with an increased risk of water disasters. We quantified secular and monthly glacier-mass-balance and TWS changes in water basins from April 2002 to December 2021 through the Gravity Recovery and Climate Experiment and its Follow-on satellite mission (GRACE/GRACE-FO). Adequate data postprocessing with destriping filters and gap filling and two regularization methods implemented in the spectral and space domain were applied. The largest glacier-mass losses were found in the Nyainqentanglha Mountains and Eastern Himalayas, with rates of −4.92 ± 1.38 Gt a−1 and −4.34 ± 1.48 Gt a−1, respectively. The Tien Shan region showed strong losses in its eastern and central parts. Furthermore, we found small glacier-mass increases in the Karakoram and West Kunlun. Most of the glacier mass change can be explained by snowfall changes and, in some areas, by summer rainfall created by the Indian monsoon. Major water basins in the north and south of the TP exhibited partly significant negative TWS changes. In turn, the endorheic region and the Qaidam basin in the TP, as well as the near Three Rivers source region, showed distinctly positive TWS signals related to net precipitation increase. However, the Salween River source region and the Yarlung Zangbo River basin showed decreasing trends. We suggest that our new and improved TWS-change results can be used for the maintenance of water resources and the prevention of water disasters not only in the TP, but also in surrounding Asian countries. They may also help in global change studies.

Rapid urban development in China has aggravated land subsidence, which poses a potential threat to sustainable urban development. It is imperative to monitor and predict land subsidence over large areas. To address these issues, we chose Henan Province as the study area and applied small baseline subset-interferometric synthetic aperture radar (SBAS-InSAR) technology to obtain land deformation information for monitoring land subsidence from November 2019 to February 2022 with 364 multitrack Sentinel-1A satellite images. The current traditional time-series deep learning models suffer from the problems of (1) poor results in extracting a sequence of information that is too long and (2) the inability to extract the feature information between the influence factor and the land subsidence well. Therefore, a long short-term memory-temporal convolutional network (LSTM-TCN) deep learning model was proposed in order to predict land subsidence and explore the influence of environmental factors, such as the volumetric soil water layer and monthly precipitation, on land subsidence in this study. We used leveling data to verify the effectiveness of SBAS-InSAR in land subsidence monitoring. The results of SBAS-InSAR showed that the land subsidence in Henan Province was obvious and uneven in spatial distribution. The maximum subsidence velocity was −94.54 mm/a, and the uplift velocity was 41.23 mm/a during the monitoring period. Simultaneously, the land subsidence in the study area presented seasonal changes. The rate of land subsidence in spring and summer was greater than that in autumn and winter. The prediction accuracy of the LSTM-TCN model was significantly better than that of the individual LSTM and TCN models because it fully combined their advantages. In addition, the prediction accuracies, with the addition of environmental factors, were improved compared with those using only time-series subsidence information.

The estimation of lake depth and water storage is crucial for water resource management in inland waters. In previous studies, lake water storage estimation required in situ bathymetry data; however, the high cost of fieldwork prevented the extension of this method to all lakes. This study aims to solve this problem by studying how to estimate lake water storage without using bathymetric data. According to the similarity between the topography surrounding the lake and the underwater topography, we present an approach to estimate the lake water storage. The principle of the method is to extend the surrounding terrain into the underwater area to establish an underwater model. In this study, six lakes of the Tibetan Plateau were selected, and the results showed that the average R2 between the estimated and measured water depths of the six lakes was about 0.76. Among them, the estimated water depth of Taiyang Lake (maximum depth of 61.57 m) had the largest deviation from the measured water depth, with an RMSE (root mean square error) of 19.77 m, while the estimated water depth of Bangda Co (maximum depth of 41.04 m) had the best result, with an RMSE of 7.83 m. The relative errors of all water storage estimations were less than 20%. The lake with the best accuracy was Zigetang Co (measured water storage of 39.6 Gt), and the estimated water storage was 42.2 Gt, with a relative error of 6.57%. The lake with the worst accuracy was Taiyang Lake (measured water storage of 31.94 Gt), and the estimated water storage was 36.47 Gt, with a relative error of 14.18%. This study provides an effective method for lake depth estimation and improves the ability to monitor changes in lake storage, especially for lakes with harsh environments or in remote areas.

There is a large amount of lakes on the Tibetan Plateau (TP), which are very sensitive to climate change. Understanding the characteristics and driving mechanisms of lake change are crucial for understanding climate change and the effective use of water resources. Previous studies have mainly focused on inter-annual lake variation, but the continuous and long-term intra-annual variation of lakes on the TP remains unclear. To address this gap, we used the global surface water (GSW) dataset and the Shuttle Radar Topography Mission (SRTM) DEM to estimate the water level and storage changes on the TP. The results indicated that the average annual minimum lake water level (LWLmin) and the average annual maximum lake water level (LWLmax) increased by 3.09 ± 0.18 m (0.16 ± 0.01 m/yr) and 3.69 ± 0.12 m (0.19 ± 0.01 m/yr) from 2000 to 2018, respectively, and the largest change of LWLmin and LWLmax occurred in 2002–2003 (0.45 m) and 2001–2002 (0.39 m), respectively. Meanwhile, the annual minimum lake water storage change (LWSCmin) and annual maximum lake water storage change (LWSCmax) were 125.34 ± 6.79 Gt (6.60 ± 0.36 Gt/yr) and 158.07 ± 4.52 Gt (8.32 ± 0.24 Gt/yr) from 2000 to 2018, and the largest changes of LWSCmin and LWSCmax occurred in the periods of 2002–2003 (17.67 Gt) and 2015–2016 (17.51 Gt), respectively. The average intra-year changes of lake water level (LWLCintra-year) and the average intra-year changes of lake water storage (LWSCintra-year) were 0.98 ± 0.23 m and 40.19 ± 10.67 Gt, respectively, and the largest change in both LWLCintra-year (1.44 m) and LWSCintra-year (62.46 Gt) occurred in 2018. The overall trend of lakes on the TP was that of expansion, where the LWLC and LWSC in the central and northern parts of the TP was much faster than that in other regions, while the lakes in the southern part of the TP were shrinking, with decreasing LWLC and LWSC. Increased precipitation was found to be the primary meteorological factor affecting lake expansion, and while increasing glacial meltwater also had an important influence on the LWSC, the variation of evaporation only had a little influence on lake change.

Lakes and glaciers are widely distributed on the Tibetan Plateau and are linked via hydrological processes. They are experiencing rapid changes due to global warming, but their relationships during the Holocene are less well known due to limited coupled geological records. Here, we analyzed the δ13C-VPDB and δ18O-VPDB values and ion content of calcite and aragonite in a 407-cm-long sediment core from Guozha Co, a closed basin on the northwestern Tibetan Plateau supplied by glacial meltwater, in order to understand how the lake responded to glacier changes during the Holocene. Our results indicate that the glacial meltwater lowered the lake’s temperature and the δ18Olake water and δ18Oendogenic + authigenic carbonate values and diluted the ion concentrations in the lake water. Three stages of evolution, 8.7–4.0, 4.0–1.5, and 1.5 kyr BP to present, are distinguished based on the decrease in glacial meltwater recharge. Guozha Co has been a closed basin since at least 8.7 kyr BP, and it has changed from a fresh water lake during 8.7–1.5 kyr BP to a brackish lake from 1.5 kyr BP to present due to several climate events. The famous 4.2 kyr BP cold event was identified in the core at 4.0 kyr BP, while warm events occurred at 6.2, 3.9, 2.2, 0.9, and 0.4 kyr BP. Both glaciers and lakes in this area are controlled by climate, but they exhibit opposite changes, that is, glaciers retreat and lakes expand, and vice versa. Our results provide an accurate interpretation of the cold events based on carbonate minerals and carbon–oxygen isotopes in glacial meltwater–recharged lake sediments.

Time series of the Gravity Recovery and Climate Experiment (GRACE) satellite mission have been successfully used to reveal changes in terrestrial water storage (TWS) in many parts of the world. This has been hindered in the interior of the Tibetan Plateau since the derived TWS changes there are very sensitive to the selections of different available GRACE solutions, and filters to remove north-south-oriented (N-S) stripe features in the observations. This has resulted in controversial distributions of the TWS changes in previous studies. In this paper, we produce aggregated hydrology signals (AHS) of TWS changes from 2003 to 2009 in the Tibetan Plateau and test a large set of GRACE solution-filter combinations and mascon models to identify the best combination or mascon model whose filtered results match our AHS. We find that the application of a destriping filter is indispensable to remove correlated errors shown as N-S stripes. Three best-performing destriping filters are identified and, combined with two best-performing solutions, they represent the most reliable solution-filter combinations for determination of weak terrestrial water storage changes in the interior of the Tibetan Plateau from GRACE. In turn, more than 100 other tested solution-filter combinations and mascon solutions lead to very different distributions of the TWS changes inside and outside the plateau that partly disagree largely with the AHS. This is mainly attributed to less effective suppression of N-S stripe noises. Our results also show that the most effective destriping is performed within a maximum degree and order of 60 for GRACE spherical harmonic solutions. The results inside the plateau show one single anomaly in the TWS trend when additional smoothing with a 340-km-radius Gaussian filter is applied. We suggest using our identified best solution-filter combinations for the determination of TWS changes in the Tibetan Plateau and adjacent areas during the whole GRACE operation time span from 2002 to 2017 as well as the succeeding GRACE-FO mission.

Soil is an important natural resource. The excessive amount of heavy metals in soil can harm and threaten human health. Therefore, monitoring of soil heavy metal content is urgent. Monitoring soil heavy metals by traditional methods requires many human and material resources. Remote sensing has shown advantages in the field of monitoring heavy metals. Based on 971 heavy metal samples and Sentinel-2 multi-spectral images in Tai Lake, China, we analyzed the correlation between six heavy metals (Cd, Hg, As, Pb, Cu, Zn) and spectral factors, and selected As and Hg as the input factors of inversion model. The correlation coefficient of the best model of As was 0.53 (p < 0.01), and of Hg was 0.318 (p < 0.01). We used the methods of partial least squares regression (PLSR) and back propagation neural network (BPNN) to establish inversion models with different combinations of spectral factors by using 649 measured samples. In addition, 322 measured samples were used for accuracy evaluation. Compared with the PLSR model, the BP neural network builds the model with higher accuracy, and B1-B4 combined with LnB1-LnB4 builds the model with the highest accuracy. The accuracy of the best model was verified, with an average error of 19% for As and 45% for Hg. Analyzing the spatial distribution of heavy metals by using the interpolation method of Kriging and IDW. The overall distribution trend of the two interpolations is similar. The concentration of As elements tends to increase from north to south, and the relatively high value of Hg elements is distributed in the east and west of the study area. The factories in the study area are distributed along rivers and lakes, which is consistent with the spatial distribution of heavy metal enrichment areas. The relatively high-value areas of heavy metal elements are related to the distribution of metal products factories, refractory porcelain factories, tile factories, factories and mining enterprises, etc., indicating that factory pollution is the main reason for the enrichment of heavy metals.

Water resources are rich on the Tibetan Plateau, with large amounts of glaciers, lakes, and permafrost. Terrestrial water storage (TWS) on the Tibetan Plateau has experienced a significant change in recent decades. However, there is a lack of research about the spatial difference between TWSC and lake water storage change (LWSC), which is helpful to understand the response of water storage to climate change. In this study, we estimate the change in TWS, lake water storage (LWS), soil moisture, and permafrost, respectively, according to satellite and model data during 2005−2013 in the inner Tibetan Plateau and glacial meltwater from previous literature. The results indicate a sizeable spatial difference between TWSC and LWSC. LWSC was mainly concentrated in the northeastern part (18.71 ± 1.35 Gt, 37.7% of the total) and southeastern part (22.68 ± 1.63 Gt, 45.6% of the total), but the increased TWS was mainly in the northeastern region (region B, 18.96 ± 1.26 Gt, 57%). Based on mass balance, LWSC was the primary cause of TWSC for the entire inner Tibetan Plateau. However, the TWS of the southeastern part increased by 3.97 ± 2.5 Gt, but LWS had increased by 22.68 ± 1.63 Gt, and groundwater had lost 16.91 ± 7.26 Gt. The increased TWS in the northeastern region was equivalent to the increased LWS, and groundwater had increased by 4.47 ± 4.87 Gt. Still, LWS only increased by 2.89 ± 0.21 Gt in the central part, and the increase in groundwater was the primary cause of TWSC. These results suggest that the primary cause of increased TWS shows a sizeable spatial difference. According to the water balance, an increase in precipitation was the primary cause of lake expansion for the entire inner Tibetan Plateau, which contributed 73% (36.28 Gt) to lake expansion (49.69 ± 3.58 Gt), and both glacial meltwater and permafrost degradation was 13.5%.

Lake water storage is essential information for lake research. Previous studies usually used bathymetric data to acquire underwater topography by interpolation method, and to therefore estimate water storage. However, due to the large area of Tibetan Plateau (TP) lakes, the method of bathymetry was challenging to cover the whole region of one lake, and the accuracy of the underwater topography, in which no bathymetric data covered, was low, which resulted in a comparatively large error of lake water storage estimation and its change. In this study, we used Shuttle Radar Topography Mission (SRTM) and in situ bathymetric data to establish the underwater topography of Hohxil Lake (HL) and Lexiewudan Lake (LL) in the Hohxil Region of North TP and estimate and analyzed the changes of lake level and water storage. The results showed HL and LL’s water storage was 5.12 km3 and 5.31 km3 in 2019, respectively, and their level increased by 0.5 m/y and 0.57 m/y during 2003−2018, respectively. They were consistent with those (0.5 m/y and 0.5 m/y) from altimetry data, and they were much more accurate than those results (0.077 m/y and 0.156 m/y) from bathymetric data. These findings indicated that this method could improve the accuracy of lake water storage and change estimation. We estimated water storage of two lakes by combining with multitemporal Landsat images, which had doubled since 1976. Our results suggested that the increasing precipitation may dominate the lake expansion by comparing with the change of temperature and precipitation and the increasing glacial meltwater contributed approximately 4.8% and 10.7% to lake expansion of HL and LL during 2000–2019 based on the glacier mass balance data, respectively.