基于LSTM地形迁移学习的湖泊水储量演化异质性

Heterogeneity in water storage evolution of lakes based on LSTM bathymetry transfer learning

摘要

为厘清内蒙古高原湖泊演化对湖盆地形的响应机制，克服以面积代替水储量进行评估易产生偏差的问题，利用深度学习定量刻画不同湖盆形态对湖泊水储量演化的影响。采用全局预训练-逐湖残差学习策略，将长短期记忆网络（long short-term memory network，LSTM）地形预测模型迁移至27个中大型湖泊，重建湖盆三维地形并反演1990−2024年水储量序列。结果表明：模型交叉验证决定系数为0.758，整体精度良好；27个湖泊可划分为宽浅开阔型、过渡混合型与深碗紧凑型，3类湖泊的面积-储量速率比（V/A）总体差异具有统计学意义（P=0.001 6），且平均水深与V/A显著正相关（r_s=0.78）；湖盆深度是决定水体面积与水储量耦合度的主控因子，湖盆形态是影响内蒙古高原湖泊演化异质性的关键内因。研究结果可为高原湖泊分类施策与生态阈值设定提供科学依据。

Abstract

The Inner Mongolia Plateau's lakes have significantly shrunk due to human activity and climate change, endangering northern China's ecological security. Most long-term assessments used lake area as a proxy for water storage because it was easily obtained through remote sensing. However, basin morphology governed the conversion of water level, surface area, and stored volume, allowing area and water storage to become decoupled, and relying solely on area misrepresented the evolution of water resources. Furthermore, lakes with different morphologies showed noticeably different shrinkage behavior under generally similar anthropogenic and climatic conditions. High-precision underwater topography modeling, however, was still mostly limited to the scale of individual lakes. As a result, the mechanism by which lake evolution responded to basin topography remained poorly constrained, and a regional, morphology-resolved understanding of water-storage evolution was still lacking. A previously published model for predicting underwater topography based on a long short-term memory network (LSTM) was expanded from a single lake to twenty-seven medium-to-large lakes across the plateau. The large differences in lake size and the uneven density of field measurements were reconciled using a two-stage approach that combined global pre-training with per-lake ensemble residual learning. Multi-source data were integrated, including long-term satellite imagery for the extraction of annual water-area series from 1990 to 2024, a global elevation model corrected with satellite laser altimetry, and field bathymetric soundings together with surveyed shoreline elevations. The three-dimensional underwater topography of each lake was reconstructed, and annual water-storage series were calculated by combining the water-surface boundary with the predicted basin floor. A dual-threshold classification framework based on mean water depth and the ratio of the semi-major axis to mean depth was built after morphological parameters were extracted within a uniformly defined maximum stable boundary. Evolution rates were characterized by ordinary least-squares slopes, breakpoints were identified and verified by trend and structural-change tests, and differences among groups were examined with non-parametric tests. Leave-one-out cross-validation produced a mean coefficient of determination of 0.758, and an independent hold-out validation for Daihai yielded 0.761, confirming that the method could be applied to lakes with sparse bathymetric data. The twenty-seven lakes were divided into wide-shallow-open, transitional-mixed, and deep-bowl-compact types; among them, twelve were wide-shallow-open, seven were transitional-mixed, and eight were deep-bowl-compact. The area-to-water-storage rate ratio (V/A) showed an overall difference among the three types that was statistically significant (P = 0.001 6), and mean water depth was significantly and positively correlated with V/A (r_s = 0.78, P < 0.001). These findings indicated that vertical basin depth was the most important morphological factor influencing the relationship between area and water storage. For wide-shallow-open lakes, substituting the rate of area change for the rate of water-storage change overestimated the actual change in water volume by more than sixfold. Basin morphology also modulated the consistency of shrinkage, the timing of breakpoints, and the magnitude of interannual fluctuation. Specifically, wide-shallow-open lakes responded earliest to climatic signals and exhibited the strongest amplification of area fluctuation, whereas deep-bowl-compact lakes shrank most consistently and responded most stably. Basin morphological structure was identified as the key internal factor that controlled the heterogeneity of lake evolution on the Inner Mongolia Plateau. Water storage, rather than area, was shown to be a more reliable and more universal indicator of lake evolution, particularly for wide-shallow-open lakes, for which area-based assessment could substantially overstate the magnitude of change. Basin morphological type should be incorporated as an essential reference dimension in the analysis of lake evolution and in the classified management of water resources. A monitoring and early-warning scheme based on water storage should be given priority in wide, shallow, open lakes so that changes in surface area are not mistaken for the true variation of water resources.

关键词

内蒙古高原 / 湖盆形态 / 双阈值分类 / 水储量 / 面积演化差异

Key words

Inner Mongolia Plateau / lake basin morphology / dual-threshold classification / water storage / difference in area evolution

参考文献

[1] Tao Shengli, Fang Jingyun, Zhao Xia, et al. Rapid loss of lakes on the Mongolian Plateau[J]. Proceedings of the National Academy of Sciences, 2015, 112(7): 2281-2286. DOI: 10.1073/pnas.1411748112.

[2] Zhou Yan, Dong Jinwei, Xiao Xiangming, et al. Continuous monitoring of lake dynamics on the Mongolian Plateau using all available Landsat imagery and Google Earth Engine[J]. Science of the Total Environment, 2019, 689: 366-380. DOI: 10.1016/j.scitotenv.2019.06.341.

[3] Liu Qianqian, Xu Xibao, Chen Minkun. Changing pattern and driving factors of ecosystem service value of the lakes in northern China since 1990[J]. Ecological Indicators, 2024, 158: 111370. DOI: 10.1016/j.ecolind.2023.111370.

[4] Feng Shuzhen, Zheng Shiyu, Guan Weijin, et al. Evolution of the lake area and its drivers during 1990-2021 in Inner Mongolia[J]. Environmental Earth Sciences, 2024, 83: 414. DOI: 10.1007/s12665-024-11709-9.

[5] Xu Haowei, Zhang Fei, Jim C Y, et al. Regional variations in lake areas in China due to human and natural environmental factors since 1990[J]. Ecological Indicators, 2025, 172: 113307. DOI: 10.1016/j.ecolind.2025.113307.

[6] Zhang Mengmeng, Wang Shuai, Gao Guangyao, et al. Exploring responses of lake area to river regulation and implications for lake restoration in arid regions[J]. Ecological Engineering, 2019, 128: 18-26. DOI: 10.1016/j.ecoleng.2018.12.031.

[7] Zhang Yanfei, Liao Zilong, Xu Xiaomin, et al. Investigation of lake shrinkage attributed to climate change over the past 33 years in Inner Mongolia, China[J]. Climatic Change, 2023, 176(2): 10. DOI: 10.1007/s10584-023-03487-3.

[8] Hu Yue, Sun Peng, Liao Fu, et al. Dual 222Rn models for tracing groundwater-lake water exchange in a flow-through lake[J]. Water Resources Research, 2025, 61: e2024WR038689. DOI: 10.1029/2024WR038689.

[9] Gharari S, Vanderkelen I, Tefs A, et al. A flexible framework for simulating the water balance of lakes and reservoirs from local to global scales: Mizu Route-Lake[J]. Water Resources Research, 2024, 60: e2022WR032400. DOI: 10.1029/2022WR032400.

[10] Wan Fang, Zhang Fei, Zheng Xiaokang, et al. Study on ecological water demand and ecological water supplement in Wuliangsuhai Lake[J]. Water, 2022, 14: 1262. DOI: 10.3390/w14081262.

[11] Jiang Ruirui, Wang Likang, Ma Rong, et al. Influence of lake basin topography on the differential shrinkage of lakes on the Inner Mongolia Plateau, China[J/OL]. Journal of Groundwater Science and Engineering, 2026-05-18. https://doi.org/10.26599/JGSE.2026.9280099.

[12] Zhang Jing, Ji Dongli, Liu Futian, et al. Global assessment of lake surface morphology and its impact on water volume estimation[J]. Water Resources Research, 2025, 61(7): e2024WR039627. DOI: 10.1029/2024WR039627.

[13] Kraemer B M, Anneville O, Chandra S. Morphometry and average temperature affect lake stratification responses to climate change[J]. Geophysical Research Letters, 2015, 42(12): 4981-4988. DOI: 10.1002/2015GL064097.

[14] Jiang Ruirui, Ma Rong, Zhao Lefan, et al. Impacts of lake basin topography predicted by long short-term memory on difference of lake area and water resource evolution[J]. Journal of Cleaner Production, 2025, 490: 144781. DOI: 10.1016/j.jclepro.2025.144781.

[15] Liu Kai, Song Chunqiao. Modeling lake bathymetry and water storage from DEM data constrained by limited underwater surveys[J]. Journal of Hydrology, 2022, 604: 127260. DOI: 10.1016/j.jhydrol.2021.127260.

[16] Deng Chunnuan, Li Dafeng, Li Ju, et al. Impacts of underwater topography on the distribution of microplastics in lakes: A case from Dianchi Lake, China[J]. Science of the Total Environment, 2022, 837: 155708. DOI: 10.1016/j.scitotenv.2022.155708.

[17] Wilcox D A, Bateman J A, Kowalski K P, et al. Extent of sedge-grass meadow in a Lake Michigan drowned river mouth wetland dictated by topography and lake level[J]. Wetlands, 2022, 42: 34. DOI: 10.1007/s13157-022-01534-w.

[18] Yu Wei, Liu Fali, Jiao Xianglan, et al. Human-induced N-P imbalances will aggravate GHG emissions from lakes and reservoirs under persisting eutrophication[J]. Water Research, 2025, 276: 123240. DOI: 10.1016/j.watres.2025.123240.

[19] 崔亮, 李永平, 黄国和, 等. 基于Landsat-TM影像的洞庭湖水面动态变化[J]. 南水北调与水利科技（中英文）, 2015, 13(1): 63-66. DOI: 10.13476/j.cnki.nsbdqk.2015.01.015.

[20] 文京川, 赵红莉, 蒋云钟. 卫星测高数据筛选方法研究: 以Jason-3数据和洪泽湖为例[J]. 南水北调与水利科技（中英文）, 2018, 16(3): 194-200. DOI: 10.13476/j.cnki.nsbdqk.2018.0088.

[21] Gorelick N, Hancher M, Dixon M, et al. Google Earth Engine: planetary-scale geospatial analysis for everyone[J]. Remote Sensing of Environment, 2017, 202: 18-27. DOI: 10.1016/j.rse.2017.06.031.

[22] Xu Hanqiu. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery[J]. International Journal of Remote Sensing, 2006, 27(14): 3025-3033. DOI: 10.1080/01431160600589179.

[23] McFeeters S K. The use of the normalized difference water index (NDWI) in the delineation of open water features[J]. International Journal of Remote Sensing, 1996, 17(7): 1425-1432. DOI: 10.1080/01431169608948714.

[24] Otsu N. A threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1979, 9(1): 62-66. DOI: 10.1109/TSMC.1979.4310076.

[25] Pekel J F, Cottam A, Gorelick N, et al. High-resolution mapping of global surface water and its long-term changes[J]. Nature, 2016, 540(7633): 418-422. DOI: 10.1038/nature20584.

[26] Neuenschwander A, Pitts K. The ATL08 land and vegetation product for the ICESat-2 mission[J]. Remote Sensing of Environment, 2019, 221: 247-259. DOI: 10.1016/j.rse.2018.11.005.

[27] Chen Tianqi, Guestrin C. XGBoost: A scalable tree boosting system[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, CA, USA: ACM, 2016: 785-794. DOI: 10.1145/2939672.2939785.

[28] Chen Chuanfa, Yang Shuai, Li Yanyan. Accuracy assessment and correction of SRTM DEM using ICE-Sat/GLAS data under data coregistration[J]. Remote Sensing, 2020, 12(20): 3435. DOI: 10.3390/rs12203435.

[29] Ram P, Sinha K. Revisiting kd-tree for nearest neighbor search[C]// Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. Anchorage, AK, USA: ACM, 2019: 1378-1388. DOI: 10.1145/3292500.3330875.

[30] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780. DOI: 10.1162/neco.1997.9.8.1735.

[31] Håkanson L. The importance of lake morphometry and catchment characteristics in limnology: Ranking based on statistical analyses[J]. Hydrobiologia, 2005, 541(1): 117-137. DOI: 10.1007/s10750-004-5032-7.

[32] Wetzel R G, Likens G E. Lake basin characteristics and morphometry[M].//Wetzel R G, Likens G E. Limnological Analyses. New York: Springer, 1991: 1-14. DOI: 10.1007/978-1-4757-4098-1_1.

[33] 尹明波, 窦明, 刘玉倩, 等. 基于XGBoost-MIKE21耦合模型的缺资料水库富营养化模拟[J]. 南水北调与水利科技（中英文）, 2026, 24(1): 224-234. DOI: 10.13476/j.cnki.nsbdqk.2026.0022.

[34] 钟学敏, 文军, 王卓元, 等. 黄河源区径流预报的机器学习算法研究[J]. 南水北调与水利科技（中英文）, 2026, 24(1): 111-122. DOI: 10.13476/j.cnki.nsbdqk.2026.0011.

[35] Magee M R, Wu C H. Response of water temperatures and stratification to changing climate in three lakes with different morphometry[J]. Hydrology and Earth System Sciences, 2017, 21(12): 6253-6274. DOI: 10.5194/hess-21-6253-2017.

[36] Seekell D A, Byström P, Karlsson J. Lake morphometry moderates the relationship between water color and fish biomass in small boreal lakes[J]. Limnology and Oceanography, 2018, 63(5): 2171-2178. DOI: 10.1002/lno.10931.

[37] McMahon T A, Peel M C, Lowe L, et al. Estimating actual, potential, reference crop and pan evaporation using standard meteorological data: A pragmatic synthesis[J]. Hydrology and Earth System Sciences, 2013, 17(4): 1331-1363. DOI: 10.5194/hess-17-1331-2013.

[38] Du Yajun, Wan Weifeng, Li Qingbo, et al. Impacts of climate and human activities on Daihai Lake in a typical semi-arid watershed, northern China[J]. PLoS ONE, 2022, 17(5): e0266049. DOI: 10.1371/journal.pone.0266049.

[39] Wang Shuhang, Xu Chi, Zhang Wanchang, et al. Human-induced water loss from closed inland lakes: Hydrological simulations in China's Daihai Lake[J]. Journal of Hydrology, 2022, 607: 127552. DOI: 10.1016/j.jhydrol.2022.127552.

[40] Scanlon B R, Keese K E, Flint A L, et al. Global synthesis of groundwater recharge in semiarid and arid regions[J]. Hydrological Processes, 2006, 20(15): 3335-3370. DOI: 10.1002/hyp.6335.

[41] Chen Jiaqi, Lyu Jiming, Li Ning, et al. External groundwater alleviates the degradation of closed lakes in semi-arid regions of China[J]. Remote Sensing, 2019, 12(1): 45. DOI: 10.3390/rs12010045.

[42] Ma Rong, Chen Peng, Zhang Bing, et al. Identification of ecological restoration standard for lakes on the Inner Mongolia Plateau, China[J]. Ecological Indicators, 2024, 162: 111993. DOI: 10.1016/j.ecolind.2024.111993.

基金资助

国家重点研发计划项目(2024YFC3211600)

中国地质调查局黑河流域水循环野外站联合开放基金项目(WCSHR-2024-08)