A runoff reconstruction method for glacier-covered areas without observed data

Abstract

The application discloses a runoff reconstruction method for a glacier-covered area without actual measurement data, relates to the technical field of geographic information and hydrology, and comprises the following steps: acquiring multi-source heterogeneous data of a target basin; extracting an area time sequence of an ice lake based on remote sensing images, constructing an area-water level-storage capacity relationship curve based on a digital elevation model, and determining an ice lake storage capacity change time sequence; constructing an ice hydrology physical model and integrating an ice lake reservoir module based on ice distribution data and meteorological reanalysis forcing data; determining a prior distribution through a cross-basin parameter migration technology, constructing a likelihood function based on the ice lake storage capacity change time sequence, determining an optimal parameter set based on Bayes theorem, and running the ice hydrology physical model to obtain a primary simulated runoff sequence; and performing deviation correction through a correction model based on the primary simulated runoff sequence, the ice lake storage capacity simulation time sequence and an environmental characteristic sequence to generate a reconstructed runoff sequence. The application can improve the accuracy and reliability of runoff reconstruction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geographic information and hydrology science, and relates to but is not limited to a runoff reconstruction method for an ice-covered area without actual measurement data. BACKGROUND

[0002] Accurate runoff simulation in ice-covered high mountain basins is crucial for regional water resources management and disaster prevention. However, due to the lack of observation stations or imperfect observation station settings in most areas, traditional hydrological model methods face several key limitations. First, the lack of continuous flow observation data makes it impossible to calibrate the hydrological model, resulting in a large uncertainty in the prediction results. Second, the complex cryosphere processes, especially the dynamic evolution of glacial lakes and the occurrence of glacial lake outburst floods, can greatly change the runoff pattern, but these situations are often oversimplified in existing models. In addition, although satellite products provide valuable data sources, remote sensing observations often have temporal discontinuity and inversion errors due to complex terrain and cloud cover. At the same time, pure deep learning methods have inherent limitations in generalization ability in environments where data is scarce and physical consistency is crucial.

[0003] In the prior art, the runoff simulation methods in the ice-covered area mainly include conceptual or distributed physical hydrological model methods, runoff estimation methods based on remote sensing empirical regression, and runoff prediction methods based on deep learning. However, under the condition of no or lack of actual hydrological data, the existing technologies have obvious limitations. Conceptual or distributed physical hydrological models usually rely on runoff observations for parameter calibration and structure verification. Under the condition of lack of data, the reliability of parameters decreases significantly, which can lead to increased equivalence and state drift, and thus weaken the credibility of the runoff sequence. Remote sensing empirical regression or runoff inversion methods based on a single observation (such as ice lake area, snow cover area, etc.) are affected by factors such as spatial and temporal resolution, cloud and snow shielding, and inversion error, and can only provide discrete phase or seasonal scale estimation results, making it difficult to generate continuous, process-consistent daily runoff sequences. Pure deep learning methods can achieve good fitting results in areas with data, but in high mountain cryospheres with significant sample scarcity and environmental heterogeneity, there are problems such as insufficient stability of model extrapolation, and due to the lack of explicit expression of energy-water constraints and ice-snow meltwater physical mechanisms, physical inconsistency occurs. Moreover, the formation, expansion, and drainage processes of glacial lakes have a significant impact on runoff storage and mutation, but existing methods often ignore or simplify the expression of ice lake change processes, making it difficult to reconstruct credible runoff processes in ice lake-distributed lack of measurement ice-covered areas.

[0004] Therefore, there is an urgent need for a multi-source coupled runoff reconstruction method for ice lake distribution in an ungauged glacier basin, which couples hydrological models, remote sensing inversion and deep learning models to solve the related problems existing in the prior art, achieve reliable hydrological simulation in a data-scarce glaciated basin, and provide a generalizable, interpretable and uncertainty characterization capable methodological support for water resource assessment and risk analysis under the background of cryosphere change. SUMMARY

[0005] The embodiment of the present application provides a runoff reconstruction method for an ice-covered area without actual measurement data.

[0006] The technical scheme of the embodiment of the present application is as follows:

[0007] In a first aspect, the embodiment of the present application provides a runoff reconstruction method for an ice-covered area without actual measurement data, which comprises: acquiring a digital elevation model, ice distribution data, meteorological reanalysis forcing data and optical and radar remote sensing images of a target basin; extracting an ice lake area time series based on the optical and radar remote sensing images, and constructing an area-water level-storage relationship curve based on the ice lake area time series and the digital elevation model to determine an ice lake storage change amount time series; based on the ice distribution data and the meteorological reanalysis forcing data, constructing an ice hydrological physical model using an ice hydrological model as a simulation skeleton, and integrating an ice lake reservoir module for simulating dynamic changes of ice lake storage in the ice hydrological physical model; determining a prior distribution of a model parameter set in the ice hydrological physical model through a cross-basin parameter migration technology, constructing a likelihood function based on the ice lake storage change amount time series, and determining an optimal parameter set based on the Bayes theorem in combination with the prior distribution and the likelihood function, running the ice hydrological physical model based on the optimal parameter set to obtain a primary simulated runoff sequence; and performing bias correction through a pre-trained correction model based on the primary simulated runoff sequence, the ice lake storage simulation time series and the environmental characteristic sequence of the target basin to generate a reconstructed runoff sequence of the target basin.

[0008] The technical scheme provided in the application comprises the following steps: obtaining a digital elevation model of a target basin, glacial distribution data, meteorological reanalysis forcing data and optical and radar remote sensing images; extracting an ice lake area time sequence based on the optical and radar remote sensing images, and constructing an area-water level-storage relationship curve based on the ice lake area time sequence and the digital elevation model to determine an ice lake storage variation time sequence, so as to determine a state consistency constraint condition in an optimization process of a glacial hydrological physical model parameter; constructing the glacial hydrological physical model by taking the glacial hydrological model as a simulation skeleton based on the glacial distribution data and the meteorological reanalysis forcing data, and integrating an ice lake reservoir module for simulating dynamic changes of ice lake storage in the glacial hydrological physical model, which is different from a processing mode of regarding the ice lake as a static water surface or ignoring the regulating effect of the ice lake in the prior art, and the ice lake reservoir module is embedded in the glacial hydrological physical model to explicitly express the regulation and storage effect of the ice lake on runoff and the instantaneous peak discharge process of the burst discharge process, so that the response simulation of the ice lake flood burst event has high physical authenticity, thereby enabling the runoff reconstruction result to reflect the nonlinear characteristics of the ice-snow water cycle under an extreme climate background; determining a prior distribution of a model parameter set in the glacial hydrological physical model through a cross-basin parameter migration technology, and constructing a likelihood function based on the ice lake storage variation time sequence, so as to directly introduce the ice lake storage variation time sequence obtained by remote sensing inversion into a parameter optimization link of the glacial hydrological physical model, and based on a state constraint mechanism of the remote sensing observed storage, the limitation that the remote sensing data is only used for posterior verification in the prior art is changed, and by constructing the likelihood function, the remote sensing data is used to correct intermediate state variables of the glacial hydrological physical model in real time, and this “process constraint” ensures that the runoff generation law of the model still conforms to the physical fluctuation characteristics of the ice lake storage even in the case of lacking ground measured runoff, thereby greatly improving the physical reliability of the simulation result; determining an optimal parameter set based on the Bayes theorem in combination with the prior distribution and the likelihood function, running the glacial hydrological physical model based on the optimal parameter set to obtain a primary simulation runoff sequence; performing bias correction through a pre-trained correction model based on the primary simulation runoff sequence of the target basin, the ice lake storage simulation time sequence and an environmental characteristic sequence to generate a reconstructed runoff sequence of the target basin, and the final runoff reconstruction result has strict physical logic support, each time point of runoff component (glacial meltwater, snowmelt, rainfall runoff and ice lake contribution) can be traced and traced, and conforms to the mass balance principle, through the uncertainty estimation of the Bayes inference and deep learning, the model can simultaneously output the uncertainty confidence interval of the simulation result, thereby simultaneously considering the physical interpretability and uncertainty quantification, and providing an important risk reference basis for decision makers in water resource allocation and disaster assessment.The technical scheme provided in the application breaks away from the strong dependence of traditional hydrological models on parameter calibration of runoff records, realizes long-time series continuous runoff reconstruction under the condition of missing hydrological data, and experiments show that the runoff reconstruction method provided in the application can output a runoff sequence with high time resolution in an unpopulated area with high altitude lacking hydrological station observation, thereby providing reliable data support for cross-border river management and remote mountainous area water resource assessment.

[0009] Optionally, the ice lake area time sequence is extracted based on the optical and radar remote sensing images, and an area-water level-storage capacity relationship curve is constructed based on the ice lake area time sequence and the digital elevation model to determine the ice lake storage capacity change time sequence, including: based on the optical and radar remote sensing images, an ice lake area time sequence of a target ice lake in the target basin is extracted by a remote sensing image automatic extraction algorithm; based on the ice lake area time sequence, the digital elevation model and the terrain features, an area-water level-storage capacity relationship curve of the target ice lake is constructed; based on each phase ice lake area value in the ice lake area time sequence, the ice lake storage capacity time sequence and the ice lake storage capacity change time sequence of the target ice lake are inversely determined according to the area-water level-storage capacity relationship curve.

[0010] Optionally, the mathematical expression of the glacial hydrological physical model is represented by the following formula:

[0011] ;

[0012] In the formula, represents the primary simulation runoff at the moment t; represents a function set of the GSM-Socont model describing hydrological physical processes; represents the air temperature forcing data at the moment t; represents the precipitation forcing data at the moment t; represents the ice lake area; represents a set of model parameters to be determined, and the set of model parameters is independent of the measured runoff data calibration.

[0013] Optionally, the ice lake reservoir module simulates the dynamic change of the storage capacity of the ice lake through a mass balance equation, and the mass balance equation is represented by the following formula:

[0014] ;

[0015] In the formula, represents the ice lake storage capacity simulation value at the moment t; represents ice lake storage simulation value at a time point; indicates total inflow water volume of the ice lake at a time point; indicates total outflow water volume of the ice lake at a time point, the total outflow water volume being defined by a condition-triggered segmented function for describing a regular outflow process, a spillway process and a sudden breaching process of the ice lake; indicates lake surface evaporation of the ice lake at a time point.

[0016] Optionally, the step of determining the prior distribution of the model parameter set in the ice hydrological physical model through the cross-basin parameter migration technology, constructing a likelihood function based on the ice lake storage variation time series, and determining the optimal parameter set based on Bayes' theorem in combination with the prior distribution and the likelihood function comprises: calculating the basin attribute similarity between a target basin and a reference basin through a basin attribute vector, establishing a parameter mapping space, performing a cross-basin parameter migration technology, and obtaining the prior distribution of the model parameter set in the ice hydrological physical model, wherein the basin attribute vector is composed according to terrain slope, ice coverage, latitude and vegetation index; taking the ice lake storage variation time series as a state consistency constraint condition, constructing a likelihood function based on ice lake state observation, the likelihood function representing the matching probability of the ice lake storage simulation time series simulated by the ice hydrological physical model and the ice lake storage variation time series under the condition of a given model parameter set; determining the posterior distribution of the model parameter set based on Bayes' theorem in combination with the prior distribution and the likelihood function, and determining the optimal parameter set of the ice hydrological physical model according to the posterior distribution, the calculation formula of the posterior distribution being represented by the following formula:

[0017] ;

[0018] In the formula, indicates the ice lake storage variation time series inversed by remote sensing; indicates the posterior distribution; indicates the likelihood function; indicates the prior distribution.

[0019] Optionally, the preliminary simulation runoff sequence, the ice lake storage simulation time sequence and the environmental characteristic sequence of the target basin are subjected to bias correction by a pre-trained correction model to generate a reconstructed runoff sequence of the target basin, comprising: constructing a deep learning-based correction model, taking the preliminary simulation runoff reference sequence, the ice lake storage simulation time reference sequence and the environmental characteristic reference sequence as input, pre-training the correction model on a reference basin having measured runoff data and similar characteristics to the target basin, learning the residual sequence between the preliminary simulation runoff reference sequence and the corresponding measured runoff sequence of the reference basin, wherein the correction model is a long short-term memory network model or a time convolution network model; inputting the preliminary simulation runoff sequence, the ice lake storage simulation time sequence and the environmental characteristic sequence of the target basin into the pre-trained correction model, outputting a predicted residual sequence corresponding to the preliminary simulation runoff sequence of the target basin, adding the predicted residual sequence and the preliminary simulation runoff sequence to obtain the reconstructed runoff sequence of the target basin, and synchronously generating diagnostic information, the diagnostic information including the storage contribution of the glacial lake, the contribution degree of the extreme drainage event to the peak flow and the uncertainty confidence interval of the reconstructed runoff sequence, the uncertainty confidence interval being determined by Monte Carlo simulation on the posterior distribution of the model parameter set.

[0020] In a second aspect, an embodiment of the present application provides an electronic device, comprising a memory and a processor, the memory storing a computer program capable of running on the processor, and the processor implements the steps of the above-mentioned runoff reconstruction method for glacier-covered areas without measured data when executing the program.

[0021] In a third aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the above-mentioned runoff reconstruction method for glacier-covered areas without measured data.

[0022] The technical scheme provided by the embodiments of the present application has at least the following beneficial effects:

[0023] The application provides a glacier-covered area runoff reconstruction method, obtains the digital elevation model, glacier distribution data, meteorological reanalysis forcing data and optical and radar remote sensing images of a target basin; extracts an ice lake area time sequence based on the optical and radar remote sensing images, constructs an area-water level-storage capacity relationship curve based on the ice lake area time sequence and the digital elevation model, determines an ice lake storage capacity change time sequence, and determines a state consistency constraint condition in an ice hydrology physical model parameter optimization process; based on the glacier distribution data and the meteorological reanalysis forcing data, an ice hydrology model is used as a simulation skeleton to construct an ice hydrology physical model, and an ice lake reservoir module for simulating dynamic changes of ice lake storage capacity is integrated in the ice hydrology physical model, unlike the processing mode of regarding the ice lake as a static water surface or ignoring the regulation effect of the ice lake in the prior art, the ice lake reservoir module is embedded in the ice hydrology physical model to explicitly express the regulation effect of the ice lake on runoff and the instantaneous peak discharge process of the burst discharge process, the response simulation of the ice lake flood burst event has high physical reality, so that the runoff reconstruction result can reflect the nonlinear characteristics of the ice-snow water cycle under an extreme climate background; the prior distribution of the model parameter set in the ice hydrology physical model is determined through a cross-basin parameter migration technology, a likelihood function is constructed based on the ice lake storage capacity change time sequence, so that the ice lake storage capacity change time sequence obtained by remote sensing inversion is directly introduced into the parameter optimization link of the ice hydrology physical model, and based on the state constraint mechanism of the remote sensing observation storage capacity, the limitation that the remote sensing data is only used for posterior verification in the past is changed, and by constructing the likelihood function, the intermediate state variable of the ice hydrology physical model is corrected in real time by using the remote sensing data, the "process constraint" ensures that the runoff generation law of the model still conforms to the physical fluctuation characteristics of the ice lake storage capacity even in the absence of ground measured runoff, thereby greatly improving the physical reliability of the simulation result, the optimal parameter set is determined based on the Bayes theorem in combination with the prior distribution and the likelihood function, the ice hydrology physical model based on the optimal parameter set is run, and a primary simulation runoff sequence is obtained; based on the primary simulation runoff sequence, the ice lake storage capacity simulation time sequence and the environmental characteristic sequence of the target basin, deviation correction is performed through a pre-trained correction model to generate a reconstructed runoff sequence of the target basin, and the final runoff reconstruction result has strict physical logic support, the runoff components (glacier melt water, snowmelt water, rainfall runoff and ice lake contribution) at each moment can be traced and traced, and meet the mass balance principle, through the uncertainty estimation of the Bayes inference and deep learning, the model can simultaneously output the uncertainty confidence interval of the simulation result, thereby simultaneously considering the physical interpretability and uncertainty quantification, and providing an important risk reference basis for decision makers in water resource allocation and disaster assessment.The technical scheme provided in the application breaks away from the strong dependence of the traditional hydrological model on parameter calibration of runoff records, realizes long-time series continuous runoff reconstruction under the condition of missing hydrological data, and experiments show that the runoff reconstruction method provided in the application can output a runoff sequence with high time resolution in the high-altitude uninhabited area lacking hydrological station observation, thereby providing reliable data support for cross-border river management and remote mountainous area water resource assessment. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor on the premise of not deviating from the concept of the application.

[0025] Figure 1 A flowchart of a runoff reconstruction method for an ice-covered area without actual measurement data provided by the embodiments of the application is shown in the figure.

[0026] Figure 2 A hardware entity schematic diagram of an electronic device provided by the embodiments of the application is shown in the figure. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical scheme and advantages of the embodiments of the application more clear, the technical scheme in the embodiments of the application will be clearly and completely described below in combination with the drawings in the embodiments of the application. Obviously, the described embodiments are some of the embodiments of the application, not all the embodiments of the application. The following embodiments are used to illustrate the application, but not to limit the scope of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0028] In the following description, "some embodiments" are described, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subset of all possible embodiments, and can be combined with each other without conflict.

[0029] It should be noted that the terms "first, second, third" involved in the embodiments of the application are only to distinguish similar objects, and do not represent a specific order of the objects. It can be understood that "first, second, third" can be interchanged in a specific order or sequence as allowed, so that the embodiments of the application described here can be implemented in an order other than that illustrated or described here.

[0030] Those skilled in the art can understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiments belong. It should also be understood that the terms, such as those defined in a general dictionary, should be interpreted as having a meaning consistent with the meaning in the context of the relevant art and should not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0031] The embodiments of the present application are further described below with reference to the accompanying drawings.

[0032] In view of the problems in the field of geographic information and hydrological science technology for the reconstruction of runoff in the ice-covered area without measured data, the embodiments of the present application provide a method for reconstructing runoff in the ice-covered area without measured data.

[0033] The technical solutions of the present application are introduced as follows. First, the method embodiments of the present application are introduced.

[0034] Please refer to Figure 1 , which shows a flowchart of a method for reconstructing runoff in the ice-covered area without measured data according to an embodiment of the present application, as shown in Figure 1 , the method comprises at least the following steps S110 to S140.

[0035] In step S110, the digital elevation model, the ice distribution data, the meteorological reanalysis forcing data and the optical and radar remote sensing images of the target basin are obtained; the ice lake area time series is extracted based on the optical and radar remote sensing images, and the area-water level-storage relationship curve is constructed based on the ice lake area time series and the digital elevation model to determine the ice lake storage capacity change time series.

[0036] In this embodiment, multi-source heterogeneous data of the target watershed is systematically acquired. This multi-source heterogeneous data includes a high-resolution digital elevation model, glacier distribution data selected from the Randolph Glacier catalog, meteorological reanalysis forcing data (such as ERA5-Land), and long-sequence optical and radar remote sensing images selected from the Landsat and Sentinel series. Furthermore, the time series of glacial lake area is extracted based on the optical and radar remote sensing images, and an area-water level-reservoir capacity relationship curve is constructed based on the glacial lake area time series and the digital elevation model to determine the glacial lake reservoir capacity time series and the time series of glacial lake reservoir capacity change. Specifically, based on optical and radar remote sensing images, the time series of glacial lake area of ​​the target glacial lake within the target watershed is extracted using an automatic remote sensing image extraction algorithm. Based on the glacial lake area time series, digital elevation model, and topographic features, the area-water level-storage capacity relationship curve of the target glacial lake is constructed. Based on the glacial lake area values ​​at each time phase in the glacial lake area time series, and according to the area-water level-storage capacity relationship curve, the time series of glacial lake storage capacity and the time series of glacial lake storage capacity change are determined by inversion. This transforms discrete remote sensing observations into continuous time series of glacial lake storage capacity and glacial lake storage capacity change, providing key physical state constraints for subsequent parameter estimation of glacial hydrophysical models.

[0037] Step S120: Based on the glacier distribution data and the meteorological reanalysis forcing data, a glacier hydrological physical model is constructed using the glacier hydrological model as the simulation framework, and a glacier lake reservoir module simulating the dynamic changes in glacier lake capacity is integrated into the glacier hydrological physical model.

[0038] In this embodiment, the glacier hydrological model GSM-Socont is used as the simulation framework to construct the glacier hydrophysical model. The reliance on measured runoff data for calibration is eliminated, and external observation constraint variables are introduced to participate in state updates. The mathematical expression of this glacier hydrophysical model is represented by the following formula:

[0039] ;

[0040] In the formula, express Primary simulated runoff at a given time; This represents the set of functions used by the GSM-Socont model to describe hydrophysical processes. express Temperature forcing data at any given time; express Precipitation forcing data at specific times; Indicates the area of ​​a glacial lake; This represents the set of model parameters to be determined. This set of parameters is not calibrated based on measured runoff data from the target watershed outlet section, but is determined through remote sensing reservoir capacity dynamic constraints and cross-watershed parameter migration techniques. Furthermore, a glacial lake-reservoir module is integrated into the glacial hydrophysical model to simulate dynamic changes in glacial lake capacity, explicitly expressing the impact of glacial lakes on runoff regulation and abrupt changes. The glacial lake-reservoir module simulates dynamic changes in glacial lake capacity, such as dynamic water storage behavior, through a mass balance equation, which is expressed as follows:

[0041] ;

[0042] in, express Simulated value of the ice lake reservoir capacity at any given time; express Simulated value of the ice lake reservoir capacity at any given time; express The total inflow of water into the glacial lake includes glacial meltwater, precipitation runoff, and groundwater runoff. express The total outflow of the glacial lake at any given time is defined by a conditionally triggered piecewise function. This function is used to describe the controlled outflow process of the glacial lake under normal conditions, the overflow process under abnormal conditions, and the sudden bursting process. This allows the model to capture the peak shaving and valley filling effect of the glacial lake on runoff and the sudden drainage response under extreme conditions. express The evaporation rate of the glacial lake surface at any given time. The technical solution provided in this application embeds the glacial lake as a reservoir module with nonlinear regulation and storage functions directly into the main loop of the glacial hydrological physical model. By constructing the mass balance equation of the glacial lake, the accumulation, evaporation, and controlled or natural discharge process of glacial meltwater in the glacial lake are simulated. Through this deep coupling and mechanism improvement at the physical structure dimension, the model can more realistically depict the hydrological feedback mechanism between "glacier-glacial lake-runoff" in high-altitude cold mountainous areas, thereby solving the pain point of insufficient representation of glacial lake regulation and sudden drainage processes in traditional physical models.

[0043] Step S130: Determine the prior distribution of the model parameter set in the glacier hydrophysical model through cross-basin parameter migration technology; construct a likelihood function based on the time series of glacier-lake reservoir capacity change; determine the optimal parameter set based on Bayes' theorem, combining the prior distribution and the likelihood function; and run the glacier hydrophysical model based on the optimal parameter set to obtain the primary simulated runoff sequence.

[0044] In this embodiment, the prior distribution of the model parameter set in the glacier hydrophysical model is determined by introducing cross-basin parameter transfer technology. Specifically, a basin attribute vector is constructed based on basin attribute characteristics such as topographic slope, glacier cover, latitude, and vegetation index. ,in, Indicates the slope of the terrain. Indicates glacier coverage. Indicates latitude, The vegetation index is used to calculate the similarity of watershed attributes between the target watershed and the reference watershed with measured runoff data using watershed attribute vectors. A parameter mapping space is established, and after performing cross-watershed parameter transfer technology, the prior distribution of the model parameter set in the glacier hydrophysical model can be obtained. Furthermore, a likelihood function is constructed based on the time series of glacier-lake storage capacity changes. Based on Bayes' theorem, the optimal parameter set is determined by combining the prior distribution and the likelihood function. Specifically, since using the remotely sensed glacier-lake storage capacity change time series as the core constraint information can reduce the uncertainty in the cross-watershed parameter transfer process, the glacier-lake storage capacity change time series is introduced as a state consistency constraint. The expression of this state consistency constraint is given by the following formula:

[0045] ;

[0046] In the formula, express Simulated value of the ice lake reservoir capacity at any given time; express Real-time remote sensing inversion of changes in the reservoir capacity of the glacial lake; The permissible error threshold is represented by the state consistency constraint, which aims to ensure that the simulated time series of glacial lake capacity by the glacial hydrophysical model is as close as possible to the time series of glacial lake capacity change retrieved from remote sensing. To achieve this constraint, a likelihood function based on glacial lake state observations is constructed. This likelihood function characterizes the probability of matching between the simulated time series of glacial lake capacity and the time series of glacial lake capacity change, given a set of model parameters. Based on Bayes' theorem, and combining the prior distribution and the likelihood function, the posterior distribution of the model parameter set is determined. The formula for calculating this posterior distribution is expressed as follows:

[0047] ;

[0048] In the formula, This represents the time series of changes in the reservoir capacity of glacial lakes retrieved through remote sensing. Indicate the posterior distribution; Represents the likelihood function; The prior distribution is represented, and in the determination process of the posterior distribution, the optimization of the key model parameters in the glacial hydrological physical model, such as the glacial ablation factor and the runoff coefficient which are difficult to determine, is driven by minimizing the difference between the ice lake storage simulation time sequence and the ice lake storage change time sequence. The optimal value is sampled or selected from the posterior distribution, that is, the optimal parameter set of the glacial hydrological physical model of the target basin is determined, so as to ensure that the glacial meltwater quantity and the runoff process simulated by the model are consistent with the physical fluctuation law of the actual ice lake storage. Finally, the glacial hydrological physical model based on the optimal parameter set is run, and the primary simulation runoff sequence is obtained. The technical scheme provided in the embodiments of the present application changes the limitation that the remote sensing data is only used for posterior verification in the past, and directly introduces the ice lake storage change time sequence obtained by remote sensing inversion into the parameter optimization link of the glacial hydrological physical model, constructs a likelihood function, and uses the remote sensing data to correct the intermediate state variable of the glacial hydrological physical model in real time. This 'process constraint' ensures that the runoff law of the model still conforms to the physical fluctuation characteristics of the ice lake storage even in the case of lacking ground measured runoff, thereby greatly improving the physical reliability of the simulation result.

[0049] In step S140, based on the primary simulation runoff sequence of the target basin, the ice lake storage simulation time sequence and the environmental characteristic sequence, the deviation correction is performed on the pre-trained correction model, and the reconstructed runoff sequence of the target basin is generated.

[0050] In the embodiments of the present application, the correction model based on deep learning is constructed and coupled at the output end of the glacial hydrological physical model to capture the nonlinear residual that cannot be simulated by the glacial hydrological physical model. The correction model is a long short-term memory network model or a time convolution network model. The primary simulation runoff reference sequence, the ice lake storage simulation time reference sequence and the environmental characteristic reference sequence (including precipitation, air temperature, evapotranspiration, etc.) are used as cooperative inputs, the correction model is pre-trained on the reference basin which has measured runoff data and has similar characteristics to the target basin, so that the correction model learns the residual sequence between the primary simulation runoff reference sequence and the corresponding measured runoff sequence of the reference basin.

[0051] Further, in the target basin, the deviation correction is performed on the primary simulation runoff sequence by the pre-trained correction model, and the reconstructed runoff sequence of the target basin is generated. Specifically, the primary simulation runoff sequence of the target basin, the ice lake storage simulation time sequence and the environmental characteristic sequence are input into the pre-trained correction model, and the predicted residual sequence corresponding to the primary simulation runoff sequence of the target basin is output. The predicted residual sequence and the primary simulation runoff sequence are added to obtain the high time resolution reconstructed runoff sequence of the target basin. The calculation formula of the reconstructed runoff sequence is represented by the following formula:

[0052] ;

[0053] in which, denotes reconstructed runoff at the time instant; denotes primary simulated runoff at the time instant; denotes predicted residual quantity at the time instant. And, diagnostic information is synchronously generated, including the storage contribution of the glacial lake, the contribution degree of the extreme drainage event to the peak of the flow, and the uncertainty confidence interval of the reconstructed runoff sequence, which is determined by Monte Carlo simulation on the posterior distribution of the model parameter set, thereby providing a scientific basis for glacial hydrological evolution analysis and downstream water safety management.

[0054] The technical scheme provided by the embodiment of the application constructs a collaborative computing paradigm of model prior driving and deep learning bias correction, rather than a single black box prediction, so as to exert the mechanism extrapolation capability of the glacial hydrological physical model in the ungauged area. After generating the preliminary estimation of the runoff sequence, the deep learning network model is used as a residual learner, the output of the glacial hydrological physical model and the multi-source environmental variables are used as feature inputs, the nonlinear system error that is difficult to be described by the glacial hydrological physical model is captured, through this coupling mode, the dependence of deep learning on massive samples is overcome, and the generalization error in the parameter determination process of the glacial hydrological physical model is made up, so that the unification of high precision and high interpretability is realized.

[0055] The technical scheme provided by the embodiment of the application does not depend on the ground measured runoff data, solves the problem of runoff reconstruction in the ungauged or little gauged glacial area by integrating the published remote sensing data, meteorological reanalysis forcing data and through the cross-basin parameter transfer technology, and realizes the continuous runoff sequence reconstruction under the condition of complex geographical environment and lack of hydrological station support, thereby providing a feasible engineering method for water resource investigation in high-altitude uninhabited area, cross-border river risk assessment and climate change impact research.

[0056] The above-mentioned runoff reconstruction method for the ungauged glacial covered area is described below in combination with a specific embodiment, however, it should be noted that the specific embodiment is only for better illustrating the application, and does not constitute an improper limitation on the application.

[0057] In a specific embodiment, the selected research period is from 2015 to 2024, and the selected application object is the Kelerqing River Basin in the Karakoram Mountains. The basin is located in a high-altitude area with concentrated distribution of glaciers, and there are multiple glacial lakes in the basin, among which the Kyaigil Glacier Lake and the Telamukalili Glacier Lake are representative. The basin has high glacier coverage and complex terrain, lacks long-term continuous and stable measured runoff data, and belongs to a typical little-data glacier-covered basin, which is suitable for verifying the applicability and implementability of the runoff reconstruction method provided in the application under the condition of missing data. First, multi-source heterogeneous data is obtained, including digital elevation model, glacier distribution data, meteorological reanalysis forcing data, and optical and radar remote sensing images. Specifically, a 30-meter resolution digital elevation model is used for basin terrain analysis and construction of the area-water level-storage relationship of the glacial lake; the global glacier catalog is used to obtain the glacier boundary and geometric characteristics of the Kelerqing River Basin, i.e., the glacier distribution data; ERA5-Land reanalysis data is used to obtain daily air temperature, precipitation and other meteorological variables, and the underlying surface is corrected according to the elevation difference to obtain meteorological reanalysis forcing data; Landsat-8 / 9 and Sentinel-1 / 2 remote sensing image sequences from 2015 to 2024 are selected, and long-term monitoring of the glacial lakes in the basin is performed. Further, the remote sensing images are processed by an automatic water body recognition algorithm to extract the time series of the lake area of the Kyaigil Glacier Lake and the Telamukalili Glacier Lake, and the area-water level-storage relationship curve of the glacial lake is constructed by combining the digital elevation model to obtain the time series of the glacial lake storage and the time series of the glacial lake storage change. For example, part of the meteorological reanalysis forcing data of the Kelerqing River Basin in the Karakoram Mountains in June 2024 is shown in Table 1:

[0058] Table 1 (Meteorological reanalysis forcing data table)

[0059] ;

[0060] Further, the glacier hydrological model GSM-Socont is used for runoff simulation, and the corrected air temperature, precipitation and glacier distribution data are input to simulate the glacier meltwater, runoff yield and confluence process. An important point is that in this embodiment, the model parameters do not depend on the measured runoff data at the mountain pass for calibration. For the Kyaigil Glacier Lake and the Telamukalili Glacier Lake, an equivalent reservoir module is introduced into the corresponding glacier hydrological physical model, the dynamic storage and drainage process of the glacial lake is described by a mass balance equation, and the total outflow of the glacial lake is described by a conditional trigger type segmented function, which respectively depicts the drainage process under normal and extreme conditions, so that the model can reflect the regulation and storage effect of the glacial lake on the runoff and the influence of sudden drainage events on the downstream flow.

[0061] Further, based on the terrain, glacier coverage and climate conditions of the Kuerqiang River Basin, a basin attribute vector is constructed, and then the prior distribution of the key model parameters in the glacier hydrological physical model is obtained; the time series of the ice lake storage capacity change quantity inverted by remote sensing is taken as the state consistency constraint condition, introduced into the parameter optimization process, and a likelihood function based on the ice lake storage capacity deviation is constructed to perform Bayesian updating on the model parameters. Through this process, the key parameters such as the glacier ablation factor and the runoff coefficient are constrained, so that the simulated melt water quantity and the actual water storage change of the glacier lake remain physically consistent, thereby determining the optimal parameters, wherein the ablation threshold temperature is-6.50 degrees Celsius, the runoff coefficient is 1000, the snowmelt coefficient is 10000, the glacier recession coefficient is 15.00, and the non-ice recession coefficient is 7.00.

[0062] After completing the model construction and parameter posterior optimization, further deep learning bias correction is performed, taking the primary simulated runoff sequence, the ice lake storage capacity simulation time series and the environmental feature sequence output by the glacier hydrological physical model as inputs, and a correction model based on the long short-term memory network and the time convolution network is constructed; the correction model is fine-tuned in the Kuerqiang River Basin to obtain the final reconstructed runoff sequence. Specifically, the continuous daily-scale runoff sequence from 2015 to 2024 in the Kuerqiang River Basin is successfully reconstructed, and for example, part of the primary simulated runoff sequence, the actual observed runoff sequence and the reconstructed runoff sequence in June 2024 are shown in Table 2.

[0063] Table 2 (primary simulated runoff sequence, actual observed runoff sequence and reconstructed runoff sequence data table)

[0064] ;

[0065] The following implementation effects are obtained: firstly, the reconstructed runoff sequence can reflect obvious seasonal variation characteristics, which is consistent with the hydrological process dominated by glacial meltwater; secondly, the Kuyajier Glacier Lake and the Telamukalili Glacier Lake show the effect of regulating and storing in the peak period of meltwater, and the model simulation result shows that the runoff peak is reduced and appears time lag; thirdly, in individual years, the model successfully identifies the runoff surge process corresponding to the rapid decrease of the ice lake storage, which shows that the runoff reconstruction method provided in the application based on the remote sensing storage dynamic constraint and the cross-basin parameter migration of the Kelenghe River Basin can depict the influence of ice lake sudden drainage on runoff; finally, the runoff uncertainty confidence interval obtained by Monte Carlo simulation sampling remains within a reasonable range, which shows that under the condition of no measured runoff, the runoff reconstruction method still has stability and implementability. In combination with the above implementation effects, it can be seen that the runoff reconstruction method provided in the application for the ice-covered area without measured data can effectively operate in the ice-covered basin lacking measured hydrological data, realize continuous reconstruction of the runoff process, and has good physical consistency and application feasibility, and through the implementation verification in the Kelenghe River Basin, the technical solution provided in the application has good applicability and implementability in the high mountain cryosphere region with multiple ice lakes and lacking measured data, and can provide reliable technical support for runoff simulation, water resource assessment and disaster risk analysis in similar regions.

[0066] In summary, the method for reconstructing runoff in a glacier-covered area without actual measurement data provided by the embodiments of the present application comprises the following steps: obtaining a digital elevation model, glacier distribution data, meteorological reanalysis forcing data and optical and radar remote sensing images of a target basin; extracting an ice lake area time sequence based on the optical and radar remote sensing images, and constructing an area-water level-storage capacity relationship curve based on the ice lake area time sequence and the digital elevation model to determine an ice lake storage capacity change time sequence, so as to determine a state consistency constraint condition in an optimization process of a glacier hydrological physical model parameter; constructing the glacier hydrological physical model by using the glacier hydrological model as a simulation skeleton based on the glacier distribution data and the meteorological reanalysis forcing data, and integrating an ice lake reservoir module for simulating dynamic changes of ice lake storage capacity in the glacier hydrological physical model, which is different from a processing mode of regarding the ice lake as a static water surface or ignoring the regulating effect of the ice lake in the prior art, and the ice lake reservoir module is embedded in the glacier hydrological physical model to explicitly express the regulation and storage effect of the ice lake on runoff and the instantaneous peak discharge process of a sudden burst of the ice lake, so that the response simulation of the ice lake flood and other sudden events has high physical authenticity, thereby enabling the runoff reconstruction result to reflect the nonlinear characteristics of snow and ice water cycle under an extreme climate background; determining a prior distribution of a model parameter set in the glacier hydrological physical model through a cross-basin parameter migration technology, and constructing a likelihood function based on the ice lake storage capacity change time sequence, so as to directly introduce the ice lake storage capacity change time sequence obtained by remote sensing inversion into a parameter optimization link of the glacier hydrological physical model, and based on a state constraint mechanism of the remote sensing observed storage capacity, the limitation that the remote sensing data is only used for posterior verification in the prior art is changed, and by constructing the likelihood function, the remote sensing data is used to correct intermediate state variables of the glacier hydrological physical model in real time, and this “process constraint” ensures that the runoff generation law of the model still conforms to the physical fluctuation characteristics of the ice lake storage capacity even in the case of lacking ground actual measurement runoff, thereby greatly improving the physical reliability of the simulation result; determining an optimal parameter set based on the Bayes theorem in combination with the prior distribution and the likelihood function, running the glacier hydrological physical model based on the optimal parameter set to obtain a primary simulation runoff sequence; performing bias correction on the primary simulation runoff sequence, the ice lake storage capacity simulation time sequence and the environmental characteristic sequence of the target basin by using a pre-trained correction model to generate a reconstructed runoff sequence of the target basin, and the final runoff reconstruction result has strict physical logic support, each time point of runoff components (glacier melt water, snowmelt water, rainfall runoff and ice lake contribution) can be traced and sourced, and conforms to the mass balance principle; through the uncertainty estimation of the Bayes inference and deep learning, the model can simultaneously output the uncertainty confidence interval of the simulation result, thereby simultaneously considering the physical interpretability and uncertainty quantification, and providing an important risk reference basis for decision makers in water resource allocation and disaster assessment.The technical solution provided in this application, by introducing cross-basin parameter migration technology and remote sensing reservoir capacity dynamic constraints, breaks away from the strong dependence of traditional hydrological models on parameter calibration of runoff records, and realizes long-term continuous runoff reconstruction under the condition of missing hydrological data. Experiments show that in high-altitude uninhabited areas lacking hydrological station observations, the runoff reconstruction method provided in this application can output runoff sequences with high temporal resolution, providing reliable data support for transboundary river management and water resource assessment in remote mountainous areas.

[0067] It should be noted that, in the embodiments of this application, if the above-mentioned method for reconstructing runoff in glacier-covered areas without measured data is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0068] Correspondingly, embodiments of this application provide a computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the steps in the runoff reconstruction method for glacier-covered areas without measured data as described in any of the above embodiments. Correspondingly, embodiments of this application also provide a computer program product, which, when executed by a processor of an electronic device, is used to implement the steps in the runoff reconstruction method for glacier-covered areas without measured data as described in any of the above embodiments.

[0069] Based on the same technical concept, this application provides an electronic device for implementing a runoff reconstruction method for glacier-covered areas without measured data, as described in the above method embodiments. Figure 2 This is a schematic diagram of the hardware entity of an electronic device provided in an embodiment of this application, such as... Figure 2 As shown, the electronic device 200 includes a memory 210 and a processor 220. The memory 210 stores a computer program that can run on the processor 220. When the processor 220 executes the program, it implements the steps in the method for reconstructing runoff in glacier-covered areas without measured data as described in any embodiment of this application.

[0070] The memory 210 is configured to store instructions and applications executable by the processor 220, and can also cache data (e.g., image data, audio data, voice communication data, and video communication data) to be processed by the processor 220 and modules in the electronic device, and can be implemented by a FLASH or a Random Access Memory (RAM).

[0071] The processor 220 implements the steps of any one of the above-mentioned methods for reconstructing runoff in an ice-covered region without actual measurement data when executing a program. The processor 220 generally controls the overall operation of the electronic device 200.

[0072] The above-mentioned processor can be at least one of an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a Central Processing Unit (CPU), a controller, a microcontroller, or a microprocessor. It can be understood that the electronic device implementing the above-mentioned processor functions can also be other, and the embodiments of the present application are not specifically limited.

[0073] The computer storage medium / memory can be a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a Ferromagnetic Random Access Memory (FRAM), a Flash Memory, a magnetic surface storage, an optical disc, a Compact Disc Read-Only Memory (CD-ROM), or the like memory; or can be various electronic devices including one or any combination of the above memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, and the like.

[0074] It should be noted that the above description of the storage medium and device embodiments is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments. For technical details not disclosed in the storage medium and device embodiments of the present application, please refer to the description of the method embodiments of the present application for understanding.

[0075] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that the size of the sequence number of the above processes in various embodiments of the present application does not mean the execution order, and the execution order of the processes should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above sequence number of the embodiments of the present application is only for description, not representing the advantages and disadvantages of the embodiments.

[0076] It should be noted that, in the present document, the terms "comprising", "containing", or any other similar term are intended to encompass non-exclusive inclusions, such that a process, method, article, or apparatus that comprises a list of elements does not necessarily include those elements only, but can include other elements not expressly listed, or can include elements inherent in such process, method, article, or apparatus. Without more limitations, an element defined by the phrase "comprising a" does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0077] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The above-described device embodiments are only illustrative, for example, the division of the units is only a logical functional division, and actual implementation can have another division manner, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed components can be through some interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0078] The units described above as separate components can or can not be physically separated, and the components shown as units can or can not be physical units; they can be located in one place or distributed on multiple network units; and part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0079] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the above integrated unit can be realized in the form of hardware or in the form of hardware plus software functional unit.

[0080] Alternatively, the above integrated unit of the present application, if realized in the form of a software function module and sold or used as an independent product, can also be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing an apparatus to automatically test a line to execute all or part of the methods described in the embodiments of the present application. The foregoing storage medium includes: mobile storage devices, ROM, magnetic or optical disks, and various media that can store program codes.

[0081] The methods disclosed in the several method embodiments provided by the present application can be combined arbitrarily without conflict to obtain new method embodiments.

[0082] The features disclosed in the several method or device embodiments provided by the present application can be combined arbitrarily without conflict to obtain new method embodiments or device embodiments.

[0083] The above merely provides the implementation manners of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for reconstructing runoff in an ungauged glacierized catchment, characterized in that, The method comprises: acquiring a digital elevation model, glacial distribution data, meteorological reanalysis forcing data and optical and radar remote sensing images of a target basin; extracting an ice lake area time sequence based on the optical and radar remote sensing images, and constructing an area-water level-storage relationship curve based on the ice lake area time sequence and the digital elevation model to determine an ice lake storage variation time sequence; based on the glacial distribution data and the meteorological reanalysis forcing data, constructing a glacial hydrological physical model by using a glacial hydrological model as a simulation skeleton and integrating an ice lake reservoir module for simulating dynamic changes of ice lake storage in the glacial hydrological physical model; calculating the basin attribute similarity between the target basin and a reference basin through a basin attribute vector, establishing a parameter mapping space, performing a cross-basin parameter migration technology, and obtaining a prior distribution of a model parameter set in the glacial hydrological physical model, wherein the basin attribute vector is composed of terrain slope, glacial coverage, latitude and vegetation index; taking the ice lake storage variation time sequence as a state consistency constraint condition, constructing a likelihood function based on ice lake state observation, and the likelihood function represents a matching probability of an ice lake storage simulation time sequence simulated by the glacial hydrological physical model and the ice lake storage variation time sequence under the condition of a given model parameter set; based on Bayes' theorem, combining the prior distribution and the likelihood function, determining a posterior distribution of the model parameter set, and determining an optimal parameter set of the glacial hydrological physical model according to the posterior distribution, and the calculation formula of the posterior distribution is represented by the following formula: ； wherein represents a time series of remote sensing-inverted ice lake storage change; represents a posterior distribution; represents a likelihood function; represents a prior distribution; running a glacio-hydrological physical model based on the optimal parameter set to obtain a primary simulated runoff series; constructing a correction model based on deep learning, taking a primary simulation runoff reference sequence, an ice lake storage simulation time reference sequence and an environmental feature reference sequence as inputs, pre-training the correction model on a reference basin having measured runoff data and similar characteristics to the target basin, and learning a residual sequence between the primary simulation runoff reference sequence of the reference basin and the corresponding measured runoff sequence, wherein the correction model is a long short-term memory network model or a time convolution network model; inputting the primary simulation runoff sequence, the ice lake storage simulation time sequence and the environmental feature sequence of the target basin into the pre-trained correction model, outputting a predicted residual sequence corresponding to the primary simulation runoff sequence of the target basin, adding the predicted residual sequence and the primary simulation runoff sequence to obtain a reconstructed runoff sequence of the target basin, and synchronously generating diagnostic information, wherein the diagnostic information includes a regulation and storage contribution of the glacial lake, a contribution degree of an extreme drainage event to a peak flow value and an uncertainty confidence interval of the reconstructed runoff sequence, and the uncertainty confidence interval is determined by Monte Carlo simulation on the posterior distribution of the model parameter set.

2. The method of claim 1, wherein, The method comprises: Based on the optical and radar remote sensing images, an ice lake area time sequence of a target glacial lake in the target basin is extracted by a remote sensing image automatic extraction algorithm; Based on the ice lake area time sequence, the digital elevation model and the terrain features, an area-water level-storage relationship curve of the target glacial lake is constructed; Based on each phase ice lake area value in the ice lake area time sequence, the ice lake storage time sequence and the ice lake storage change amount time sequence of the target glacial lake are inversely determined according to the area-water level-storage relationship curve.

3. The method of claim 1, wherein, The mathematical expression of the glacial hydrological physical model is represented by the following formula: ； wherein denotes a preliminary simulated runoff at the time instant; denotes a set of functions describing hydrological physical processes of the GSM-Socont model; denotes air temperature forcing data at the time instant; denotes precipitation forcing data at the time instant; denotes the area of the glacier lake; denotes a set of model parameters to be determined, which are independent of the rating of the measured runoff data.

4. The method of claim 3, wherein, The glacial lake reservoir module simulates the dynamic change of the storage of the glacial lake by a mass balance equation, and the mass balance equation is represented by the following formula: ； wherein, represents the simulated ice lake storage value at the time instant; represents the simulated ice lake storage value at the time instant; represents the total inflow water volume of the glacial lake at the time instant; represents the total outflow water volume of the glacial lake at the time instant, the total outflow water volume being defined by a condition-triggered piecewise function for describing a regular outflow process, a spillway process and a sudden breach process of the glacial lake; represents the lake surface evaporation volume of the glacial lake at the time instant.

Images