50 results about "Drought risk" patented technology

The invention provides a regulation factor threshold effect evaluation method and device for vegetation response sudden drought characteristics, equipment and a storage medium. Relates to the technical field of drought risk prediction and influence evaluation. The method comprises the following steps: acquiring hydro meteorological data, vegetation productivity data, topographic feature data, soil characteristic data and biological variable data of a target area; constructing a drought suddenness index, extracting a drought suddenness event according to the drought suddenness index, and determining a drought suddenness feature of the drought suddenness event; extracting vegetation productivity abnormal data before and after the sudden drought event, and determining the response characteristic difference of the vegetation to sudden drought; the method comprises the following steps: determining influence factors of vegetation on sudden drought response characteristics, fusing XGBoost, a random forest regression model and an SHAP interpretability analysis method, analyzing regulatory factors of the vegetation response sudden drought characteristics, and determining a threshold effect of the regulatory factors. According to the method, the nonlinear cooperative regulation effect of multiple factors on vegetation response sudden drought characteristics can be disclosed more comprehensively.

The invention discloses an agricultural drought risk assessment method. The method comprises the following steps: dividing a target area into a plurality of assessment units according to geographic space raster data of the target area; agricultural disaster damage data of each evaluation unit is collected, and a quantitative agricultural loss rate is obtained by adopting a K-Means classification method and through moisture-yield coefficient correction; the total output value of the target area and the agricultural output value and the regional output value of each evaluation unit are extracted, an agricultural exposure degree model is constructed, and the agricultural exposure degree of the target area is obtained; constructing an agricultural drought risk assessment model by using the drought risk, the agricultural loss rate and the agricultural exposure degree, and obtaining agricultural drought risk pattern grid data of each assessment unit by using a neighborhood spatial analysis-main filtering method and a spatial difference method; and distributing the agricultural drought risk pattern grid data of each evaluation unit in a target area according to geographic space grid data of the target area to obtain an agricultural drought risk pattern of the target area. The agricultural drought risk is accurately quantified, high-vulnerability and high-risk regions are defined, and a scientific basis is provided for agricultural resource allocation optimization, disaster prevention and reduction strategy formulation and drought emergency management.

The invention relates to the technical field of water regimen forecasting, in particular to a river channel water regimen forecasting method and system based on digital twinning. The method comprises the following steps: arranging sensor nodes along a river channel and carrying out real-time monitoring on the river channel environment water regimen along the river channel to obtain real-time data of the water level along the river channel, real-time data of the flow along the river channel and real-time data of the weather along the river channel; performing digital twinborn construction and risk estimation division along the riverway to generate a riverway water regimen flood risk estimation period and a riverway water regimen drought risk estimation period; the method comprises the following steps: performing flood risk forecast processing in a river water regimen flood risk estimation period to execute corresponding river water regimen flood risk management decision-making work; and performing drought risk forecast processing in the river channel water regimen drought risk estimation time period to execute corresponding river channel water regimen drought risk management decision-making work. According to the invention, efficient and accurate water regimen prediction and decision support can be realized.

The invention relates to the technical field of water-saving irrigation, in particular to a drought risk management-based corn planting water-saving irrigation system, which comprises a crop stress state monitoring module for acquiring a corn canopy temperature and an environment temperature and calculating a temperature difference value between the canopy and the environment. According to the method, the actual stress state of the crop is quantified by acquiring the difference value between the corn canopy temperature and the environment temperature in real time, and the crop water stress index which is dynamically calculated is combined, so that the water stress identification precision and the response time efficiency are improved, and the problems of irrigation decision lag and blindness are avoided; further, the real evapotranspiration demand of crops is accurately judged by dynamically monitoring the change rate of latent heat flux, misjudgment possibly caused by single monitoring of temperature difference is avoided, and the irrigation demand judgment accuracy is ensured; meanwhile, the prices, the flows and the available amounts of different water sources are comprehensively optimized, the taking proportion of multiple water sources is configured with the lowest total irrigation cost as the standard, water source economical efficiency optimization becomes possible, and irrigation operation cost and resource consumption are effectively reduced.

The invention belongs to the technical field of drought risk prediction and vegetation influence evaluation, and particularly discloses a vegetation-to-drought resource utilization coupling response strategy evaluation method, which comprises the following steps: acquiring target data; constructing a drought index according to the target data, and extracting a drought event; extracting vegetation total productivity, vegetation net level productivity and evapotranspiration abnormal data before and after the drought event, and calculating response intensity of vegetation variables; calculating the average variation of the light energy utilization efficiency, the carbon utilization efficiency and the moisture utilization efficiency; comprehensively considering the multivariable resistance-recovery state of the vegetation physiological process and the regulating variable of the resource utilization efficiency of the vegetation in response to drought stress, and analyzing the drought event resistance-recovery response capability of the vegetation. The method can more accurately and objectively understand the physiological mechanism difference of the vegetation to the drought stress response.

The invention discloses a soil drought risk early warning system and early warning method based on machine learning, and belongs to the field of soil drought early warning, and the system comprises a data collection module, a data restoration and fusion module, a space-time drought coupling prediction model construction module, and a real-time early warning and model updating module. Wherein the data acquisition module is used for acquiring wind cloud meteorological data and MODIS optical image data; the data restoration and fusion module is used for generating an all-weather soil humidity chart; the space-time drought coupling prediction model construction module is used for constructing a space-time drought coupling prediction model; and the real-time early warning and model updating module is used for obtaining the dynamic drought index and dynamically updating the dynamic drought index. By adopting the soil drought risk early warning system and early warning method based on machine learning, high-precision drought risk prediction and decision support are realized through multi-source data fusion, a space-time coupling prediction model and a dynamic grading early warning mechanism.

The invention discloses a risk dynamic analysis method for a vegetation phenological period in an arid region, and the method comprises the steps: constructing a spring phenological model and an autumn phenological model in different partitions according to the partitions of different growth season initial periods SOS and growth season final periods EOS of plants in the arid region; respectively calculating vegetation growth risk coefficients of different growth season initial stage SOS partitions and growth season final stage EOS partitions; carrying out weight fusion on the vegetation growth risk coefficients to obtain a long-period vegetation growth risk comprehensive coefficient F in the arid region; calculating an ecological groundwater drought risk coefficient; calculating a vegetation risk dynamic coefficient of the arid region; and evaluating the growth risk of the vegetation in the arid region in the phenological period. According to the method, the drought risk influence coefficient is comprehensively calculated by utilizing representative feature data corresponding to different subareas and phenological areas, and comprehensive analysis of vegetation risk assessment in the arid area is realized by combining the ecological underground water drought risk coefficient on the annual scale.

The invention discloses a water-electricity coupling system toughness evaluation method considering drought risk evolution, and relates to the technical field of water-electricity coupling systems. Comprising the following steps: S1, modeling a water-electricity coupling system network; s2, modeling based on seepage water-electricity coupling network drought risk propagation; and S3, carrying out system simulation and toughness evaluation on the water-electricity coupling system network on the basis of the two models. According to the method, a heterogeneous weighted double-layer network model is constructed, a cross-scale dynamic modeling method is developed, a seepage theory and physical flow are dynamically combined, a two-way seepage equation of disaster cascade propagation is established, a toughness evaluation framework based on the seepage theory is provided, and laws of different coupling effects of a water-electricity system are disclosed; and a scientific basis can be provided for risk assessment, toughness improvement and emergency management strategy formulation of the key infrastructure.

The invention provides an agricultural drought risk quantitative evaluation method and system considering a drought process, and relates to the technical field of agriculture and climate risk management. According to the method, drought accumulation duration and intensity of all grid points in a research area are determined by using preset drought indexes, a multi-dimensional drought feature joint probability model of a Copula function is constructed based on the data, drought grades and conversion thresholds are divided, and the farmland drought area is counted in combination with land utilization data. By establishing the Markov decision process and optimizing the drought management policy, the prediction precision of the drought risk is effectively improved, the agricultural loss is remarkably reduced, and the resource utilization efficiency is improved. The method provides a reliable basis for scientific decision making.

The invention relates to the field of climate change influence assessment, and discloses a method for quantizing contribution of natural internal variability to drought intensity based on a circulation similarity method, and the method comprises the steps: data collection; evaluating the drought strength based on the difference between the rainfall distance and the potential evapotranspiration distance, and quantifying the relative contribution of rainfall deficit and evapotranspiration demand to the drought strength; based on a circulation similarity method, a regional large-scale atmospheric circulation background is obtained, and contribution of a natural internal variation rate to drought intensity, rainfall deficit and evapotranspiration requirements is evaluated; estimating long-term evolution trends of drought intensity, rainfall deficit and evapotranspiration demands under different carbon emission scenes in the future; the method has the advantages that the natural internal influence mechanism of drought intensity change is defined, the relative contribution of historical and future rainfall deficit and evapotranspiration requirements to the drought intensity is quantified, the reliability of future drought trend estimation is enhanced, theoretical support is provided for making disaster prevention and reduction policies, and the accuracy of drought risk prediction is improved.

The application provides a drought risk identification and early warning method based on multi-source satellite fusion NDVI and machine learning, acquires NDVI time series data corresponding to a research area, and extracts feature information of a disaster-bearing body corresponding to the research area based on the NDVI time series data; acquires water resource supply data corresponding to the research area, and extracts water resource supply features corresponding to the water resource supply data; through a machine learning model obtained by pre-training, a disaster-bearing body NDVI prediction value of the research area at a future time is predicted based on the disaster-bearing body feature information and the water resource supply features; the disaster-bearing body NDVI prediction value is used for identifying the drought risk of the research area in combination with precipitation forecast data, or dynamically adjusting the anti-drought irrigation strategy data. The application can significantly improve the accuracy of drought risk prediction and effectively improve the dynamic adjustment effect of the anti-drought irrigation strategy.

The present invention relates to a method for dynamically assessing the comprehensive risk of drought disasters. Taking the comprehensive risk of drought disasters as the assessment object, comprehensive risk assessment factors are selected from four aspects: disaster-causing factors, disaster-forming environments, disaster-bearing bodies, and disaster prevention and mitigation capabilities of drought disasters, and a dynamic assessment mathematical model for the comprehensive risk of drought disasters is constructed to dynamically assess the comprehensive risk of regional drought disasters in terms of time and space. The method of the present invention takes into account the evolution process of each risk factor in time and space, can characterize the comprehensive risk characteristics of drought disasters in different regions at different times, and to a certain extent makes up for the deficiency of the traditional method that only relies on drought frequency to represent the risk of disaster-causing factors of drought. At the same time, the method of the present invention also considers the influence of factors such as disaster-forming environments, disaster-bearing bodies, and disaster prevention and mitigation capabilities on risks, realizing the leap from drought risk to drought disaster risk.

The invention relates to the field of hydrological disaster risk assessment, and discloses a man-made signal identification and future estimation method for hydrological drought risks, and the method comprises the steps: collecting data; hydrological drought identification; hydrological drought occurrence probability trend quantification; quantification of contribution of human climate change and human water taking and land management change; human activity signal detection of hydrological drought occurrence probability; hydrological drought attribution risk score quantification; identifying human activity signals in hydrological drought based on a signal-to-noise ratio criterion; the uncertainty source of the future hydrological drought occurrence probability trend is quantified; according to the method, the global hydrological drought occurrence risk change in historical and future periods under the influence of human activities is evaluated, and the human activity signals in the global hydrological drought occurrence risk change are identified, so that the understanding of the global hydrological drought risk change and influence mechanism is enhanced, and a scientific basis is provided for making a hydrological drought disaster prevention and reduction decision.

The invention provides an agricultural drought risk assessment method based on machine learning, and relates to the technical field of agricultural big data. Comprising the steps of performing time sequence segmentation on multi-source monitoring data based on a crop phenological period standard, and constructing a time-space coupling data set; determining a moisture sensitivity coefficient by utilizing development key nodes and a physiological water demand threshold value, generating a dynamic attention weight matrix, and extracting a comprehensive drought feature vector set through a long short-term memory network; and then, constructing a nonlinear vulnerability fitting curve of the yield response mapping space by adopting a support vector regression algorithm, and constructing a reinforcement learning simulation environment according to the nonlinear vulnerability fitting curve. And constructing a state space vector, performing iterative optimization training of a water resource allocation action in a simulated environment by using a deep Q network algorithm to obtain an optimal risk management and control strategy network, and outputting an agricultural drought risk assessment result and a water resource allocation instruction. According to the invention, accurate quantification of drought risks and global dynamic optimal configuration of water resources are realized.

The invention relates to the technical field of risk assessment, in particular to a grassland ecosystem comprehensive drought risk assessment system based on multi-source data. The risk feature highlighting characterization value is calculated by determining the feature highlighting factors of the different risk areas of the grassland, so that the feature highlighting sorting sequence of the risk areas of the grassland is rapidly determined, the precision of grassland comprehensive drought risk assessment is improved, and meanwhile, the risk assessment accuracy is improved. According to the grassland actual data uploaded by the user side and the drought risk index information corresponding to the grassland ecological system, the maximum feature highlighting characterization value of each risk area of the grassland can be accurately obtained, and the accuracy of comprehensive drought risk assessment of the grassland ecological area is improved.

The invention discloses an intelligent drought disaster risk estimation system which comprises a data acquisition module, a feature dimension reduction module, a risk sample selection module, a risk interval loss function design module, a drought risk model design module, a hyper-parameter optimization module and a drought disaster risk estimation module. The invention belongs to the field of risk estimation, and particularly relates to an intelligent drought disaster risk estimation system, which introduces a dynamic risk sample selection mechanism to improve the drought risk capture capability. The risk disorder loss is designed in a targeted manner, and the critical risk missed judgment probability is reduced; designing a dynamic threshold to realize dynamic self-adaption of a sample screening threshold; the dynamic evolution trend of the drought risk is incorporated into the loss function design, and the early warning capability is improved; the gradient amplitude is dynamically adjusted in combination with drought risk prediction confidence, and an unknown risk mode is effectively explored; and difference traction search and focusing traction deviation optimization are performed on the hyper-parameters, so that the final drought disaster risk estimation effect is improved.

The present invention discloses an agricultural drought risk assessment method, which comprises the following steps: dividing a target area into a plurality of assessment units according to its geospatial raster data; collecting agricultural disaster loss data for each assessment unit, using the K-Means classification method and correcting it by the moisture-yield coefficient to obtain a quantitative agricultural loss rate; extracting the total output value of the target area and the agricultural output value and regional output value of each assessment unit, constructing an agricultural exposure model, and obtaining the agricultural exposure of the target area; constructing an agricultural drought risk assessment model using drought hazard, the agricultural loss rate, and the agricultural exposure, and using the neighborhood spatial analysis-primary filtering method and spatial difference method to obtain agricultural drought risk pattern raster data for each assessment unit; and distributing the agricultural drought risk pattern raster data for each assessment unit over the target area according to its geospatial raster data to obtain the agricultural drought risk pattern of the target area. By accurately quantifying agricultural drought risk and identifying high-vulnerability and high-risk areas, the present invention provides a scientific basis for optimizing agricultural resource allocation, formulating disaster prevention and mitigation strategies, and managing drought emergencies.

The invention discloses an ecological groundwater drought risk dynamic simulation method, which comprises the following steps of: constructing a short-term driving force layer of a groundwater drought influence ecological effect, and fusing a hysteresis effect and a cumulative effect to obtain a calculation model of a short-term driving force coefficient; constructing a long-term driving force layer of the underground water drought influencing the ecological effect, and obtaining an optimal calculation model of a long-term driving force coefficient; performing weight fusion on the calculation model of the short-term driving force coefficient and the optimal calculation model of the long-term driving force coefficient, and calculating a comprehensive driving force coefficient to obtain a hierarchical memory-threshold response ecological groundwater drought risk model; and realizing dynamic process simulation of the ecological drought risk by using the hierarchical memory-threshold response ecological groundwater drought risk model. According to the method, three types of ecological response mechanisms of lagging, accumulation and leaving are integrated, and conversion of ecological drought risk assessment from a static index to dynamic process simulation is realized.

The present invention discloses an intelligent drought risk management method and system based on disaster attribution analysis, and the method includes data collection, data enhancement, disaster attribution analysis, disaster dynamic risk assessment and intelligent drought risk management. The present invention relates to the field of drought risk management, and adopts an integrated method of intelligent drought risk management combining disaster attribution analysis and disaster dynamic risk assessment. By performing dynamic analysis of drought disasters from the perspective of attribution, and combining multiple feature optimization models to perform multi-factor feature mining, the complexity processing capability and real-time accuracy of intelligent drought risk management are improved, and the fusion performance of multi-domain data is also improved; an improved support vector machine model combining expert model and graph neural network feature optimization is used to perform disaster attribution analysis; and a gated recurrent unit combining generative adversarial network and graph convolution network feature deepening is used to perform dynamic disaster risk assessment.

The application provides a farmland water content collaborative diagnosis method and system based on multi-source remote sensing, and relates to the technical field of agricultural informatization and satellite remote sensing monitoring. The scheme comprises the following steps: obtaining target farmland water content condition data and identifying a risk area, triggering encrypted observation to obtain refined data; sorting the risk area based on the refined data and determining a monitoring scheme; adjusting the collection parameters according to the energy consumption state of the monitoring equipment during the execution process to update the water content data; fusing the updated data and synchronous remote sensing data to generate enhanced diagnosis data; and finally evaluating the water content condition and resource efficiency based on the enhanced diagnosis data. The application can realize dynamic collaboration and resource optimization scheduling of a multi-source observation platform, give consideration to large-scale coverage and local refined observation, and improve the accuracy and timeliness of drought risk diagnosis.

The invention provides a drought risk assessment method for an irrigated farmland, and belongs to the technical field of agricultural drought monitoring and risk analysis. Comprising the following steps: S1, acquiring basic data; s3, constructing an irrigation capacity factor ACz; s4, extending the irrigation amount time sequence; s5, drought risk index calculation; and S6, carrying out risk space distribution and result verification. According to the method, the irrigation capability factor is introduced, the farmland drought risk model considering the irrigation effect is constructed, the defect that an existing model does not consider the irrigation regulation effect is overcome, and precise quantification of the irrigation farmland drought risk is achieved.

The present application relates to a kind of multi-region meteorological drought and hydrological drought response relationship identification modeling method, comprising the following steps: drought index construction and time sequence acquisition, obtain the comparable drought index sequence;Drought event identification and feature extraction, extract the duration D of each drought and drought amplitude M;Regional response relationship modeling, parameter fitting is carried out to historical drought event sample by least square method;Space joint drought modeling and estimation, the joint probability distribution model of hydrological drought amplitude between two adjacent regions is established using Copula function.The present application can be used for drought risk early warning and cross-region hydrological drought estimation, especially in the basin with serious data missing or significant human intervention has strong applicability, can provide certain scientific basis and technical support for hydrological drought management and control strategy formulation.

The application discloses a kind of water-electricity coupling system resilience assessment method considering drought risk evolution, it is related to water-electricity coupling system technical field. Including: S1, water-electricity coupling system network modeling;S2, water-electricity coupling network drought risk propagation modeling based on seepage;S3, water-electricity coupling system network is simulated and resilience is evaluated based on the above two models basis system.This application constructs heterogeneous weighted double-layer network model, develops cross-scale dynamics modeling method, combines seepage theory and physical flow dynamics, establishes bidirectional seepage equation of disaster cascade propagation, proposes resilience assessment framework based on seepage theory and reveals the law of different coupling effects of water-electricity system, which can provide scientific basis for risk assessment, resilience improvement and emergency management strategy formulation of key infrastructure.

The invention provides a monthly-scale meteorological drought risk quantitative estimation method based on climate mode forecasting, and the method comprises the steps: determining an evaluation time period according to the occurrence condition of a drought process in an early-stage rainfall time period of an evaluation region, and carrying out the evaluation of the early-stage rainfall time period through the historical day-by-day rainfall observation data of a meteorological station in the evaluation time period for many years. Counting the number of duration days without effective precipitation and the average percentage of the accumulated precipitation in the evaluation period of each year, calculating a meteorological drought risk index of each year, and determining a grading threshold value of a meteorological drought risk grade by adopting a percentile method; and calculating a meteorological drought risk index of an estimated month and determining a risk level through the estimated data of the evaluation time period. According to the method, the non-effective rainfall duration day index considering the characteristic of less rainfall duration and the accumulated rainfall average percentage index reflecting the characteristic of less rainfall total amount are selected as estimation indexes, the complexity of an existing monthly-scale meteorological drought estimation method is overcome, and the estimation operability and reliability are improved.

The present application provides a kind of drought forecasting method and system based on artificial intelligence and atmospheric circulation mechanism, comprising: S1, obtaining alternative precipitation runoff influence factor;S2, the importance of alternative precipitation runoff influence factor is sorted and the final precipitation runoff influence factor and precipitation runoff state number k value are determined;S3, with the Bayesian-NHMM model of minimum precipitation runoff prediction error as the best precipitation runoff prediction model, obtain precipitation and runoff prediction result;S4, realize the classification of land water storage forecast factor;S5, train machine learning model, and use the trained machine learning model as land water storage forecast model;S6, realize drought forecasting and early warning based on land water storage forecast model.The present application makes full use of the advantages of Bayesian-NHMM model and current machine learning model, makes up for the problem of low accuracy of traditional drought forecasting, and can deduce future drought risk and carry out drought risk early warning.

The invention discloses a drought monitoring method and system based on coupling of crop types and multi-dimensional drought characteristics. The method comprises the following steps: acquiring and preprocessing multi-source remote sensing data; based on high-resolution remote sensing data, farmland plots are extracted in combination with terrain and road network data; combining optical remote sensing data to extract crop time sequence characteristics of the farmland plot; inputting the crop time sequence characteristics into a pre-trained crop identification model; constructing a specific three-dimensional drought index model of different types of crops based on the spatial distribution of various types of crops; and according to the specific three-dimensional drought index model, fusing a soil moisture dynamic threshold value and remote sensing drought loss inversion data to generate a crop specific drought risk grading map. According to the method, different crop types can be accurately distinguished, the difference of different crops in the aspects of water demand characteristics and drought response is fully reflected, and refined monitoring and evaluation of drought conditions of different crops are realized.

The present application relates to the field of natural disaster risk assessment and water resource informatization, and discloses a drought risk assessment method based on soil humidity, which comprises the following steps: obtaining multi-source data of a region to be evaluated and preprocessing the data; constructing unified input features based on the preprocessed multi-source data; establishing a spatial neighborhood graph and a spatial weight matrix according to the position information of each monitoring site; inputting the unified input features, the spatial neighborhood graph and the spatial weight matrix into a trained soil humidity spatio-temporal prediction model to output soil humidity prediction values; obtaining a soil humidity field based on the soil humidity prediction values and the position information of each monitoring site, and generating soil grid data according to the soil humidity field; calculating a risk index according to the soil grid data and risk assessment auxiliary data; and outputting a drought risk spatial distribution map and risk grade classification results of the region to be evaluated based on the risk index, which can effectively improve the spatial continuity and prediction accuracy of drought assessment.

The invention discloses a composite high-temperature drought dynamic early warning method based on multi-source data fusion, and belongs to the technical field of big data analysis and mining, and the method comprises the following steps: collecting multi-source observation data in a region, forming a unified grid data matrix, carrying out the marking and interpolation processing of missing data, generating a weight matrix, recognizing potential abnormality, and carrying out the early warning of the high-temperature drought. The composite high-temperature drought dynamic early-warning method based on multi-source data fusion comprises the following steps of: acquiring data, dividing grid abnormal grades, adjusting data risk fusion weight factors, analyzing regional composite risk indexes, and determining regional composite high-temperature drought risk early-warning grades, and realizes unified acquisition, space alignment and quality evaluation of multi-source data; the method is advantaged in that the data reliability is improved, grid risk calculation is carried out, dynamic area risk early warning is realized through combination of moving average and threshold value determination, the composite effect of high temperature and drought can be simultaneously considered, the risk level is quantified, and early warning accuracy, timeliness and reliability are improved.

The invention provides an estuary area composite salt and drought multivariable risk assessment method and system, and relates to the technical field of hydrology and water resources. Aiming at the lack of recurrence period evaluation means capable of simultaneously quantifying the drought and salt tide coupling effect and accurately depicting multivariable features and joint probability of a composite event during estuary composite salt drought risk prediction in the prior art, a standardized composite salt drought index is constructed, and event feature variables are identified based on a run-length theory; and a Copula function is used to establish joint probability distribution of duration and severity, and finally a technical path of a co-occurrence recurrence period and a joint recurrence period is calculated, so that the accuracy and comprehensiveness of a traditional single variable recurrence period in evaluating the multi-dimensional risk of the composite event are improved. According to the method, through multi-distribution optimization, experience fitting and multi-scale verification, the precision and reliability of composite salt and drought event identification are improved, and a hierarchical and quantifiable risk decision basis is provided for estuary region water resource management, disaster early warning and engineering fortification.

The application discloses a soil drought risk early warning system and method based on machine learning, belongs to the field of soil drought early warning, and comprises a data acquisition module, a data repair and fusion module, a space-time drought coupling prediction model construction module and a real-time early warning and model updating module; wherein the data acquisition module is used for acquiring wind cloud weather data and MODIS optical image data; the data repair and fusion module is used for generating all-weather soil humidity maps; the space-time drought coupling prediction model construction module is used for constructing a space-time drought coupling prediction model; and the real-time early warning and model updating module is used for obtaining a dynamic drought index and dynamically updating. The soil drought risk early warning system and method based on machine learning have the advantages that through multi-source data fusion, a space-time coupling prediction model and a dynamic grading early warning mechanism, high-precision drought risk prediction and decision support are realized.