CN116796649A - SPEI coarse resolution data space downscaling method and device based on machine learning - Google Patents

SPEI coarse resolution data space downscaling method and device based on machine learning Download PDF

Info

Publication number
CN116796649A
CN116796649A CN202310895388.5A CN202310895388A CN116796649A CN 116796649 A CN116796649 A CN 116796649A CN 202310895388 A CN202310895388 A CN 202310895388A CN 116796649 A CN116796649 A CN 116796649A
Authority
CN
China
Prior art keywords
spei
downscaling
data
model
resolution
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310895388.5A
Other languages
Chinese (zh)
Inventor
贺倩
汪明
刘凯
李博浩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Normal University
Original Assignee
Beijing Normal University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Normal University filed Critical Beijing Normal University
Priority to CN202310895388.5A priority Critical patent/CN116796649A/en
Publication of CN116796649A publication Critical patent/CN116796649A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Geometry (AREA)
  • Computer Hardware Design (AREA)
  • Probability & Statistics with Applications (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Image Analysis (AREA)

Abstract

The application relates to a SPEI coarse resolution data space downscaling method and device based on machine learning, wherein the method comprises the following steps: step 1: acquiring original coarse resolution SPEI raster data of a long-time sequence in a research area; step 2: acquiring downscaling factor data in a research area; step 3: resampling the downscaling factor to the same spatial resolution as the original SPEI; step 4: constructing sample data of all pixels, including SPEI values and downscaling factor values, and dividing the sample data into a test set and a training set; step 5: constructing a space downscaling model based on the training set, constructing the downscaling model by utilizing Gaussian process regression of a machine learning algorithm, and evaluating the precision of the model by using a test set; step 6: inputting the high-resolution downscaling factor raster data in the step 2 into a trained model to generate a long-time sequence high-spatial-resolution downscaling SPEI data product.

Description

SPEI coarse resolution data space downscaling method and device based on machine learning
Technical Field
The application relates to the technical field of agricultural information, in particular to a machine learning-based Standardized Precipitation Evapotranspiration Index (SPEI) coarse resolution data space downscaling method and device.
Background
Global warming will result in more frequent, more severe drought, with severe impact on regional water resource supplies and land ecosystems. China is one of the countries that is often affected by drought. In recent decades, drought in China is frequently affected by climate change, and serious consequences are brought to ecology, agriculture and social economy. Therefore, monitoring and alleviating drought is of paramount importance.
Drought types typically include meteorological drought, agricultural drought, hydrographic drought, and socioeconomic drought. Drought index is the basis for quantification and description of drought, and is an effective tool for drought early warning, monitoring and assessment. The weather drought index standardized precipitation evapotranspiration index (Standardized Precipitation Evapotranspiration Index, SPEI) is one of the most widely used weather drought indexes at present. The existing SPEI data products, although time series are long enough (1901-2021), are relatively coarse in spatial resolution, being 0.5 °. The method is suitable for global research, and the spatial resolution is rough for evaluating the drought space-time variation of China. The spatial scale of drought can range from a few kilometers to the whole area, so that evaluation on the high-resolution grid scale of drought in China is necessary. The lack of a long-time sequence high-spatial resolution drought data set is inconvenient to quantify long-term drought in China, and complex space-time characteristics of the drought are difficult to master.
At present, a high-resolution space continuous SPEI data set is obtained by interpolation through a meteorological site SPEI, however, the method is limited by site non-uniform distribution and lack of meteorological site data in certain areas, and the interpolation result often has larger space uncertainty. Geographic weighted regression is a statistical downscaling method commonly used at present, but has limited applicability in non-stationary and non-linear problems.
Therefore, development of a technical scheme aiming at the defects is urgently needed at present, an effective SPEI space downscaling model is constructed, spatial resolution of SPEI long-time sequence data in China is improved, drought space-time patterns and occurrence rules in China are mastered more finely by assistance, and reliability of drought risk assessment in China is improved.
Disclosure of Invention
In order to achieve the above purpose, the application provides a SPEI space downscaling method based on machine learning, which improves the space resolution of SPEI data products, meets the application requirements of drought monitoring and evaluation and water resource management of fine scale, and solves the defect of low space resolution of the existing SPEI data products.
More specifically, the present application provides a method for spatial downscaling of SPEI coarse resolution data based on machine learning, comprising:
step 1: acquiring original coarse resolution SPEI raster data of a long-time sequence in a research area;
step 2: acquiring downscaling factor data in a research area, wherein the downscaling factor data comprises high-resolution climate, topography and geographic position factor grid data, and the climate factors correspond to long-time sequences of SPEI;
step 3: resampling the downscaling factor to the same spatial resolution as the original SPEI;
step 4: constructing sample data of all pixels, taking the pixels as a statistical unit, extracting SPEI values and downscaling factor values of all pixels, taking all pixel values as samples, and dividing the sample data into a test set and a training set;
step 5: constructing a space downscaling model based on a training set, taking a SPEI value as a dependent variable, taking a downscaling factor as an independent variable, constructing the downscaling model by utilizing Gaussian Process Regression (GPR) of a machine learning algorithm, determining model parameters by utilizing Bayesian optimization, and evaluating the precision of the model by using a testing set; and
step 6: inputting the high-resolution downscaling factor raster data in the step 2 into a trained model to generate a long-time sequence high-spatial-resolution downscaling SPEI data product.
According to an embodiment of the application, wherein the method further comprises step 7: the original SPEI of step 1 was used for cross-validation and downscaled SPEI data products were validated by studying drought events in the regional history.
According to an embodiment of the application, wherein in step 2, the climate factors comprise a month average temperature, a month maximum temperature, a month minimum temperature and a month precipitation; the topography factors include elevation, grade and slope; the geographic location factor includes longitude and latitude. For example, month precipitation, month average temperature, maximum temperature and minimum temperature data may be obtained from the national Qinghai-Tibet plateau science data center (https:// data.tpdc.ac.cn /), with a spatial resolution of 0.0083333 ° (about 1 km) for a period of from 1 month in 1901 to 12 months in 2021. Other high-resolution weather data products can be selected by the person skilled in the art according to specific conditions; the terrain factors can be calculated by using a Digital Elevation Model (DEM), the DEM data can be SRTM DEM Version4, and other DEM data products can be selected by a person skilled in the art according to specific situations.
According to an embodiment of the present application, wherein in step 1, the coarse resolution is 0.5 ° x 0.5 °, e.g. the original SPEI database is constructed by the SPEI's provider Sergio m.vicenter-Serrano, with a spatial resolution of 0.5 °, and a spatial resolution of 0.0083 ° with a downscaling factor.
According to an embodiment of the present application, in step 5, the super parameters of the Gaussian Process Regression (GPR) model are obtained using bayesian optimization, and the downscaling model is respectively constructed based on monthly data.
The GPR method is a non-parametric probability model based on a kernel function, and the kernel theory is a Bayesian method. GPR performs well on small datasets, and is generally not affected by fitting. GPR with gaussian noise can be expressed as follows:
y＝f(x)+ε
where ε is the noise (ε -N (0, σ) that follows a Gaussian distribution 2 ) X represents input and y represents output. f (x) is a regression function, which can be expressed as:
f(x)～GP(m(x),k(x,x′))
where m (x) is the mean function and k (x, x') is the covariance function.
According to an embodiment of the application, wherein step 5 further comprises evaluating the accuracy of the model, determining coefficients (R 2 ) Root Mean Square Error (RMSE), mean Absolute Error (MAE)The model is evaluated.
Precision evaluation index R 2 The calculation formula of RMSE and MAE is as follows:
wherein: o (O) i Actual value for the ith data point; e (E) i Modeling values for the ith site; n represents the number of samples;is the average of the actual values.
According to an embodiment of the present application, step 7 includes determining that the downscaled data is not degraded in accuracy by calculating a correlation of the original SPEI and the downscaled SPEI, and evaluating the effectiveness of the downscaled SPEI in identifying drought using historical drought events.
According to an embodiment of the application, wherein in step 4, the samples are divided into 70% training sets to train the model, 30% test sets to evaluate the model accuracy.
It should be appreciated that the SPEI data may include multiple time scales, such as 1 month SPEI,3 month SPEI,6 month SPEI, and the like.
The application also provides a SPEI coarse resolution data space downscaling method device based on machine learning, which comprises the following steps:
the coarse resolution data acquisition module is used for acquiring original coarse resolution SPEI raster data of a long-time sequence in the research area;
the downscaling factor data acquisition module is used for acquiring downscaling factor data in the research area;
a downscaling factor resampling module for resampling the downscaling factor to the same spatial resolution as the original SPEI;
the sample data construction module is used for constructing sample data of all pixels and dividing the sample data into a test set and a training set;
the space downscaling model construction module is used for constructing a space downscaling machine learning model based on the training set and evaluating the precision of the model based on the test set; and
the downscaling SPEI data product forming module is used for inputting the downscaling factor raster data with high resolution into the trained model to generate a downscaling SPEI data product with long-time sequence and high spatial resolution.
According to an embodiment of the application, the apparatus further comprises a verification module for cross-verifying using the original SPEI of step 1 and verifying the downscaled SPEI data product by studying drought events of the regional history.
The application also provides an electronic device comprising a memory and one or more processors;
the memory is used for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the methods described herein.
The application provides a coarse resolution SPEI space downscaling method based on a machine learning GPR algorithm, factors of a constructed downscaling model are easy to obtain, site actual measurement data is not needed, and the model precision is high. The method can generate SPEI data with high spatial resolution, and the generated data can provide data support for drought monitoring and evaluation with small range and high precision.
Drawings
FIG. 1 is a flowchart of a SPEI coarse resolution data spatial downscaling method based on machine learning in accordance with an embodiment of the present application;
FIG. 2 is a schematic diagram of a result of accuracy of a downscaling model on SPEI-6 in a machine learning based SPEI coarse resolution data spatial downscaling method according to an embodiment of the present application;
FIG. 3 is a spatial contrast plot of an original SPEI and a downscaled SPEI in a machine learning based SPEI coarse resolution data spatial downscaling method in accordance with an embodiment of the present application;
FIG. 4 is a monthly correlation coefficient of an original SPEI and a downscaled SPEI in a machine learning based SPEI coarse resolution data spatial downscaling method according to an embodiment of the present application;
FIG. 5 is a spatial distribution diagram of a machine-learning based SPEI coarse resolution data spatial downscaling method applied to a research area, SPEI-6 in 2006, in accordance with an embodiment of the present application;
FIG. 6 is a schematic diagram of a machine learning based SPEI coarse resolution data spatial downscaling apparatus in accordance with an embodiment of the present application; and
fig. 7 is a schematic diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
FIG. 1 is a flow chart of a SPEI coarse resolution data spatial downscaling method based on machine learning in accordance with an embodiment of the present application. The method of the present application is described in detail below with a city in China as a research area, and as shown in the figure, the method of the embodiment may include:
(1) A city in China is selected as a research area, an original coarse resolution 0.5-DEG long-time sequence SPEI space dataset (SPEIbase v 2.7) constructed by Sergio M.Vicente-Serrano is obtained, and a SPEI with a 6-month scale is taken as an example, and the time span is 1901-2020 month by month.
(2) Acquiring scale factor data, namely acquiring month-by-month rainfall and temperature data of China in corresponding time 1901-2020, wherein the data are acquired from a national Qinghai-Tibet plateau science data center, and the spatial resolution is 0.0083333 degrees (about 1 km), and specifically comprise a month average temperature, a month maximum temperature, a month minimum temperature and a month rainfall; and obtaining DEM data of STRM version4, calculating elevation, gradient and slope direction, and generating longitude and latitude space data.
(3) The spatial resolution of the downscaling factor raster data is unified to 0.5 deg., maintaining the same mesh size as the coarse resolution SPEI.
(4) The SPEI value and the downscaling factor value of each pixel are summarized, the original SPEI value is used as a dependent variable, the downscaling factor temperature, the precipitation and the topography factor are used as independent variables, sample data of all pixels are constructed, the sample data are divided into 70% training sets to train a model, and 30% testing sets to evaluate the model precision.
(5) Based on a machine learning GPR algorithm, training a downscaling model for each month by using a training set, performing Bayesian optimization on model super parameters in the process of constructing the GPR model, wherein a basic function (basic function) after optimization is a constant, a kernel function is a rational quadratic kernel function, and data in the model are subjected to standardized processing; utilizing R based on test set 2 The accuracy of the model is evaluated by the RMSE and MAE indexes, and the model prediction accuracy is high. FIG. 2 shows the accuracy of SPEI-6, finding R for each month 2 The median value is above 0.95, and both RMSE and MAE are less than 0.2.
(6) By using the constructed downscaling model to generate a long-time sequence SPEI space data product in China, and comparing the original rough SPEI with the downscaled SPEI in FIG. 3, the downscaled SPEI can be seen to be smoother and have more details.
(7) Then cross-verifying the down-scale SPEI and the original SPEI, and proving that the down-scale SPEI and the original SPEI have higher correlation, wherein the correlation of each month is more than 0.9, and the result is shown in figure 4;
in addition, the reliability of the downscaled data is verified by adopting a historical drought event, and taking a summer drought event of the research area in 2006 as an example, the model can accurately identify the historical drought event of a small area, SPEI-6 is lower from June, the drought of June is the most serious, and the reliability of the data produced by the downscaled model is verified, and the result is shown in figure 5.
FIG. 6 is a schematic diagram of a machine learning based SPEI coarse resolution data spatial downscaling apparatus in accordance with an embodiment of the present application. As shown in fig. 6, the apparatus includes: a coarse resolution data acquisition module 610 for acquiring raw coarse resolution SPEI raster data of a long time sequence within a study area; a downscaling factor data acquisition module 620, configured to acquire downscaling factor data in the research area; a downscaling factor resampling module 630 for resampling the downscaling factor to the same spatial resolution as the original SPEI; a sample data construction module 640, configured to construct sample data of all pixels, and divide the sample data into a test set and a training set; a spatial downscaling model construction module 650 for constructing a spatial downscaling machine learning model based on the training set and evaluating the accuracy of the model based on the test set; the downscaling SPEI data product forming module 660 is configured to input the downscaling factor raster data with high resolution into the trained model, and generate a downscaling SPEI data product with long-time sequence and high spatial resolution; and a verification module 670 for cross-verifying the original SPEI of step 1 and verifying the downscaled SPEI data product by studying drought events of the regional history.
Fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application, and as shown in fig. 7, the electronic device includes a processor 710, a memory 720, an input device 730, and an output device 740; the number of processors 710 in the electronic device may be one or more, one processor 710 being taken as an example in fig. 7; the processor 710, memory 720, input device 730, and output device 740 in the electronic device may be connected by a bus or other means, for example in fig. 7.
The memory 720 is a computer readable storage medium that may be used to store software programs, computer executable programs, and modules, such as program instructions/modules (e.g., coarse resolution data acquisition module 610, downscale factor data acquisition module 620, downscale factor resampling module 630, sample data construction module 640, spatial downscale model construction module 650, downscale SPEI data product formation module 660, and verification module 670) corresponding to a machine-learning-based SPEI coarse resolution data spatial downscaling method in embodiments of the present application. The processor 710 executes various functional applications of the electronic device and data processing by running software programs, instructions and modules stored in the memory 720, i.e., implementing the above-described machine learning-based SPEI coarse resolution data space downscaling method.
Memory 720 may include primarily a program storage area and a data storage area, wherein the program storage area may store an operating system, at least one application program required for functionality; the storage data area may store data created according to the use of the terminal, etc. In addition, memory 720 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some examples, memory 720 may further include memory remotely located relative to processor 710, which may be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof. The input device 730 may be used to receive input numeric or character information and to generate key signal inputs related to user settings and function control of the electronic device. The output device 740 may include a display device such as a display screen.
According to the application, based on the existing coarse-resolution long-time sequence SPEI data set, a downscaling model is constructed by taking the advantages of Gaussian process regression of a machine learning algorithm into consideration of climate, topography and geographic position factors, and a long-time sequence high-spatial resolution (0.0083 ℃) SPEI spatial data product in China is obtained, so that a reference is provided for drought assessment of finer scales.
Although embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the application, the scope of which is defined in the appended claims and their equivalents.

Claims (10)

1. A machine learning-based SPEI coarse resolution data spatial downscaling method, comprising:
step 1: acquiring original coarse resolution SPEI raster data of a long-time sequence in a research area;
step 2: acquiring downscaling factor data in a research area, wherein the downscaling factor data comprises high-resolution climate, topography and geographic position factor grid data, and the climate factors correspond to long-time sequences of SPEI;
step 3: resampling the downscaling factor to the same spatial resolution as the original SPEI;
step 4: constructing sample data of all pixels, taking the pixels as a statistical unit, extracting SPEI values and downscaling factor values of all pixels, taking all pixel values as sample data, and dividing the sample data into a test set and a training set;
step 5: constructing a space downscaling model based on a training set, taking a SPEI value as a dependent variable and a downscaling factor as an independent variable, constructing the downscaling model by means of Gaussian process regression of a machine learning algorithm, determining model parameters through Bayesian optimization, and evaluating the accuracy of the model by using a testing set; and
step 6: inputting the high-resolution downscaling factor raster data in the step 2 into a trained model to generate a long-time sequence high-spatial-resolution downscaling SPEI data product.
2. The method according to claim 1, further comprising step 7: the original SPEI of step 1 was used for cross-validation and downscaled SPEI data products were validated by studying drought events in the regional history.
3. The method of claim 1, wherein in step 2, the climate factors include a month average temperature, a month maximum temperature, a month minimum temperature, and a month precipitation; the topography factors include elevation, grade and slope; the geographic location factor includes longitude and latitude.
4. The method of claim 1, wherein in step 1, the coarse resolution is 0.5 ° x 0.5 ° and the downscaling factor spatial resolution is 0.0083 °.
5. The method according to claim 1, wherein in step 5, the hyper-parameters of the gaussian process regression model are obtained using bayesian optimization, and the downscaling model is built based on monthly data, respectively.
6. The method according to claim 1, characterized in that step 5 further comprises evaluating the accuracy of the model by calculating a decision coefficient (R 2 ) The model was evaluated for Root Mean Square Error (RMSE), mean Absolute Error (MAE).
7. The method of claim 2, wherein step 7 includes ensuring that downscaled data is not degraded in accuracy by calculating a correlation of the original SPEI to the downscaled SPEI, and evaluating the effectiveness of downscaled SPEI in identifying drought using historical drought times.
8. The method of claim 1, wherein in step 4, the samples are divided into 70% training sets to train the model and 30% test sets to evaluate model accuracy.
9. A machine learning-based SPEI coarse resolution data spatial downscaling method apparatus, comprising:
the coarse resolution data acquisition module is used for acquiring original coarse resolution SPEI raster data of a long-time sequence in the research area;
the downscaling factor data acquisition module is used for acquiring downscaling factor data in the research area;
a downscaling factor resampling module for resampling the downscaling factor to the same spatial resolution as the original SPEI;
the sample data construction module is used for constructing sample data of all pixels and dividing the sample data into a test set and a training set;
the space downscaling model construction module is used for constructing a space downscaling machine learning model based on the training set and evaluating the precision of the model based on the test set; and
the downscaling SPEI data product forming module is used for inputting the downscaling factor raster data with high resolution into the trained model to generate a downscaling SPEI data product with long-time sequence and high spatial resolution.
10. An electronic device, comprising: a memory and one or more processors;
the memory is used for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-8.
CN202310895388.5A 2023-07-20 2023-07-20 SPEI coarse resolution data space downscaling method and device based on machine learning Pending CN116796649A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310895388.5A CN116796649A (en) 2023-07-20 2023-07-20 SPEI coarse resolution data space downscaling method and device based on machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310895388.5A CN116796649A (en) 2023-07-20 2023-07-20 SPEI coarse resolution data space downscaling method and device based on machine learning

Publications (1)

Publication Number Publication Date
CN116796649A true CN116796649A (en) 2023-09-22

Family

ID=88036803

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310895388.5A Pending CN116796649A (en) 2023-07-20 2023-07-20 SPEI coarse resolution data space downscaling method and device based on machine learning

Country Status (1)

Country Link
CN (1) CN116796649A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117423476A (en) * 2023-12-18 2024-01-19 中国科学院地理科学与资源研究所 Echinococcosis epidemic rate prediction method based on downscaling and Bayesian model

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117423476A (en) * 2023-12-18 2024-01-19 中国科学院地理科学与资源研究所 Echinococcosis epidemic rate prediction method based on downscaling and Bayesian model
CN117423476B (en) * 2023-12-18 2024-03-08 中国科学院地理科学与资源研究所 Echinococcosis epidemic rate prediction method based on downscaling and Bayesian model

Similar Documents

Publication Publication Date Title
Tosunoglu et al. Application of copulas for regional bivariate frequency analysis of meteorological droughts in Turkey
Ali et al. Spatial–temporal characterization of rainfall in Pakistan during the past half-century (1961–2020)
Huerta et al. A spatiotemporal model for Mexico City ozone levels
Salih et al. Integrative stochastic model standardization with genetic algorithm for rainfall pattern forecasting in tropical and semi-arid environments
Hennemuth et al. Statistical methods for the analysis of simulated and observed climate data: applied in projects and institutions dealing with climate change impact and adaptation
CN113902580B (en) Historical farmland distribution reconstruction method based on random forest model
Palmate et al. Assessment of uncertainties in modelling land use change with an integrated cellular automata–Markov chain model
Xu et al. Evaluation of spatiotemporal dynamics of simulated land use/cover in China using a probabilistic cellular automata-Markov model
CN116796649A (en) SPEI coarse resolution data space downscaling method and device based on machine learning
CN110826689A (en) Method for predicting county-level unit time sequence GDP based on deep learning
Habitou et al. CHIRPS precipitation open data for drought monitoring: application to the Tensift basin, Morocco
Chen et al. A cellular automata approach of urban sprawl simulation with Bayesian spatially-varying transformation rules
Leung et al. Integration of air pollution data collected by mobile sensors and ground-based stations to derive a spatiotemporal air pollution profile of a city
CN112612822A (en) Beidou coordinate time series prediction method, device, equipment and storage medium
Liu et al. Identify optimal predictors of statistical downscaling of summer daily precipitation in China from three-dimensional large-scale variables
CN116881624B (en) Composite extreme event forecasting method, device, computer equipment and storage medium
de Aquino Ferreira et al. Validation of the representativeness of wind speed time series obtained from reanalysis data for Brazilian territory
CN116186493B (en) Meteorological forecasting method, device, equipment and medium based on numerical mode correction
CN117371303A (en) Prediction method for effective wave height under sea wave
CN116341841A (en) Runoff forecast error correction method, apparatus, device, medium and program product
Smith et al. Testing probabilistic adaptive real‐time flood forecasting models
Kneis et al. Verification of short-term runoff forecasts for a small Philippine basin (Marikina)
Zucchini et al. A Web‐based rainfall atlas for Southern Africa
Zhang et al. Statistical post-processing of precipitation forecasts using circulation classifications and spatiotemporal deep neural networks
Mushtaq et al. Hydrologic interpretation of machine learning models for 10-daily streamflow simulation in climate sensitive upper Indus catchments

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination