CN109375294A - A Downscaling Correction Method for Satellite Precipitation Data in Mountainous Areas - Google Patents

A Downscaling Correction Method for Satellite Precipitation Data in Mountainous Areas Download PDF

Info

Publication number
CN109375294A
CN109375294A CN201811084852.8A CN201811084852A CN109375294A CN 109375294 A CN109375294 A CN 109375294A CN 201811084852 A CN201811084852 A CN 201811084852A CN 109375294 A CN109375294 A CN 109375294A
Authority
CN
China
Prior art keywords
data
precipitation
monthly
downscaling
correction
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.)
Granted
Application number
CN201811084852.8A
Other languages
Chinese (zh)
Other versions
CN109375294B (en
Inventor
杜军凯
李晓星
刘欢
贾仰文
牛存稳
仇亚琴
郝春沣
赵红莉
冶运涛
张海涛
张双虎
郑钊
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.)
China Institute of Water Resources and Hydropower Research
Original Assignee
China Institute of Water Resources and Hydropower Research
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 China Institute of Water Resources and Hydropower Research filed Critical China Institute of Water Resources and Hydropower Research
Priority to CN201811084852.8A priority Critical patent/CN109375294B/en
Publication of CN109375294A publication Critical patent/CN109375294A/en
Application granted granted Critical
Publication of CN109375294B publication Critical patent/CN109375294B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01W—METEOROLOGY
    • G01W1/00—Meteorology
    • G01W1/18—Testing or calibrating meteorological apparatus
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01W—METEOROLOGY
    • G01W1/00—Meteorology
    • G01W1/10—Devices for predicting weather conditions

Landscapes

  • Environmental & Geological Engineering (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Atmospheric Sciences (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Ecology (AREA)
  • Environmental Sciences (AREA)
  • Image Processing (AREA)

Abstract

本发明公开一种山区卫星降水数据的降尺度校正方法，包括：TRMM 3B42.V7卫星降水数据的读取和月降水量统计；校正变量的融合与空间尺度统一；回归降尺度模型建立；交叉验证与降尺度校正执行。本发明将山区气象台站观测降水数据融入到卫星降水数据的降尺度校正过程中，以交叉验证技术进行方法优选，充分发挥了山区有限观测数据的优势，降尺度校正后的降水量的精度及其与实测数据系列的一致性大幅提升；此外，本发明提出了多方法对比复核的降尺度校正技术，初步解决了单一降尺度校正方法的系统偏差问题，丰富了卫星降水数据降尺度校正方法体系，提升了结果的可信度。该方法在典型山区卫星降水产品的降尺度校正上适用性良好，相关成果可为系统掌握山区降水时空分布特征提供有力的支撑。

The invention discloses a downscaling correction method for satellite precipitation data in mountainous areas, including: reading of TRMM 3B42.V7 satellite precipitation data and monthly precipitation statistics; fusion of correction variables and unification of spatial scale; establishment of regression downscaling model; cross-validation Perform with downscaling correction. The present invention integrates the observational precipitation data from meteorological stations in mountainous areas into the downscaling correction process of satellite precipitation data, and uses cross-validation technology to optimize the method, which fully utilizes the advantages of limited observational data in mountainous areas. The consistency with the measured data series is greatly improved; in addition, the present invention proposes a multi-method comparison and review downscaling correction technology, which preliminarily solves the problem of systematic deviation of a single downscaling correction method, and enriches the satellite precipitation data downscaling correction method system. Increased confidence in the results. The method has good applicability in downscaling correction of satellite precipitation products in typical mountainous areas, and the related results can provide strong support for systematically grasping the spatial and temporal distribution characteristics of precipitation in mountainous areas.

Description

A kind of NO emissions reduction bearing calibration of mountain area satellite precipitation data
Technical field
The present invention relates to the NO emissions reduction bearing calibrations of hydraulic engineering technical field more particularly to satellite precipitation data, specifically For a kind of NO emissions reduction bearing calibration of mountain area satellite precipitation data.
Background technique
Total input of the precipitation as water cycle process is that water cycle process is most important, one of most active element, in substance Key player is play in movement and energy exchange.Precisely parse and paddle affairs of the accurate estimation of precipitation to water cycle process Science decision it is most important.By the multifactor impacts such as mountain range trend, terrain slope and moisture source, mountainous region water cycle process tool There is vertical zonality.However, existing ground observation website is unevenly distributed, be laid in low altitude area region more, economic condition it is poor, Unfrequented High aititude mountain area, website is rare, and data is deficient.The Precipitation Distribution in Time and Space information obtained by interpolation algorithm, by It is verified in the observational data for lacking High aititude mountain area, spread result is difficult to accurately hold the spatial distribution characteristic of precipitation.
Continue to bring out in recent years satellite precipitation data (including CMAP, TMPA, GPCP, CMORPH, PERSIANN-CDR, NRL-Blend and GPM etc.) support can be provided to lack the hydrological analysis calculating in data mountain area, but there are spaces point for these data Relatively thick and precision deficiency the problem of resolution, it is still necessary to carry out space NO emissions reduction and accuracy correction to meet application demand.
Present satellites precipitation data carries out NO emissions reduction timing in mountain area and has the following problems: (1) using single method more NO emissions reduction correction, the NO emissions reduction knot when laying less High aititude mountain area for Rainfall Monitoring website are carried out to satellite precipitation data Fruit verifying is insufficient, and error is larger；(2) existing method does not consider the actual measurement precipitation data monitored by laying website, only by it For the verifying and assessment of NO emissions reduction result, fails joint and play on the accuracy benefits of ground observation and the face of satellite remote sensing precipitation Advantage.
Summary of the invention
The technical problems to be solved by the invention are that existing satellite precipitation data NO emissions reduction bearing calibration is overcome to lack Existing defect when the High aititude mountain area application of field data, propose blended based on ground observation and moonscope it is multi-party The mountain area satellite precipitation data NO emissions reduction bearing calibration of method comparison review.This method melts mountain area meteorological station observation precipitation data Enter into the NO emissions reduction correction course of satellite precipitation data, effectively reduce systematic error, improves correction result and actual measurement number According to the goodness of fit and consistency, for system grasp Precipitation in Mountain Area spatial and temporal patterns provide science and technology support, enrich satellite The method system of precipitation data NO emissions reduction correction.
The object of the present invention is achieved like this: a kind of NO emissions reduction bearing calibration of mountain area satellite precipitation data, the side Method includes four parts: the reading of I .TRMM 3B42.V7 satellite remote sensing precipitation and monthly total precipitation statistics；II, correcting variable is melted It closes unified with space scale；III, returns NO emissions reduction model foundation；IV, cross validation and NO emissions reduction correction execute.Steps are as follows:
The specific steps of the reading of I .TRMM 3B42.V7 satellite remote sensing precipitation and monthly total precipitation statistics:
Step 1: according to the vector boundary in research area, obtain the research upper left of area's rectangular space range, upper right, lower-left and The space coordinate Geo on the vertex of bottom right 4[top,left], Geo[top,right], Geo[bottom,left], Geo[bottom,right]。
Step 2: according to Geo[top,left], Geo[top,right], Geo[bottom,left], Geo[bottom,right]Four vertex institutes Determining square boundary to make full use of measured data, and preferably reflects the influence that landform is distributed Precipitation in Mountain Area, along research Area's outer boundary expands 0.5 ° outward and establishes buffer area, and TRMM 3B42.V7 precipitation is shown in Table with HDF stored in file format, arrangement mode 1, the TRMM precipitation information within the scope of buffer area is read, research area's time interval 3h, 0.25 ° of spatial resolution of TRMM drop are obtained Water number is according to A.
Table 1TRMM 3B42.V7 satellite Precipitation Products arrangement mode
Annotation: table middle latitude negative value indicates that south latitude, positive value indicate north latitude；Longitude negative value indicates west longitude, and positive value indicates east Through.
Step 3: the satellite precipitation information extracted is counted by pixel, obtains the TRMM precipitation number in each 1~December of pixel According to B.
The fusion and the unified specific steps of space scale of II, correcting variable:
Step 1: the Daily rainfall amount that meteorological station monitors in Revision area, and count each website month by month Precipitation data C.
Step 2: the TRMM precipitation grid where meteorological station is determined according to geographical coordinate, with the actual measurement precipitation of weather station Obs is measured instead of the satellite remote sensing precipitation on the TRMM grid, the TRMM precipitation data B moon for being modified to " star-ground " fusion is dropped Water number is according to D.
Step 3: digital elevation (DEM) and normalized differential vegetation index (NDVI) data are cut according to research area's range, are obtained The NDVI data F after dem data E and cutting after cutting.
Step 4: dem data E is subjected to resampling, obtains the dem data G of 0.25 ° of spatial resolution, and according to DEM Data G calculates the Gradient H and slope aspect data J on the research each pixel in area.
Step 5: the NDVI data F after cutting is subjected to resampling, obtains 0.25 ° of spatial resolution, NDVI number month by month According to K.
Step 6: moon precipitation data D, dem data G, Gradient H, slope aspect data J, NDVI data K spatial and temporal scales are united After one, the centroid point of 0.25 ° of spatial resolution grid is calculated, obtains the longitude data L of each grid, latitude data M.
The specific steps of III, recurrence NO emissions reduction model foundation:
Step 1: determining the independent variable and dependent variable for returning NO emissions reduction model, and the D of precipitation data month by month that fusion is obtained makees For dependent variable, by the dem data G that spatial and temporal scales are unified, Gradient H, slope aspect data J, NDVI data K, longitude data L, latitude Degree is according to M as independent variable.
Step 2: dem data E is subjected to resampling, obtains the dem data N of 0.05 ° of spatial resolution, is counted according to data N Calculation obtains the Gradient O of 0.05 ° of spatial resolution, slope aspect data P.
Step 3: NDVI data F is subjected to resampling, obtains the NDVI data Q of 0.05 ° of spatial resolution.
Step 4: the Gradient O of 0.05 ° of spatial resolution grid, the space lattice of slope aspect data P, NDVI data Q are Completely the same, the centroid of NO emissions reduction grid is calculated in an optional raster data, is calculated under the spatial resolution each The longitude data R of grid, latitude data S.
Step 5: using multiple linear regression analysis method, establishes the multiple regression relationship between precipitation data D and independent variable month by month MLR。
Step 6: using partial least-square regression method, establishes the offset minimum binary between precipitation data D and independent variable month by month Regression relation PLSR.
Step 7: using Geographically weighted regression procedure, establishes the Geographical Weighted Regression between precipitation data D and independent variable month by month Relationship GWR.
Step 8: Gradient O, slope aspect data P, NDVI data Q, longitude data R, latitude data S are brought into polynary time Return in relationship MLR, executes NO emissions reduction model, the precipitation T1 month by month after obtaining NO emissions reduction correction.
Step 9: Gradient O, slope aspect data P, NDVI data Q, longitude data R, latitude data S are brought into partially minimum Two multiply in regression relation PLSR, execute NO emissions reduction model, the precipitation T2 month by month after obtaining NO emissions reduction correction.
Step 10: Gradient O, slope aspect data P, NDVI data Q, longitude data R, latitude data S are brought into geographical add It weighs in regression relation GWR, executes NO emissions reduction model, the precipitation T3 month by month after obtaining NO emissions reduction correction.
The specific steps that IV, cross validation and NO emissions reduction correction execute:
Step 1: assuming that meteorological station number is Count, reduce by 1 meteorological station every time, II, executes correcting variable It merges the step two unified with space scale to blend TRMM precipitation and ground observation precipitation, obtains " the star-not comprising the point The moon precipitation data V of ground " fusion.
Step 2: moon precipitation data V obtained using step 1 as dependent variable, by dem data G, Gradient H, slope aspect number It repeats III, as independent variable according to J, NDVI data K, longitude data L, latitude data M and returns NO emissions reduction model foundation part The step of five~step 10 Count times.
Step 3: the MLR NO emissions reduction correction monthly total precipitation for calculating Count times is done sums average, obtains multiple linear and returns The NO emissions reduction of method is returned to correct monthly total precipitation raster data W1；
Step 4: the PLSR NO emissions reduction correction monthly total precipitation for calculating Count times is done sums average, obtains geographical weight back The NO emissions reduction of method is returned to correct monthly total precipitation raster data W2；
Step 5: the GWR NO emissions reduction correction monthly total precipitation for calculating Count times is done sums average, obtains geographical weight back The NO emissions reduction of method is returned to correct monthly total precipitation raster data W3；
Step 6: the precipitation number generated according to the geographical coordinate of meteorological station, matching step three, step 4 and step 5 According to W1、W2、W3The monthly total precipitation Y corrected with meteorological station spatial position X, the NO emissions reduction of grid where extracting meteorological station, Calculate the coefficient of determination R surveyed between precipitation data obs and data Y month by monthj 2, root-mean-square error RMSEjAnd average relative error AREj。
Step 7: the cross validation of multiple linear regression and Geographically weighted regression procedure is completed.
Step 8: according to cross validation results, using moon precipitation data D of " star-ground " fusion as dependent variable, with dem data G, Gradient H, slope aspect data J, NDVI data K, longitude data L and latitude data M are independent variable, use optimal method pair It studies area's monthly total precipitation and carries out NO emissions reduction correction, the monthly total precipitation data Z after being corrected, and extract in optimum regression relationship The regression coefficient AA of dependent variable and elevation.
Step 9: the regression coefficient AA of the Z of precipitation data month by month, precipitation and elevation after correction are converted into grid map Piece obtains research area and corrects the gradient grid map that precipitation and monthly total precipitation change along elevation month by month.
Step 10: gradient grid of the precipitation with it along elevation variation is cut month by month with the vector boundary batch in research area Figure obtains research average precipitation in 1~December of area (see Figure 24), and the research monthly precipitation in area along the change of gradient of elevation (see Figure 25).
Further, IV, cross validation and NO emissions reduction correction execute, coefficient of determination R in step 6j 2, root-mean-square error RMSEjWith average relative error AREjCalculation formula be respectively as follows:
In formula:CountTo survey website number, obsiFor the actual measurement precipitation of i-th of website,For all actual measurement websites Average precipitation, YiPrecipitation, R are corrected for the NO emissions reduction of i-th of websitej 2、RMSEjAnd AREjWhat respectively jth time was verified determines Determine coefficient, root-mean-square error and mean relative deviation.
Further, IV, cross validation and NO emissions reduction correction execute, the principle that cross validation is deferred in step 7 are as follows:
The beneficial effect comprise that:
(1) by the NO emissions reduction correction course of the precipitation measurement data fusion of meteorological stations to satellite precipitation data, By cross validation, increases the number of numerical experiment to ensure the stability of analysis result, actual measurement number is farthest utilized According to advantage；
(2) the synchronous recurrence NO emissions reduction for carrying out satellite precipitation datas using three kinds of methods correct and to carry out method preferred, logical The comparison review for crossing a variety of methods, effectively reduces systematic error.
Satellite precipitation data NO emissions reduction correction upper applicability of this method in High aititude mountain area is good, and correlated results can be to be System grasps Precipitation in Mountain Area spatial and temporal patterns and provides science and technology support.
Detailed description of the invention
In the following with reference to the drawings and specific embodiments, invention is further described in detail；
Fig. 1 is the flow chart of the embodiment of the present invention the method；
Fig. 2 is Wild jujube in Taihang Mountain Area boundary, buffer area boundary, original TRMM raster data centroid point and the meteorology that the present invention uses Station distribution figure；
Fig. 3 be present invention determine that 0.25 ° of buffer area spatial resolution digital elevation data；
Fig. 4 be present invention determine that the 0.25 ° of spatial resolution in buffer area Gradient；
Fig. 5 be present invention determine that 0.25 ° of buffer area spatial resolution slope aspect data；
Fig. 6 be present invention determine that 0.25 ° of buffer area spatial resolution NDVI data；
Fig. 7 be present invention determine that 0.05 ° of buffer area spatial resolution digital elevation data；
Fig. 8 be present invention determine that the 0.05 ° of spatial resolution in buffer area Gradient；
Fig. 9 be present invention determine that 0.05 ° of buffer area spatial resolution slope aspect data；
Figure 10 be present invention determine that 0.05 ° of buffer area spatial resolution NDVI data；
Figure 11 be present invention determine that NO emissions reduction to after 0.05 ° precipitation raster data centroid point distribution；
Figure 12 be present invention determine that cross validation after January average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 13 be present invention determine that cross validation after February average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 14 be present invention determine that cross validation after March average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 15 be present invention determine that cross validation after April average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 16 be present invention determine that cross validation after May average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 17 be present invention determine that cross validation after June average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 18 be present invention determine that cross validation after July average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 19 be present invention determine that cross validation after August average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 20 be present invention determine that cross validation after September average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 21 be present invention determine that cross validation after October average original TRMM precipitation, the correction of MLR NO emissions reduction The comparison diagram of precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 22 be present invention determine that cross validation after November average original TRMM precipitation, MLR NO emissions reduction school The comparison diagram of positive precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 23 be present invention determine that cross validation after December average original TRMM precipitation, MLR NO emissions reduction school The comparison diagram of positive precipitation, PLSR NO emissions reduction correction precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation；
Figure 24 a~Figure 24 l be present invention determine that the research Qu Yueping that corrects of NO emissions reduction is carried out using best practice The spatial distribution of equal precipitation；
Figure 25 a~Figure 25 b be present invention determine that the research Qu Nianping that corrects of NO emissions reduction is carried out using best practice The spatial distribution and isopleth of equal precipitation；
Figure 26 a~Figure 26 l be present invention determine that the research Qu Yueping that corrects of NO emissions reduction is carried out using best practice The spatial distribution for the gradient value that equal precipitation changes along elevation.
Specific embodiment
In the following with reference to the drawings and specific embodiments, the present invention is described in further details:
Taihang mountain range extends to the Wangwu Shan Mountain of Shanxi, Henan border land North gets Beijing's Western Hills southwards, and west connects Shanxi plateau, The North China Plain is faced in east, and in northeast~southwest trend, it is the second ladder of China's landform east edge that be continuous more than 400 kilometers.The present invention chooses Wild jujube in Taihang Mountain Area is as research area.It is as follows to study area's overview: about 12.78 ten thousand km of the gross area2, height above sea level section is -65~3059m, is indulged Across Beijing, Hebei, Shanxi and Henan Si Sheng (city), Haihe River, two, the Yellow River level-one basin are traversed；In China's subhumid and half Evergreen conifruticeta, theropencedrymion, broad-leaved deciduous forest, artificial forest, fallen leaves shrubbery and Cao Po is distributed in arid biogeographic zone intermediate zone Etc. vegetation patterns；Belong to monsoon climate of medium latitudes, summer is burning hot and rainy by wet southeast wind effect is warmed up, and winter is by dry and cold northwester It influences and cold short of rain, mean annual precipitation is in 400~600mm；Main River Systems include Yellow River basin the Yellow River mainstream and Qin He, The tributaries such as Dan He, and belong to the rivers such as Caobai River, the Yongdinghe River, Daqinghe River, the Zhanghe River of Haihe basin.Taihang Mountain is the allusion quotation of monsoon region Type mountain range has preferable representative using Wild jujube in Taihang Mountain Area as research object.
The reading of I .TRMM 3B42.V7 satellite remote sensing precipitation and monthly total precipitation statistics；
90 × 90m used from NASA (NASA) and U.S. National Imagery and Mapping Agency (NIMA) joint publication is empty Between resolution ratio digital elevation (DEM) data set, 137 meteorological site Daily rainfall data of selection are from China national gas Basic data as 2000~2011 years meteorological data data sets that office reorganizes, as the expansion of this example.Example with ArcGIS10.2 software is display platform, and program calculation is realized by Python and MATLAB language.
Step 1: according to the vector boundary in research area, obtain the research upper left of area's rectangular space range, upper right, lower-left and The space coordinate Geo on the vertex of bottom right 4[top,left], Geo[top,right], Geo[bottom,left], Geo[bottom,right]。
Step 2: according to Geo[top,left], Geo[top,right], Geo[bottom,left], Geo[bottom,right]Four vertex institutes Determining square boundary to make full use of measured data, and preferably reflects the influence that landform is distributed Precipitation in Mountain Area, along research Area's outer boundary expands 0.5 ° outward and establishes buffer area, and TRMM 3B42.V7 precipitation is shown in Table with HDF stored in file format, arrangement mode 1.Using the hdfread function of MATLAB software, the TRMM precipitation information within the scope of buffer area is read, is obtained between research area's time Every 3h, 0.25 ° of spatial resolution of TRMM precipitation data A.
Step 3: the satellite precipitation information extracted is counted by pixel, obtains the TRMM precipitation number in each 1~December of pixel According to B, Fig. 2 is shown in the distribution of TRMM precipitation grid central point.
The fusion and the unified specific steps of space scale of II, correcting variable:
Step 1: the Daily rainfall amount of meteorological station monitoring within the scope of buffer area is arranged, and counts each website Precipitation data C month by month, research area's range, the range of buffer area are shown in Fig. 2.
Step 2: the TRMM precipitation grid where meteorological station is determined according to geographical coordinate, with the actual measurement precipitation of weather station Amount obs replaces the satellite remote sensing precipitation on the TRMM grid, the TRMM precipitation data B moon for being modified to " star-ground " fusion is dropped Water number is according to D.
Step 3: the Extract by under 10.2 tool box platform Spatial Analyst tools Arcgis is used Mask tool cuts digital elevation (DEM) and normalized differential vegetation index (NDVI) data according to buffer area range, after obtaining cutting Dem data E and cut after NDVI data F.
Step 4: the Resample tool in the tool box Data Management Tools of 10.2 platform of Arcgis is used Dem data E is subjected to resampling, obtains the dem data G (see Fig. 3) of 0.25 ° of spatial resolution, and calculate and delay according to data G Area is rushed by the Gradient H (see Fig. 4) and slope aspect data J of pixel (see Fig. 5).
Step 5: the Resample tool in the tool box Data Management Tools of 10.2 platform of Arcgis is used NDVI data F after cutting is subjected to resampling, obtains 0.25 ° of spatial resolution, NDVI data K month by month (see Fig. 6).
Step 6: moon precipitation data D, dem data G, Gradient H, slope aspect data J, NDVI data K spatial and temporal scales are united After one, calculated using the Raster To Point tool under 10.2 tool box platform Conversion tools Arcgis The centroid of each raster data under 0.25 ° of spatial resolution obtains the longitude data L (can obtain from Fig. 2) of each grid, latitude Data M (can be obtained) from Fig. 2.
The specific steps of III, recurrence NO emissions reduction model foundation:
Step 1: determining the independent variable and dependent variable for returning NO emissions reduction model, and the D of precipitation data month by month that fusion is obtained makees For dependent variable, by the dem data G that spatial and temporal scales are unified, Gradient H, slope aspect data J, NDVI data K, longitude data L, latitude Degree is according to M as independent variable.
Step 2: the Resample work in the tool box Data Management Tools of 10.2 platform of Arcgis is used Dem data E is carried out resampling, obtains the dem data N (see Fig. 7) of 0.05 ° of spatial resolution, use Arcgis 10.2 by tool The Gradient of 0.05 ° of spatial resolution is calculated in Slope tool under the tool box platform Spatial Analyst tools O (see Fig. 8) is calculated 0.05 ° using the Aspect tool under 10.2 tool box platform Spatial Analysis Arcgis The slope aspect data P of spatial resolution (see Fig. 9).
Step 3:, will using the Resample tool under 10.2 tool box platform Spatial Analysis Arcgis NDVI data F carries out resampling, obtains the NDVI data Q (example is shown in Figure 10) of 0.05 ° of spatial resolution.
Step 4: Gradient O, the space lattice of slope aspect data P, NDVI data Q of 0.05 ° of spatial resolution grid are complete It is complete consistent, use the Raster To Point tool optional one under 10.2 tool box platform Conversion tools Arcgis The centroid point (see Figure 11) of NO emissions reduction grid is calculated in a raster data, and each grid under the spatial resolution is calculated Longitude data R (can be read) from Figure 11, and latitude data S (can be read) from Figure 11.
Step 5: using multiple linear regression analysis method, writes MATLAB program, establishes precipitation data D and independent variable month by month Multiple regression relationship MLR.
Step 6: using partial least-square regression method, writes MATLAB program, establishes precipitation data D month by month and becomes certainly The Partial Least Squares Regression relationship PLSR of amount.
Step 7: using Geographically weighted regression procedure, write MATLAB program, determines geographical power using Gaussian function method Weight establishes month by month the Geographical Weighted Regression relationship GWR of precipitation data D and independent variable.
Step 8: using the geotiffread function of MATLAB software, read and by Gradient O, slope aspect data P, NDVI data Q, longitude data R, latitude data S are brought into respectively in multiple regression relationship MLR, are executed NO emissions reduction model, are dropped Precipitation T month by month after dimension correction1。
Step 9: using the geotiffread function of MATLAB software, read and by Gradient O, slope aspect data P, NDVI data Q, longitude data R, latitude data S are brought into respectively in Partial Least Squares Regression relationship PLSR, execute NO emissions reduction model, Precipitation T month by month after obtaining NO emissions reduction correction2。
Step 10: using the geotiffread function of MATLAB software, read and by Gradient O, slope aspect data P, NDVI data Q, longitude data R, latitude data S are brought into respectively in Geographical Weighted Regression relationship GWR, are executed NO emissions reduction model, are obtained Precipitation T month by month to after NO emissions reduction correction3。
The specific steps of IV, cross validation and NO emissions reduction correction:
Step 1: assuming that meteorological station number is Count, reduce by 1 meteorological station every time, execution II be " correcting variable The step of fusion and space scale unification " part two, blends TRMM precipitation and ground observation precipitation, obtains not including the point " star-ground " fusion moon precipitation data V.
Step 2: moon precipitation data V obtained using step 1 as dependent variable, by dem data G, Gradient H, slope aspect number III " returning NO emissions reduction model foundation " portion is repeated as independent variable according to J, NDVI data K, longitude data L, latitude data M The step of dividing, five~step 10 was Count times total.
Step 3: the MLR NO emissions reduction correction monthly total precipitation for calculating Count times is done sums average, obtains multiple linear and returns The NO emissions reduction of method is returned to correct monthly total precipitation raster data W1；
Step 4: the PLSR NO emissions reduction correction monthly total precipitation for calculating Count times is done sums average, obtains geographical weight back The NO emissions reduction of method is returned to correct monthly total precipitation raster data W2；
Step 5: the GWR NO emissions reduction correction monthly total precipitation for calculating Count times is done sums average, obtains geographical weight back The NO emissions reduction of method is returned to correct monthly total precipitation raster data W3；
Step 6: the precipitation number generated according to the geographical coordinate of meteorological station, matching step three, step 4 and step 5 According to W1、W2、W3The monthly total precipitation corrected with the spatial position X of meteorological station, the NO emissions reduction of grid where extracting meteorological station Y calculates the coefficient of determination R surveyed between precipitation data obs and data Y month by monthj 2, root-mean-square error RMSEjAnd average relative error AREj.2~Figure 23 of the result is shown in Figure 1 of each moon, summarizing for comparing result are shown in Table 2 and table 3.
In formula:CountTo survey website number, obsiFor the actual measurement precipitation of i-th of website,For all actual measurement websites Average precipitation, YiPrecipitation, R are corrected for the NO emissions reduction of i-th of websitej 2、RMSEjAnd AREjWhat respectively jth time was verified determines Determine coefficient, root-mean-square error and mean relative deviation.
Table 2 be after the embodiment of the present invention cross validation each month original TRMM precipitation, MLR NO emissions reduction correction precipitation Amount；PLSR NO emissions reduction corrects precipitation, the coefficient of determination of GWR NO emissions reduction correction precipitation and website actual measurement precipitation series and equal The statistical form of square error average value；
Table 2R2Result is cross-checked with RMSE
Table 3 be after the embodiment of the present invention cross validation each month original TRMM precipitation, MLR NO emissions reduction correction precipitation Amount；PLSR NO emissions reduction corrects the mean relative deviation of precipitation, GWR NO emissions reduction correction precipitation and website actual measurement precipitation series The statistical form of mean value.
Table 3ARE cross-checks result
Step 7: according to principle listed by formula (4), multiple linear regression is completed, minimum two partially adds at recurrence and geography Weigh the cross validation of homing method.
Step 8: according to cross validation results, using moon precipitation data D of " star-ground " fusion as dependent variable, with dem data G, Gradient H, slope aspect data J, NDVI data K, longitude data L, latitude data M are independent variable, using best practice to grinding Study carefully area's monthly total precipitation carry out NO emissions reduction correction, the monthly total precipitation data Z after being corrected, and extract in optimum regression relationship because The regression coefficient AA of variable and elevation.
Step 9: it using the geotiffwrite function of MATLAB software, obtains research area and corrects precipitation and the moon month by month The gradient grid map that precipitation changes along elevation.
Step 10: being based on Python translation and compiling environment, calls the arcpy secondary development bag of Arcgis 10.2, according to research area Vector boundary (see Fig. 2), using arcpy.sa.Extractbymask function batch cut precipitation month by month and its along elevation The gradient grid map of variation obtains research average precipitation in 1~December of area (see Figure 24), and the research monthly precipitation edge in area The change of gradient of elevation (see Figure 25).
Finally it should be noted that being only used to illustrate the technical scheme of the present invention and not to limit it above, although referring to preferable cloth The scheme of setting describes the invention in detail, those skilled in the art should understand that, it can be to technology of the invention Scheme (such as utilization, sequencing of step of various formula etc.) is modified or replaced equivalently, without departing from the present invention The spirit and scope of technical solution.

Claims (4)

1.一种山区卫星降水数据的降尺度校正方法，其特征在于，所述降水降尺度方法包括四个部分：Ⅰ.TRMM 3B42.V7卫星遥感降水的读取和月降水量统计；Ⅱ.校正变量的融合与空间尺度统一；Ⅲ.回归降尺度模型建立；Ⅳ.交叉验证与降尺度校正执行。1. A downscaling correction method for satellite precipitation data in mountainous areas, characterized in that the precipitation downscaling method comprises four parts: I. TRMM 3B42.V7 satellite remote sensing precipitation reading and monthly precipitation statistics; II. Correction Fusion of variables and unification of spatial scale; Ⅲ. Regression downscaling model establishment; Ⅳ. Cross-validation and downscaling correction execution. 2.根据权利要求1所述的山区卫星降水数据的降尺度校正方法，其特征在于，其具体步骤是：2. the downscaling correction method of mountain satellite precipitation data according to claim 1, is characterized in that, its concrete steps are: Ⅰ.TRMM 3B42.V7卫星遥感降水的读取和月降水量统计：Ⅰ. TRMM 3B42.V7 satellite remote sensing precipitation reading and monthly precipitation statistics: 步骤一：根据研究区的矢量边界，获取研究区矩形空间范围的左上、右上、左下和右下4个顶点的空间坐标Geo[top,left]，Geo[top,right]，Geo[bottom,left]，Geo[bottom,right]；Step 1: According to the vector boundary of the study area, obtain the spatial coordinates of the four vertices Geo [top,left] , Geo [top,right] , Geo [bottom,left ] of the upper left, upper right, lower left and lower right of the rectangular space of the study area ] , Geo [bottom, right] ; 步骤二：根据Geo[top,left]，Geo[top,right]，Geo[bottom,left]，Geo[bottom,right]四个顶点所确定的矩形边界，沿研究区外边界向外拓展0.5°建立缓冲区，读取缓冲区范围内的TRMM降水信息，得到研究区时间间隔3h、空间分辨率0.25°的TRMM降水数据A；Step 2: According to the rectangular boundary determined by the four vertices of Geo [top,left] , Geo [top,right] , Geo [bottom,left] , Geo [bottom,right] , extend 0.5° outward along the outer boundary of the study area Set up a buffer zone, read the TRMM precipitation information within the buffer zone, and obtain the TRMM precipitation data A with a time interval of 3 hours and a spatial resolution of 0.25° in the study area; 步骤三：逐像元统计抽取出的卫星降水信息，得到每个像元1～12月的TRMM降水数据B；Step 3: Statistically extract the satellite precipitation information pixel by pixel, and obtain the TRMM precipitation data B for each pixel from January to December; Ⅱ.校正变量的融合与空间尺度统一：Ⅱ. Fusion of correction variables and unification of spatial scale: 步骤一：整理研究区内气象台站监测的逐日降水量，并且统计出每个站点的逐月降水量数据C；Step 1: Collate the daily precipitation monitored by the meteorological stations in the study area, and count the monthly precipitation data C of each station; 步骤二：根据地理坐标判定气象台站所在的TRMM降水网格，用气象站的实测降水量obs代替该TRMM网格上的卫星遥感降水量，将TRMM降水数据B修正为“星-地”融合的月降水数据D；Step 2: Determine the TRMM precipitation grid where the meteorological station is located according to the geographical coordinates, replace the satellite remote sensing precipitation on the TRMM grid with the measured precipitation obs of the meteorological station, and correct the TRMM precipitation data B to "star-ground" fusion Monthly precipitation data D; 步骤三：根据研究区范围裁剪数字高程和归一化植被指数数据，得到裁剪后的DEM数据E和裁剪后的NDVI数据F；Step 3: Cut the digital elevation and normalized vegetation index data according to the scope of the study area, and obtain the cut DEM data E and the cut NDVI data F; 步骤四：将DEM数据E进行重采样，得到0.25°空间分辨率的DEM数据G，并且根据DEM数据G计算研究区每个像元上的坡度数据H和坡向数据J；Step 4: Resample the DEM data E to obtain the DEM data G with a spatial resolution of 0.25°, and calculate the slope data H and the slope aspect data J on each pixel in the study area according to the DEM data G; 步骤五：将裁剪后的NDVI数据F进行重采样，得到0.25°空间分辨率、逐月的NDVI数据K；Step 5: Resampling the cropped NDVI data F to obtain monthly NDVI data K with a spatial resolution of 0.25°; 步骤六：将月降水数据D、DEM数据G、坡度数据H、坡向数据J、NDVI数据K时空尺度统一后，计算0.25°空间分辨率栅格的形心点，得到每个栅格的经度数据L，纬度数据M；Step 6: After unifying the temporal and spatial scales of monthly precipitation data D, DEM data G, slope data H, aspect data J, and NDVI data K, calculate the centroid point of the 0.25° spatial resolution grid, and obtain the longitude of each grid data L, latitude data M; Ⅲ.回归降尺度模型建立：Ⅲ. Establishment of regression downscaling model: 步骤一：确定回归降尺度模型的自变量和因变量，将融合得到的逐月降水数据D作为因变量，将时空尺度统一的DEM数据G，坡度数据H，坡向数据J，NDVI数据K，经度数据L，纬度数据M作为自变量；Step 1: Determine the independent variables and dependent variables of the regression downscaling model, use the monthly precipitation data D obtained by fusion as the dependent variable, and use the unified time and space scale DEM data G, slope data H, aspect data J, NDVI data K, Longitude data L, latitude data M as independent variables; 步骤二：将DEM数据E进行重采样，得到0.05°空间分辨率的DEM数据N，根据数据N计算得到0.05°空间分辨率的坡度数据O，坡向数据P；Step 2: Resampling the DEM data E to obtain the DEM data N with a spatial resolution of 0.05°, and calculate the slope data O and the slope aspect data P with a spatial resolution of 0.05° according to the data N; 步骤三：将NDVI数据F进行重采样，得到0.05°空间分辨率的NDVI数据Q；Step 3: Resampling the NDVI data F to obtain the NDVI data Q with a spatial resolution of 0.05°; 步骤四：0.05°空间分辨率的坡度数据O、坡向数据P、NDVI数据Q的空间网格完全一致，任选一个栅格数据计算得到降尺度栅格的形心，计算得到该空间分辨率下的每个栅格的经度数据R，纬度数据S；Step 4: The spatial grids of the slope data O, the aspect data P, and the NDVI data Q with a spatial resolution of 0.05° are completely consistent, and the centroid of the downscaled grid can be calculated from any grid data, and the spatial resolution can be obtained by calculation. The longitude data R and latitude data S of each grid below; 步骤五：使用多元线性回归方法，建立逐月降水数据D和自变量间的多元回归关系MLR；Step 5: Use the multiple linear regression method to establish the multiple regression relationship MLR between the monthly precipitation data D and the independent variables; 步骤六：使用偏最小二乘回归方法，建立逐月降水数据D和自变量间的偏最小二乘回归关系PLSR；Step 6: Use the partial least squares regression method to establish the partial least squares regression relationship PLSR between the monthly precipitation data D and the independent variables; 步骤七：使用地理加权回归方法，建立逐月降水数据D和自变量间的地理加权回归关系GWR；Step 7: Use the geographically weighted regression method to establish the geographically weighted regression relationship GWR between the monthly precipitation data D and the independent variable; 步骤八：将坡度数据O、坡向数据P、NDVI数据Q、经度数据R、纬度数据S带入多元回归关系MLR中，执行降尺度模型，得到降尺度校正后的逐月降水量T1；Step 8: Bring the slope data O, the aspect data P, the NDVI data Q, the longitude data R, and the latitude data S into the multiple regression relationship MLR, execute the downscaling model, and obtain the downscaled corrected monthly precipitation T 1 ; 步骤九：将坡度数据O、坡向数据P、NDVI数据Q、经度数据R、纬度数据S带入偏最小二乘回归关系PLSR中，执行降尺度模型，得到降尺度校正后的逐月降水量T2；Step 9: Bring the slope data O, the aspect data P, the NDVI data Q, the longitude data R, and the latitude data S into the partial least squares regression relation PLSR, execute the downscaling model, and obtain the monthly precipitation after downscaling correction T 2 ; 步骤十：将坡度数据O、坡向数据P、NDVI数据Q、经度数据R、纬度数据S带入地理加权回归关系GWR中，执行降尺度模型，得到降尺度校正后的逐月降水量T3；Step 10: Bring the slope data O, the aspect data P, the NDVI data Q, the longitude data R, and the latitude data S into the geographically weighted regression relationship GWR, execute the downscaling model, and obtain the downscaled corrected monthly precipitation T 3 ; Ⅳ.交叉验证与降尺度校正执行：Ⅳ. Cross-validation and downscaling correction performed: 步骤一：假设气象台站个数为Count，每次减少1个气象台站，执行Ⅱ.校正变量的融合与空间尺度统一的步骤二对TRMM降水和地面观测降水相融合，得到不包含该点的“星-地”融合的月降水数据V；Step 1: Assuming that the number of meteorological stations is Count, reduce one meteorological station each time, and perform II. The fusion of correction variables and the unification of the spatial scale. "Star-Earth" fusion monthly precipitation data V; 步骤二：以步骤一得到的月降水数据V为因变量，将DEM数据G，坡度数据H，坡向数据J，NDVI数据K，经度数据L，纬度数据M作为自变量，重复执行Ⅲ.回归降尺度模型建立部分的步骤五～步骤十Count次；Step 2: Take the monthly precipitation data V obtained in Step 1 as the dependent variable, DEM data G, slope data H, slope aspect data J, NDVI data K, longitude data L, and latitude data M as independent variables, and repeat Ⅲ. Regression Steps 5 to 10 in the downscaling model building part are performed Count times; 步骤三：将计算Count次的MLR降尺度校正月降水量做算术平均，得到多元线性回归方法的降尺度校正月降水量栅格数据W1；Step 3: Perform arithmetic mean of the MLR downscaled corrected monthly precipitation calculated Count times to obtain the downscaled corrected monthly precipitation raster data W 1 of the multiple linear regression method; 步骤四：将计算Count次的PLSR降尺度校正月降水量做算术平均，得到地理加权回归方法的降尺度校正月降水量栅格数据W2；Step 4: Perform the arithmetic mean of the PLSR downscaling corrected monthly precipitation calculated Count times to obtain the downscaling corrected monthly precipitation raster data W 2 of the geographically weighted regression method; 步骤五：将计算Count次的GWR降尺度校正月降水量做算术平均，得到地理加权回归方法的降尺度校正月降水量栅格数据W3；Step 5: Perform arithmetic mean of the GWR downscaled corrected monthly precipitation calculated Count times to obtain the downscaled corrected monthly precipitation raster data W 3 of the geographically weighted regression method; 步骤六：根据气象台站的地理坐标，匹配步骤三、步骤四和步骤五生成的降水量数据W1、W2、W3和气象台站的空间位置X，提取气象台站所在栅格的降尺度校正得到的月降水量Y，计算逐月实测降水数据obs与数据Y间的决定系数Rj 2、均方根误差RMSEj和平均相对误差AREj；Step 6: According to the geographical coordinates of the meteorological station, match the precipitation data W 1 , W 2 , W 3 generated in Step 3, Step 4 and Step 5 with the spatial position X of the meteorological station, and extract the downscaling correction of the grid where the meteorological station is located For the obtained monthly precipitation Y, calculate the coefficient of determination R j 2 , the root mean square error RMSE j and the average relative error ARE j between the monthly measured precipitation data obs and the data Y; 步骤七：完成多元线性回归和地理加权回归方法的交叉验证；Step 7: Complete the cross-validation of multiple linear regression and geographically weighted regression methods; 步骤八：根据交叉验证结果，以“星-地”融合的月降水数据D为因变量，以DEM数据G，坡度数据H，坡向数据J，NDVI数据K，经度数据L，纬度数据M为自变量，使用最优化方法对研究区月降水量进行降尺度校正，得到校正后的月降水量数据Z，并且提取最优回归关系中因变量与高程的回归系数AA；Step 8: According to the cross-validation results, the monthly precipitation data D fused by "star-ground" is used as the dependent variable, and the DEM data G, slope data H, aspect data J, NDVI data K, longitude data L, and latitude data M are used as the dependent variable. For the independent variable, use the optimization method to downscale the monthly precipitation in the study area, obtain the corrected monthly precipitation data Z, and extract the regression coefficient AA of the dependent variable and the elevation in the optimal regression relationship; 步骤九：将校正后的逐月降水量数据Z、降水量与高程的回归系数AA转换为栅格图片，得到研究区逐月校正降水量和月降水量沿高程变化的梯度栅格图；Step 9: Convert the corrected monthly precipitation data Z, the regression coefficient AA of precipitation and elevation into raster images, and obtain the gradient raster images of monthly corrected precipitation and monthly precipitation changes along the elevation in the study area; 步骤十：用研究区的矢量边界批量裁剪逐月降水量和其沿高程变化的梯度栅格图，得到研究区1～12月份平均降水量，以及研究区月均降水量沿高程的梯度变化。Step 10: Batch crop the gradient raster map of monthly precipitation and its change along the elevation with the vector boundary of the study area to obtain the average precipitation in the study area from January to December, and the gradient change of the average monthly precipitation in the study area along the elevation. 3.根据权利要求2所述的山区卫星降水数据的降尺度校正方法，其特征在于，Ⅳ.交叉验证与降尺度校正执行，步骤六中决定系数Rj 2、均方根误差RMSEj和平均相对误差AREj的计算公式分别为：3. The down-scaling correction method of satellite precipitation data in mountainous areas according to claim 2, is characterized in that, IV. cross-validation and down-scaling correction are performed, in step 6, determination coefficient R j 2 , root mean square error RMSE j and average The calculation formulas of the relative error ARE j are: 式中：Count为实测站点个数，obsi为第i个站点的实测降水量，为所有实测站点的平均降水量，Yi为第i个站点的降尺度校正降水量，Rj 2、RMSEj和AREj分别为第j次验证的决定系数、均方根误差和平均相对偏差。In the formula: Count is the number of measured stations, obs i is the measured precipitation of the ith station, is the average precipitation of all measured stations, Yi is the downscaled corrected precipitation of the ith station, R j 2 , RMSE j and ARE j are the coefficient of determination, root mean square error and average relative deviation of the jth verification, respectively . 4.根据权利要求3所述的山区卫星降水数据的降尺度校正方法，其特征在于，Ⅳ.交叉验证与降尺度校正执行，步骤七中交叉验证遵从的原则为：4. the downscaling correction method of mountain satellite precipitation data according to claim 3 is characterized in that, Ⅳ. cross-validation and down-scaling correction are performed, and the principle that cross-validation follows in step 7 is:
CN201811084852.8A 2018-09-18 2018-09-18 A kind of NO emissions reduction bearing calibration of mountain area satellite precipitation data Expired - Fee Related CN109375294B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811084852.8A CN109375294B (en) 2018-09-18 2018-09-18 A kind of NO emissions reduction bearing calibration of mountain area satellite precipitation data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811084852.8A CN109375294B (en) 2018-09-18 2018-09-18 A kind of NO emissions reduction bearing calibration of mountain area satellite precipitation data

Publications (2)

Publication Number Publication Date
CN109375294A true CN109375294A (en) 2019-02-22
CN109375294B CN109375294B (en) 2019-11-05

Family

ID=65404861

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811084852.8A Expired - Fee Related CN109375294B (en) 2018-09-18 2018-09-18 A kind of NO emissions reduction bearing calibration of mountain area satellite precipitation data

Country Status (1)

Country Link
CN (1) CN109375294B (en)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110118982A (en) * 2019-04-12 2019-08-13 大连理工大学 A kind of satellite precipitation data bearing calibration based on space optimization interpolation
CN110597873A (en) * 2019-08-23 2019-12-20 北京师范大学 Precipitation data estimation method, device, equipment and storage medium
CN110738252A (en) * 2019-10-14 2020-01-31 广州地理研究所 Machine learning satellite precipitation data downscaling method and system for spatial autocorrelation
CN110865425A (en) * 2019-11-28 2020-03-06 中国有色金属长沙勘察设计研究院有限公司 Rain gauge gross error detection method based on prior information
CN111078678A (en) * 2019-12-18 2020-04-28 中国气象局乌鲁木齐沙漠气象研究所 Satellite precipitation data correction method based on multi-source information fusion and scale reduction
CN112699951A (en) * 2021-01-06 2021-04-23 中国气象局乌鲁木齐沙漠气象研究所 Precipitation data fusion method and device, terminal equipment and readable storage medium
CN112989557A (en) * 2021-01-14 2021-06-18 中国空间技术研究院 Method for improving water reserve change prediction reliability based on neural network selectable model
CN114281870A (en) * 2022-01-07 2022-04-05 青海省水文水资源测报中心 Method for analyzing space of characteristic of refined rainfall time variation
CN114463616A (en) * 2022-01-28 2022-05-10 国能大渡河流域水电开发有限公司 Multi-source satellite precipitation fusion method based on Stacking and EMOS-CSG
CN115166874A (en) * 2022-07-13 2022-10-11 北京师范大学 Meteorological drought index SPI construction method based on machine learning
CN116187159A (en) * 2023-04-11 2023-05-30 郑州大学 A Downscaling Method for Remote Sensing Precipitation Data by Coupling Machine Learning and GWR Correction
CN116228046A (en) * 2023-05-09 2023-06-06 成都信息工程大学 Mountain area space precipitation estimation method based on satellite remote sensing and geographic data
CN116910041A (en) * 2023-06-21 2023-10-20 中国水利水电科学研究院 A daily correction method for remote sensing precipitation products based on scale analysis

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100040260A1 (en) * 2007-12-20 2010-02-18 Image Tree Corp. Remote sensing and probabilistic sampling based method for determining the carbon dioxide volume of a forest
CN104820754A (en) * 2015-05-13 2015-08-05 南京信息工程大学 Space statistical downscaling rainfall estimation method based on geographical difference analysis method
CN105160192A (en) * 2015-09-17 2015-12-16 浙江大学 TRMM (Tropical Rainfall Measuring Mission) satellite rainfall data downscaling method based on M5-Local
CN106776481A (en) * 2016-11-29 2017-05-31 河海大学 A kind of NO emissions reduction bearing calibration for acting on satellite precipitation data
CN107608939A (en) * 2017-08-16 2018-01-19 北京师范大学 TRMM precipitation data NO emissions reduction methods based on high resolution satellite remote sensing image

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100040260A1 (en) * 2007-12-20 2010-02-18 Image Tree Corp. Remote sensing and probabilistic sampling based method for determining the carbon dioxide volume of a forest
CN104820754A (en) * 2015-05-13 2015-08-05 南京信息工程大学 Space statistical downscaling rainfall estimation method based on geographical difference analysis method
CN105160192A (en) * 2015-09-17 2015-12-16 浙江大学 TRMM (Tropical Rainfall Measuring Mission) satellite rainfall data downscaling method based on M5-Local
CN106776481A (en) * 2016-11-29 2017-05-31 河海大学 A kind of NO emissions reduction bearing calibration for acting on satellite precipitation data
CN107608939A (en) * 2017-08-16 2018-01-19 北京师范大学 TRMM precipitation data NO emissions reduction methods based on high resolution satellite remote sensing image

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110118982B (en) * 2019-04-12 2022-11-25 大连理工大学 A Correction Method of Satellite Precipitation Data Based on Spatial Optimal Interpolation
CN110118982A (en) * 2019-04-12 2019-08-13 大连理工大学 A kind of satellite precipitation data bearing calibration based on space optimization interpolation
CN110597873A (en) * 2019-08-23 2019-12-20 北京师范大学 Precipitation data estimation method, device, equipment and storage medium
CN110738252A (en) * 2019-10-14 2020-01-31 广州地理研究所 Machine learning satellite precipitation data downscaling method and system for spatial autocorrelation
CN110865425A (en) * 2019-11-28 2020-03-06 中国有色金属长沙勘察设计研究院有限公司 Rain gauge gross error detection method based on prior information
CN110865425B (en) * 2019-11-28 2021-10-26 中国有色金属长沙勘察设计研究院有限公司 Rain gauge gross error detection method based on prior information
CN111078678A (en) * 2019-12-18 2020-04-28 中国气象局乌鲁木齐沙漠气象研究所 Satellite precipitation data correction method based on multi-source information fusion and scale reduction
CN111078678B (en) * 2019-12-18 2021-03-23 中国气象局乌鲁木齐沙漠气象研究所 Satellite precipitation data correction method based on multi-source information fusion and scale reduction
CN112699951A (en) * 2021-01-06 2021-04-23 中国气象局乌鲁木齐沙漠气象研究所 Precipitation data fusion method and device, terminal equipment and readable storage medium
CN112699951B (en) * 2021-01-06 2023-02-03 中国气象局乌鲁木齐沙漠气象研究所 Precipitation data fusion method and device, terminal equipment and readable storage medium
CN112989557A (en) * 2021-01-14 2021-06-18 中国空间技术研究院 Method for improving water reserve change prediction reliability based on neural network selectable model
CN114281870A (en) * 2022-01-07 2022-04-05 青海省水文水资源测报中心 Method for analyzing space of characteristic of refined rainfall time variation
CN114463616A (en) * 2022-01-28 2022-05-10 国能大渡河流域水电开发有限公司 Multi-source satellite precipitation fusion method based on Stacking and EMOS-CSG
CN114463616B (en) * 2022-01-28 2024-04-12 国能大渡河流域水电开发有限公司 Multi-source satellite precipitation fusion method based on Stacking and EMOS-CSG
CN115166874A (en) * 2022-07-13 2022-10-11 北京师范大学 Meteorological drought index SPI construction method based on machine learning
CN116187159A (en) * 2023-04-11 2023-05-30 郑州大学 A Downscaling Method for Remote Sensing Precipitation Data by Coupling Machine Learning and GWR Correction
CN116187159B (en) * 2023-04-11 2024-10-29 郑州大学 Remote sensing precipitation data downscaling method coupling machine learning and GWR correction
CN116228046A (en) * 2023-05-09 2023-06-06 成都信息工程大学 Mountain area space precipitation estimation method based on satellite remote sensing and geographic data
CN116910041A (en) * 2023-06-21 2023-10-20 中国水利水电科学研究院 A daily correction method for remote sensing precipitation products based on scale analysis
CN116910041B (en) * 2023-06-21 2023-12-22 中国水利水电科学研究院 Daily correction method for remote sensing precipitation product based on scale analysis

Also Published As

Publication number Publication date
CN109375294B (en) 2019-11-05

Similar Documents

Publication Publication Date Title
CN109375294B (en) A kind of NO emissions reduction bearing calibration of mountain area satellite precipitation data
US7305983B1 (en) Assessment of solar energy potential on existing buildings in a region
Ninyerola et al. Monthly precipitation mapping of the Iberian Peninsula using spatial interpolation tools implemented in a Geographic Information System
Balázs et al. Simulation of the mean urban heat island using 2D surface parameters: empirical modelling, verification and extension
Hassan Climate change impact on groundwater recharge of Umm er Radhuma unconfined aquifer Western Desert, Iraq
CN115952702B (en) A Calculation Method of Forest NEP Based on FORCCHN Model and Remote Sensing Data
CN108959661A (en) 3S technology-based soil nutrient grade classification map generation method and precision evaluation method thereof
CN110440762B (en) Gridding image control point layout method for multi-rotor unmanned aerial vehicle mountainous area aerial survey image
King et al. Actual evapotranspiration estimates for Australia: Intercomparison and evaluation
Effat Mapping solar energy potential zones, using SRTM and spatial analysis, application in Lake Nasser Region, Egypt
Bhandari et al. Land capability classification and crop suitability assessment in a watershed using RS and GIS—a case study of watershed in Dehradun, Uttarakhand
CN106156756B (en) A Fast Estimation Method for Spatial Distribution of Construction Land Efficiency
Ly et al. Flood mapping along the lower Mekong River in Cambodia
Dhital et al. Rainfall-runoff simulation of Bagmati River basin, Nepal
Zhang et al. Reconstruction of cropland spatial patterns and its spatiotemporal changes over the 20th century on the Songnen Plain, Northeast China
Susilowati et al. Micro hydropower plant potential study based on Landsat 8 operational land imager satellite data
CN116629436A (en) A Method for Ecological Space Network Optimization
Jabal et al. Integrated Approach for Land Surface Temperature Assessment in Different Topography of Iraq
Al Rammahi Specification of Actual Evapotranspiration in Iraq by GIS
De Blasi Scale dependence of hydrological effects from different climatic conditions on glacierized catchments
Wang et al. Analysis of land use type in Dadukou District of Chongqing using remote sensing image inversion
Shamkhi et al. Investigation of Ahrar Irrigation Scheme Characteristics Utilizing Remote Sensing Technique
Fu et al. Assessing Terrain Effect on Evapotranspiration Using the SEBAL Model in Mountainous Regions
Huang et al. The potential impacts of climate and forest changes on streamflow for micro-, meso-and macro-scale catchments in Norway
Fu et al. Study on the Relationship Between Physical Structural Integrality and Ecosystem Services Value in the Riparian Zone

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
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20191105

CF01 Termination of patent right due to non-payment of annual fee