CN116011349B - Near-surface air temperature estimation method - Google Patents
Near-surface air temperature estimation method Download PDFInfo
- Publication number
- CN116011349B CN116011349B CN202310210569.XA CN202310210569A CN116011349B CN 116011349 B CN116011349 B CN 116011349B CN 202310210569 A CN202310210569 A CN 202310210569A CN 116011349 B CN116011349 B CN 116011349B
- Authority
- CN
- China
- Prior art keywords
- air temperature
- surface air
- data
- model
- day
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 41
- 230000005855 radiation Effects 0.000 claims description 28
- 238000007637 random forest analysis Methods 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 4
- 238000002790 cross-validation Methods 0.000 claims description 3
- 230000002354 daily effect Effects 0.000 claims 7
- 238000004364 calculation method Methods 0.000 claims 1
- 230000003203 everyday effect Effects 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000011156 evaluation Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention provides a near-surface air temperature estimation method, which relates to the technical field of surface air temperature estimation and comprises the following steps: collecting observation data of a site, and uploading the collected observation data information to a cloud platform; extracting network data corresponding to the observed data from different data sets based on the cloud platform; forming a point scale training data set through the observation data and the network data; constructing a near-surface air temperature training sub-model 1 and a near-surface air temperature training sub-model 2; combining the near-surface air temperature training sub-model 1 and the near-surface air temperature training sub-model 2 to construct a trained near-surface air temperature estimation model; and carrying out surface scale air temperature estimation on the input grid data set through the near-surface air temperature estimation model to obtain 1-km daily seamless near-surface air temperature. The invention realizes the rapid estimation of the near-surface air temperature in a long time sequence and a large range, has high efficiency and high speed in producing near-surface air temperature data, has no missing pixels in space, and can realize the time sequence intensive monitoring application in daily resolution.
Description
Technical Field
The invention belongs to the technical field of surface air temperature estimation, and particularly relates to a near-surface air temperature estimation method.
Background
The near-surface air temperature has important significance for climate change, agriculture and forestry production, ecological benefit evaluation, urban heat comfort level and the like.
Currently, the main approaches to obtain near-surface air temperature are: 1) Based on site observations; 2) Inversion is performed based on remote sensing observation; 3) Numerical simulation or re-analysis of the product. The space is not connected due to the fact that site observation distribution is sparse and lacks of space distribution when site observation is performed, space is not connected due to the fact that cloud pollution and the like are frequently caused when inversion is performed on the basis of remote sensing observation, and when the near-surface air temperature is obtained through numerical simulation or re-analysis products, the obtained space scale is large, and the accuracy of the obtained near-surface air temperature is affected.
In summary, the existing method and product for obtaining the near-surface air temperature generally have the condition of low resolution or thicker time resolution, or have the condition of only containing a single air temperature type (average value or air value) and having the condition of lack of the near-surface air temperature in space, so that the accuracy of the obtained near-surface air temperature is not high.
Disclosure of Invention
The invention specifically provides the following technical scheme: a method of estimating near-surface air temperature, comprising the steps of:
collecting observation data of a site, and uploading the collected observation data information to a cloud platform;
the cloud platform extracts network data corresponding to the observed data from a plurality of network data sets;
forming a point scale training data set through the observation data and the network data;
constructing a near-surface air temperature training sub-model 1 based on GLDAS data and a random forest model;
constructing a near-surface air temperature training sub-model 2 based on ERA5-Land data and a random forest model;
combining the near-surface air temperature training sub-model 1 with the near-surface air temperature training sub-model 2 to construct a trained near-surface air temperature estimation model;
carrying out space-time difference on the data in the point scale training data set to obtain a grid data set;
and inputting the data in the grid data set into a near-surface air temperature estimation model to obtain the near-surface air temperature which is seamless day by day under the set resolution.
Preferably, the cloud platform extracts network data corresponding to the observed data from a plurality of network data sets, including:
extracting the surface temperature, downward short wave radiation, downward long wave radiation, relative humidity, rainfall rate and wind speed of the time corresponding to the position of the observed value from the GLDAS2.1 data set;
extracting the surface temperature, downward short wave radiation, downward long wave radiation, relative humidity, rainfall rate and wind speed of the time corresponding to the position corresponding to the observed value from the ERA5-Land data set;
extracting enhanced vegetation index data corresponding to the time corresponding to the position from the MODIS data set; extracting elevation information from NASA SRTM DEM data;
demographic data is extracted from the GPWv4 dataset.
Preferably, the point-scale training dataset comprises: longitude LONG, latitude LAT, YEAR, MONTH, DAY, enhanced vegetation index EVI, digital elevation model DEM, demographic data Pop extracted from the GPWv4 dataset; extracting the surface temperature LSTg, the downward short wave radiation SWg, the downward long wave radiation LWg, the relative humidity RHg, the rainfall rate Pg and the wind speed WSg of the time corresponding to the position of the observed value from the GLDAS2.1 data set; and extracting the surface temperature LSTe, the downward short wave radiation SWe, the downward long wave radiation LWe, the relative humidity RHe, the rainfall rate Pe and the wind speed WSe of the time corresponding to the position of the observed value from the ERA5-Land data set.
Preferably, the expression of the training submodel 1 is:
NSAT1＝RF1(LONG,LAT,YEAR,MONTH,DAY,LSTg,SWg,LWg,RHg,Pg,WS g,EVI,DEM,Pop)。
preferably, the expression of the training submodel 2 is:
NSAT2＝RF2(LONG,LAT,YEAR,MONTH,DAY,LSTe,SWe,LW e ,RH e ,P e ,WS e ,
EVI,DEM,Pop)。
preferably, the estimating value of the final near-surface air temperature is output through a near-surface air temperature estimating model, and the calculating expression is as follows:
NSAT＝(NSAT1+NSAT2)/2。
preferably, a ten-fold cross validation method is adopted to evaluate the model precision, and a search method is adopted to adjust the number of trees in the random forest so as to find the optimal parameters.
Preferably, all data in the point scale training data set are normalized as follows:
V＝(V-Vmin)/(Vmax-Vmin)
where V represents the variable value at a certain position at a certain time, vmax represents the maximum value of the variable, and Vmin represents the minimum value of the variable.
Preferably, the data in the point scale training data set is subjected to space-time difference value to obtain a grid data set, which comprises the following steps:
the LONG, LAT, YEAR, MONTH, DAY, EVI, DEM, pop, LST is subjected to g 、SW g 、LW g 、RH g 、P g 、WS g 、LST e 、SW e 、LW e 、RH e 、P e 、WS e The time resolution of the data space-time interpolation is daily, the spatial resolution is 1-km, and a grid data set with the resolution of 1-km is formed.
Preferably, when the surface temperature and the site observation near-surface air temperature are daily maximum values, the method is used for estimating the daily maximum value of the near-surface air temperature of 1-km which is seamless day by day; and when the surface temperature and the site observation near-surface air temperature are the daily minimum value and the daily average value, the method is used for estimating the daily minimum value and the daily average value of the near-surface air temperature of 1-km day-by-day seamless.
Compared with the prior art, the invention has the following beneficial effects:
according to the method, a high space-time resolution near-surface air temperature estimation method is established based on a cloud platform (Google Earth Engine) through joint site observation, remote sensing data, analysis products and numerical simulation products, and a random forest based estimation method is designed based on the cloud platform, so that 1-km daily seamless near-surface air temperature estimation is realized, the method can realize long-time sequence large-range near-surface air temperature rapid estimation, the near-surface air temperature accuracy is improved, the near-surface air temperature data production efficiency is high, the speed is high, space is free of missing pixels, and daily resolution can realize time sequence intensive monitoring application.
Drawings
FIG. 1 is a flow chart of the high spatial-temporal resolution near-surface air temperature estimation of the present invention;
FIG. 2 is a graph showing the accuracy evaluation of the daily maximum value, the daily minimum value and the daily average value of near-surface air temperatures in the yellow river basin 2000-2020;
FIG. 3 is a graphical representation of near-surface air temperature data for day-by-day seamless 1-km of the present invention;
fig. 4 is a diagram of the invention in a quick extension applied to other areas.
Detailed Description
The following describes the embodiments of the present invention further with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
As shown in fig. 1, a method for estimating a near-surface air temperature includes the steps of:
s1: and collecting the observation data of the site, uploading the collected observation data to a cloud platform, and extracting network data corresponding to the observation data from a plurality of network data sets by the cloud platform.
S2: a point scale training data set is formed by the observation data and the network data.
S3: constructing a near-surface air temperature training sub-model 1 based on GLDAS data and a random forest model; and constructing a near-surface air temperature training sub-model 2 based on ERA5-Land data and a random forest model.
S4: and combining the near-surface air temperature training sub-model 1 with the near-surface air temperature training sub-model 2 to construct a near-surface air temperature estimation model.
S5: and carrying out space-time difference on the data in the point scale training data set to obtain a grid data set, and inputting the data in the grid data set into a near-surface air temperature estimation model to obtain the near-surface air temperature which is seamless day by day under the set resolution.
Before inputting the data in the grid data set into the near-surface air temperature estimation model, training the near-surface air temperature estimation model is needed, including: and inputting the data in the point scale training data set into the near-surface air temperature estimation model to obtain a trained near-surface air temperature estimation model.
Specifically, the flow of the method for rapidly estimating the near-surface air temperature with high space-time resolution is as follows:
1. collecting site observation data key information of a study area (a yellow river basin will be taken as an example for subsequent evaluation), wherein the key information comprises: near surface air temperatures (NSATobs), time of observation (i.e., YEAR (YEAR), MONTH (manth), DAY (DAY)), location of observation (i.e., longitude (LONG), latitude (LAT)), upload collected data to Google Earth Engine (GEE) cloud platform. High spatial-temporal resolution refers to the minimum time interval between two adjacent telemetry observations made in the same region.
2. Extracting a surface temperature (LSTg) of a corresponding position of the observed value corresponding to time from the GLDAS2.1 dataset based on the GEE platform, downward short wave radiation (SWg), downward long wave radiation (LWg), relative humidity (RHg), rainfall rate (Pg) and wind speed (WSg); extracting surface temperature (LSTe), downward short wave radiation (SWe), downward long wave radiation (LWe), relative humidity (RHe), rainfall rate (Pe) and wind speed (WSe) of a corresponding time of an observation position from the ERA5-Land data set; extracting Enhanced Vegetation Index (EVI) data corresponding to the position and corresponding time from the MODIS dataset; extracting elevation information (Ele) from NASA SRTM DEM data; demographic data (Pop) is extracted from the GPWv4 dataset.
3. The data extracted from steps 1 and 2 constitute a point-scale training dataset, each training data record in the training dataset being composed of an input element and an output element. Wherein the input element set comprises: LONG, LAT, YEAR, MONTH, DAY, EVI, DEM, pop, LSTg, SWg, LWg, RHg, pg, WSg, LSTe, SWe, LWe, RHe, pe, WSe the output element is NSATobs, all data are normalized as follows:
V＝(V-Vmin)/(Vmax-Vmin) (1)
where V represents the variable value at a certain position at a certain time, vmax represents the maximum value of the variable, and Vmin represents the minimum value of the variable.
4. Constructing a near-surface air temperature training sub-model 1 based on GLDAS data and a random forest model, wherein the following variables are input of the sub-model 1: longitude, latitude, year, month, day, surface temperature, downward short wave radiation, downward long wave radiation, relative humidity, rainfall rate, wind speed, enhanced vegetation index, elevation, population, wherein surface temperature, downward short wave radiation, downward long wave radiation, relative humidity, rainfall rate, wind speed is derived from GLDAS2.1 dataset, sub-model 1 can be expressed as:
NSAT1＝RF1(LONG,LAT,YEAR,MONTH,DAY,LSTg,SWg,LWg,RHg,Pg,WS g,EVI,DEM,Pop)(2)
5. constructing a near-surface air temperature training sub-model 2 based on ERA5-Land data and a random forest model, wherein the following variables are input of the sub-model 2: longitude, latitude, year, month, day, surface temperature, downward short wave radiation, downward long wave radiation, relative humidity, rainfall rate, wind speed, enhanced vegetation index, elevation, population, wherein surface temperature, downward short wave radiation, downward long wave radiation, relative humidity, rainfall rate, wind speed are derived from the ERA5-Land dataset. Sub-model 2 can be expressed as:
NSAT2＝RF2(LONG,LAT,YEAR,MONTH,DAY,LSTe,SWe,LWe,RHe,Pe,WSe,EVI,DEM,Pop)(3)
the output of the sub-models in the training process is site observation data (NSATobs), and the final near-surface air temperature estimation model is formed by combining the sub-model 1 and the sub-model 2, namely the output results (NSAT 1 and NSAT 2) of the sub-model 1 and the sub-model 2 are averaged to be used as the final near-surface air temperature estimation value (NSAT), namely
NSAT＝(NSAT1+NSAT2)/2 (4)
The model precision is evaluated by adopting a ten-fold cross validation method, and the number of trees in the random forest is adjusted by adopting a search method to find the optimal parameters.
5. And 4, using a near-surface air temperature estimation model trained based on the point scale training data set for space-time continuous near-surface air temperature estimation. Space-time interpolation is carried out on EVI, DEM, pop, LSTg, SWg, LWg, RHg, pg, WSg, LSTe, SWe, LWe, RHe, pe, WSe data (the source is the same as that of the step 2) in the research area, the time resolution is daily, the space resolution is 1-km, and a grid data set with the resolution of 1-km is formed; LONG, LAT, YEAR, MONTH, DAY is also processed into a grid data set with the resolution of 1-km, and the near-surface air temperature estimation model trained in the step 4 is utilized to carry out the face scale air temperature estimation based on the GEE cloud platform so as to obtain the 1-km daily seamless near-surface air temperature.
6. The method can be used for estimating the daily maximum value of the 1-km day-to-day seamless near-surface air temperature when the local surface temperature and the site observation near-surface air temperature are the daily maximum value and for estimating the daily minimum value and the daily average value of the 1-km day-to-day seamless near-surface air temperature when the local surface temperature and the site observation near-surface air temperature are the daily minimum value and the daily average value respectively.
The accuracy assessment of the present invention is shown in fig. 2, an exemplary spatially seamless near-surface air temperature product is shown in fig. 3, and the present invention can be rapidly extended to other areas as shown in fig. 4.
The above embodiments are merely preferred embodiments of the present invention, the protection scope of the present invention is not limited thereto, and any simple changes or equivalent substitutions of technical solutions that can be obviously obtained by those skilled in the art within the technical scope of the present invention disclosed in the present invention belong to the protection scope of the present invention.
Claims (8)
1. A method of estimating near-surface air temperature, comprising the steps of:
collecting observation data of a site, and uploading the collected observation data to a cloud platform;
the cloud platform extracts grid data corresponding to the observed data from a plurality of grid data sets;
forming a point scale training data set aiming at the maximum value, the minimum value and the average value of the daily near-surface air temperature through the observation data and the grid data;
constructing a near-surface air temperature training sub-model 1 based on a random forest model by utilizing GLDAS surface temperature and auxiliary data;
constructing a near-surface air temperature training sub-model 2 based on a random forest model by using ERA5-Land surface temperature and auxiliary data;
combining the near-surface air temperature training sub-model 1 with the near-surface air temperature training sub-model 2, and constructing a near-surface air temperature estimation model to realize high-precision air temperature estimation;
performing space-time interpolation on data corresponding to input variables in the point scale training data set to obtain a grid data set with 1-km of resolution per day and seamless space;
the point scale training dataset comprises: longitude LONG, latitude LAT, YEAR, MONTH, DAY, enhanced vegetation index EVI, digital elevation model DEM, demographic data Pop extracted from the GPWv4 dataset; extracting time corresponding to the position of the observed value from the GLDAS2.1 data setSurface temperature LST g Downward short wave radiation SW g Downward long wave radiation LW g Relative humidity RH g Rate of rainfall P g Wind speed WS g The method comprises the steps of carrying out a first treatment on the surface of the Extracting surface temperature LST of time corresponding to observation value corresponding to position from ERA5-Land data set e Downward short wave radiation SW e Downward long wave radiation LW e Relative humidity RH e Rate of rainfall P e And wind speed WS e ；
The LONG, LAT, YEAR, MONTH, DAY, EVI, DEM, pop, LST is subjected to g 、SW g 、LW g 、RH g 、P g 、WS g 、LST e 、SW e 、LW e 、RH e 、P e 、WS e The data space-time interpolation is a grid data set which is day by day, has the spatial resolution of 1-km and is seamless in space;
and inputting the data in the grid data set into a near-surface air temperature estimation model to obtain a maximum value, a minimum value and a mean value of the near-surface air temperature which are seamless every day under a set resolution.
2. The method of estimating a near-surface air temperature according to claim 1, wherein the cloud platform extracts mesh data corresponding to the observation data from a plurality of mesh data sets, comprising:
extracting the surface temperature, downward short wave radiation, downward long wave radiation, relative humidity, rainfall rate and wind speed of the time corresponding to the position of the observed value from the GLDAS2.1 data set;
extracting the surface temperature, downward short wave radiation, downward long wave radiation, relative humidity, rainfall rate and wind speed of the time corresponding to the position corresponding to the observed value from the ERA5-Land data set;
extracting enhanced vegetation index data corresponding to the time corresponding to the position from the MODIS data set; extracting elevation information from NASA SRTM DEM data;
demographic data is extracted from the GPWv4 dataset.
3. The method for estimating a near-surface air temperature according to claim 1, wherein the expression of the near-surface air temperature training submodel 1 is:
NSAT1＝RF1(LONG,LAT,YEAR,MONTH,DAY,LST g ,SW g ,LW g ,RH g ,P g ,WS g ,EVI,DEM,Pop)。
4. a method of estimating a near-surface air temperature as claimed in claim 3, wherein said near-surface air temperature training submodel 2 has the expression:
NSAT2＝RF2(LONG,LAT,YEAR,MONTH,DAY,LST e ,SW e ,LW e ,RH e ,P e ,WS e ,EVI,DEM,Pop)。
5. the method of estimating a near-surface air temperature according to claim 4, wherein the estimated value of the final near-surface air temperature is outputted by a near-surface air temperature estimation model, and the calculation expression is as follows:
NSAT＝(NSAT1+NSAT2)/2。
6. the method for estimating near-surface air temperature according to claim 1, wherein a ten-fold cross-validation method is adopted to evaluate model accuracy, and a search method is adopted to adjust the number of trees in a random forest to find optimal parameters.
7. A method of estimating a near-surface air temperature as claimed in claim 1 wherein all data in said point-scale training dataset is normalized as follows:
V＝(V-Vmin)/(Vmax-Vmin)
where V represents the variable value at a certain position at a certain time, vmax represents the maximum value of the variable, and Vmin represents the minimum value of the variable.
8. The method for estimating a near-surface air temperature according to claim 1, wherein when the surface temperature and the site observation near-surface air temperature are daily maximum values, the near-surface air temperature daily maximum value for 1-km day-by-day seamless is estimated; and when the surface temperature and the site observation near-surface air temperature are the daily minimum value and the daily average value, the method is used for estimating the daily minimum value and the daily average value of the near-surface air temperature of 1-km day-by-day seamless.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310210569.XA CN116011349B (en) | 2023-03-07 | 2023-03-07 | Near-surface air temperature estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310210569.XA CN116011349B (en) | 2023-03-07 | 2023-03-07 | Near-surface air temperature estimation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116011349A CN116011349A (en) | 2023-04-25 |
| CN116011349B true CN116011349B (en) | 2024-01-19 |
Family
ID=86035790
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310210569.XA Active CN116011349B (en) | 2023-03-07 | 2023-03-07 | Near-surface air temperature estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116011349B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112016052A (en) * | 2020-08-20 | 2020-12-01 | 广东省气象探测数据中心 | Near-surface daily maximum air temperature estimation method, system and terminal based on multi-source data |
| CN114662701A (en) * | 2022-03-21 | 2022-06-24 | 中山大学 | Method for estimating near-surface air temperature by using remote sensing |
| CN115639979A (en) * | 2022-09-15 | 2023-01-24 | 河南大学 | High-resolution SPEI data set development method based on random forest regression model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106909722B (en) * | 2017-02-10 | 2019-07-26 | 广西壮族自治区气象减灾研究所 | A kind of accurate inversion method of large area of temperature near the ground |
-
2023
- 2023-03-07 CN CN202310210569.XA patent/CN116011349B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112016052A (en) * | 2020-08-20 | 2020-12-01 | 广东省气象探测数据中心 | Near-surface daily maximum air temperature estimation method, system and terminal based on multi-source data |
| CN114662701A (en) * | 2022-03-21 | 2022-06-24 | 中山大学 | Method for estimating near-surface air temperature by using remote sensing |
| CN115639979A (en) * | 2022-09-15 | 2023-01-24 | 河南大学 | High-resolution SPEI data set development method based on random forest regression model |
Non-Patent Citations (1)
| Title |
|---|
| Use of Google Earth Engine to Generate a 20-Year 1 Km× 1 Km Monthly Air Temperature Product Over Yellow River Basin;Meiling Gao 等;IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING;第14卷;正文第10079-10090页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116011349A (en) | 2023-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111727443B (en) | Soil available nutrient inversion method based on deep neural network | |
| Song et al. | Modeling spatio-temporal distribution of soil moisture by deep learning-based cellular automata model | |
| CN111210082B (en) | Precipitation prediction method based on optimized BP neural network algorithm | |
| CN111027175B (en) | Method for evaluating social and economic influences of flood based on coupling model integrated simulation | |
| CN110751094B (en) | Crop yield estimation method based on GEE comprehensive remote sensing image and deep learning method | |
| CN108536908B (en) | Method for evaluating watershed water environment safety based on non-point source nitrogen and phosphorus loss risk | |
| CN109344865B (en) | Data fusion method for multiple data sources | |
| Sekertekin et al. | Short-term air temperature prediction by adaptive neuro-fuzzy inference system (ANFIS) and long short-term memory (LSTM) network | |
| CN101699477A (en) | Neural network method for accurately predicting dam deformation | |
| CN112016696B (en) | PM integrating satellite observation and ground observation 1 Concentration inversion method and system | |
| CN112765912A (en) | Evaluation method for social and economic exposure degree of flood disasters based on climate mode set | |
| Yu et al. | A spatial–temporal graph attention network approach for air temperature forecasting | |
| CN111062526A (en) | Winter wheat yield per unit prediction method and system | |
| CN115345076A (en) | Wind speed correction processing method and device | |
| CN112100922A (en) | Wind resource prediction method based on WRF and CNN convolutional neural network | |
| CN115640956A (en) | Future water resource supply and demand balance analysis method | |
| CN116011349B (en) | Near-surface air temperature estimation method | |
| CN116050460B (en) | Air temperature data spatial interpolation method based on attention neural network | |
| CN116522145B (en) | Drought prediction method considering space-time constraint and vegetation condition | |
| CN116432849B (en) | Leaf area index prediction method based on primary productivity data assimilation | |
| Nasseri et al. | A comparison between direct and indirect frameworks to evaluate impacts of climate change on streamflows: case study of Karkheh River basin in Iran | |
| CN113886763A (en) | Method for estimating forest fire frequency and intensity based on satellite microwave index and meteorological index | |
| CN114239274A (en) | Method for calculating root layer soil water by multi-source remote sensing data driven index filtering model | |
| CN115166866A (en) | Citrus disease and insect pest occurrence forecasting method and system based on lattice point meteorological data | |
| Mohan et al. | Crop production rate estimation using parallel layer regression with deep belief network |
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 |