CN-121997088-A - Crop actual evapotranspiration calculation method and device, electronic equipment and storage medium

CN121997088ACN 121997088 ACN121997088 ACN 121997088ACN-121997088-A

Abstract

The application relates to the technical field of agricultural irrigation, in particular to a crop actual evapotranspiration calculation method, a device, electronic equipment and a storage medium, which are used for establishing space-time units based on objective division rules and constructing stable standard cold and hot pixel sets for each unit by utilizing historical data, therefore, the pixel selection process relying on subjective experience in the traditional SEBAL model is converted into a standardized and data-driven process, and the standard set of the corresponding unit of the day to be calculated is automatically matched and called for calculation during calculation, so that the deviation caused by artificial subjective judgment and random selection is avoided. Therefore, the calculation result has higher precision, stability and repeatability, meanwhile, the comparability of calculation results in different time and different areas is ensured, and the scientificity and reliability of agricultural water management are obviously improved.

Inventors

Sun Tenghui
TIAN JINGGUO
SUN SHENGJUN
HUO LIRONG
CHEN YING
GUO DONGHAO
XU ZHONGHAO

Assignees

中恒瑞景(北京)生态科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260121

Claims (10)

1. A method for calculating actual transpiration of a crop, the method comprising: Dividing the target area into a plurality of space-time units based on objective dividing rules; Constructing a reference cold pixel set and a reference hot pixel set which represent the stable surface thermal characteristics of the space-time units by utilizing multi-source data of a historical period corresponding to the space-time units aiming at each space-time unit; According to remote sensing data of a day to be calculated and the acquired time and space coverage, determining a target space-time unit corresponding to the day to be calculated; and calling a target reference cold pixel set and a target reference hot pixel set corresponding to the target space-time unit, combining the remote sensing data and the meteorological data of the day to be calculated, and calculating based on a ground surface energy balance model to obtain the actual crop evaporation quantity.
2. The method of claim 1, wherein the step of dividing the target region into a plurality of spatio-temporal elements based on objective division rules comprises: And dividing the target area into a plurality of space-time units by adopting a standard gas-saving sequence as a time division basis and adopting a standard mapping frame as a space division basis.
3. The method of claim 1, wherein the step of constructing a set of reference cold pixels and a set of reference hot pixels characterizing stable surface thermal characteristics of the spatiotemporal unit using multi-source data of historical time periods corresponding to the spatiotemporal unit comprises: Acquiring a remote sensing image set, a meteorological site data set and digital elevation model data which cover a target area history period, wherein the remote sensing image set comprises a plurality of groups of image data, and each group of image data comprises earth surface reflectivity data and earth surface temperature data; Matching the remote sensing image set, the weather site data set and the digital elevation model data to the corresponding space-time units according to time and space information; For each group of image data matched into each time-space unit, combining weather site data and digital elevation model data which are matched with the current group of image data in a time-space mode, and respectively extracting an initial cold pixel and an initial hot pixel through a preset threshold condition; Screening the initial cold pixels based on the initial cold pixels and a first temperature threshold value for each meteorological site to obtain site cold pixel masks of the meteorological sites; meanwhile, screening the initial thermal pixels based on the initial thermal pixels and a second temperature threshold to obtain site thermal pixel masks of the meteorological sites; Combining all site cold image element masks of the meteorological sites to obtain candidate cold image element sets of the group of images; Summarizing candidate cold pixel sets and initial hot pixel sets of all groups of image data in the same space-time unit to form an initial cold and hot pixel data set of the space-time unit; And carrying out optimization processing on the initial cold and hot pixel data set to generate the reference cold pixel set and the reference hot pixel set.
4. A method according to claim 3, wherein the step of extracting the initial cold pixels and the initial hot pixels by predetermined threshold conditions respectively in combination with weather site data and digital elevation model data that are spatially-temporally matched with the current set of image data comprises: Generating a terrain screening mask based on weather site data and digital elevation model data which are matched with the current image data in a space-time mode so as to define a pixel area which accords with preset terrain conditions; In the pixel region, respectively determining a first vegetation index threshold, a first albedo threshold, a second vegetation index range and a second albedo range based on the surface reflectivity data; For each pixel in the terrain screening mask, if the vegetation index corresponding to the pixel is greater than or equal to a target vegetation index threshold determined based on a first vegetation index threshold and a first preset parameter, and the albedo corresponding to the pixel is less than or equal to a pixel of a target albedo threshold determined based on the first albedo threshold and a second preset parameter, determining the pixel as the initial cold pixel; And if the vegetation index corresponding to the pixel is in a second vegetation index range and the albedo is in the pixel in the second albedo range, determining the pixel as the initial thermal pixel.
5. The method of claim 2, wherein the step of optimizing the initial cold and hot image element dataset comprises: Acquiring land utilization type data corresponding to the pixels based on the geographic positions of the pixels in the initial cold and hot pixel data set; Removing pixels which are different from a preset hot pixel typical surface type of a preset cold pixel typical surface type in the initial cold and hot pixel data set to obtain a first cold and hot pixel data set; clustering cold pixel data and hot pixel data in the first cold and hot pixel data set by adopting a clustering algorithm to obtain a plurality of clusters; removing pixels contained in non-representative clusters according to a preset rule according to the quantity proportion of pixels contained in each cluster, and obtaining a second cold and hot pixel data set; Respectively analyzing the spatial distribution densities of cold pixel data and hot pixel data in the second cold and hot pixel data set by using a nuclear density estimation method; And removing cold pixels and hot pixels distributed in a low-density area based on the spatial distribution density to generate the reference cold pixel set and the reference hot pixel set.
6. The method of claim 5, wherein the step of removing pixels included in the non-representative clusters according to a predetermined rule to obtain the second cold and hot pixel data set according to a number ratio of pixels included in each cluster, comprises: if the statistical duty ratio of the single cluster exceeds a first preset threshold, only the cluster is reserved; If the statistical duty ratio of all the cluster memories is smaller than or equal to the first preset threshold value, at least one cluster meeting the condition is reserved according to the comparison result of the statistical duty ratio of each cluster and the second preset threshold value.
7. The method of claim 1, wherein the step of calling the target reference cold pixel set and the target reference hot pixel set corresponding to the target space-time unit, combining the remote sensing data and the meteorological data of the day to be calculated, and calculating the actual evaporation quantity of the crop based on a ground surface energy balance model comprises the following steps: Based on the remote sensing data and the meteorological data of the day to be calculated, calculating and obtaining the instantaneous surface temperature, the instantaneous net radiation, the instantaneous soil heat flux and the instantaneous aerodynamic impedance of each target pixel in the target space-time unit; calculating the instantaneous sensible heat flux of each target pixel according to the surface temperature statistic value of the target reference cold pixel set and the target reference hot pixel set, the energy balance parameter statistic value corresponding to the target reference hot pixel set, and the instantaneous surface temperature and the instantaneous aerodynamic impedance of each target pixel; calculating the instantaneous latent heat flux of each target pixel based on the instantaneous net radiation amount, the instantaneous soil heat flux and the instantaneous sensible heat flux of each target pixel; And calculating the actual evaporation quantity of the crops with the daily scale by combining the instantaneous latent heat flux, the instantaneous net radiation quantity, the instantaneous soil heat flux and the daily accumulated net radiation data.
8. A crop actual transpiration calculation device, said device comprising: The dividing module is used for dividing the target area into a plurality of space-time units based on objective dividing rules; The construction module is used for constructing a reference cold pixel set and a reference hot pixel set which characterize the stable surface thermal characteristics of the space-time units by utilizing multi-source data of the historical period corresponding to the space-time units aiming at each space-time unit; The determining module is used for determining a target space-time unit corresponding to the day to be calculated according to the acquisition time and the space coverage of the remote sensing data of the day to be calculated; And the calculation module is used for calling the target reference cold pixel set and the target reference hot pixel set corresponding to the target space-time unit, combining the remote sensing data and the meteorological data of the day to be calculated, and calculating based on a ground surface energy balance model to obtain the actual crop evaporation quantity.
9. An electronic device comprising a memory for storing a computer program and a processor that runs the computer program to cause the electronic device to perform the method of any one of claims 1 to 7.
10. A storage medium having stored therein computer program instructions which, when read and executed by a processor, perform the method of any of claims 1 to 7.

Description

Crop actual evapotranspiration calculation method and device, electronic equipment and storage medium Technical Field The invention relates to the technical field of agricultural irrigation, in particular to a crop actual evapotranspiration calculation method and device, electronic equipment and a storage medium. Background The actual evaporation of crops (Actual evapotranspiration, ETa) is a main way of agricultural water consumption, and the actual evaporation of crops is accurately calculated, so that the method is a core basis for agricultural water fine management, irrigation decision optimization and efficient utilization of regional water resources. The traditional field measurement method is high in cost and difficult to popularize in a large range, so that the regional scale calculation model based on remote sensing and meteorological data has become the mainstream of current research and application. The existing remote sensing calculation method mainly comprises a statistical model method and an energy balance model method. The statistical model method establishes the statistical relationship between the evaporation quantity and the multi-source data through means such as machine learning, but model parameters of the statistical model method often depend on training data of specific areas, so that the space-time universality is poor, and the stable large-scale business monitoring is difficult to support. The energy balance model rule is based on the principle of surface energy balance, and the physical mechanism is more definite, wherein the SEBAL model is one of representative algorithms. The model calibrates the energy distribution process by defining "cold pels" and "hot pels", but its calibration accuracy is highly dependent on manual or semi-automatic selection of the two types of feature pels. Due to the lack of objective and unified selection standards, the process has obvious randomness and subjectivity, so that the inversion result of the model is unstable, the result comparability between different time phases or areas is poor, and the reliability of the model in high-precision and business application is severely restricted. Therefore, how to overcome the dependence of the existing SEBAL model on subjective experience, and establish an objective, stable and portable cold and hot pixel calibration standard becomes a key problem for improving the actual evapotranspiration calculation accuracy and practicality of regional crops. Disclosure of Invention In view of the above, an object of the present invention is to provide a method and apparatus for calculating actual transpiration of crops, an electronic device, and a storage medium. In a first aspect, an embodiment of the present invention provides a method for calculating actual evapotranspiration of a crop, where the method includes: Dividing the target area into a plurality of space-time units based on objective dividing rules; constructing a reference cold pixel set and a reference hot pixel set which characterize the stable surface thermal characteristics of the space-time units by utilizing multi-source data of the historic period corresponding to the space-time units aiming at each space-time unit; According to the acquisition time and the space coverage range of the remote sensing data of the day to be calculated, determining a target space-time unit corresponding to the day to be calculated; And calling a target reference cold pixel set and a target reference hot pixel set corresponding to the target space-time unit, combining remote sensing data and meteorological data of a day to be calculated, and calculating based on a ground surface energy balance model to obtain the actual evaporation quantity of the crops. With reference to the first aspect, the step of dividing the target area into a plurality of space-time units based on objective division rules includes: and dividing the target area into a plurality of space-time units by adopting a standard gas saving sequence as a time division basis and adopting a standard mapping frame as a space division basis. In combination with the first aspect, the step of constructing a reference cold pixel set and a reference hot pixel set representing stable surface thermal characteristics of the space-time unit by using multi-source data of a historical period corresponding to the space-time unit includes: Acquiring a remote sensing image set, a meteorological site data set and digital elevation model data which cover a target area history period, wherein the remote sensing image set comprises a plurality of groups of image data, and each group of image data comprises earth surface reflectivity data and earth surface temperature data; matching the remote sensing image set, the weather site data set and the digital elevation model data to respective corresponding space-time units according to the time and space information; For each group of image data matched into each time-space unit, com