Remote sensing prediction method and remote sensing prediction device for natural grassland forage grass demand quantity

Abstract

The invention discloses a natural grassland forage grass demand remote sensing prediction method and a remote sensing prediction device. The prediction method comprises the following steps: obtaining daily forage grass feed intake according to the existing above-ground biomass of the grassland and the potential above-ground biomass of the grassland; according to the daily forage grass feed intake, the grassland potential above-ground biomass in the future preset time period and the natural grassland reasonable utilization rate, a forage grass demand prediction result in the future preset time period is obtained. According to the remote sensing prediction method for the forage grass demand quantity of the natural grassland provided by the invention, the forage grass demand quantity of the natural grassland can be effectively predicted by accurately calculating the daily forage grass feed intake of the natural grassland and the total amount of edible forage grass in the future preset time period; by dynamically and scientifically evaluating the grassland forage grass demand quantity, reasonable utilization of forage grass resources can be ensured, grassland degradation caused by excessive grazing is avoided, and the grassland production potential is maximized at the same time.

Description

Technical Field

[0001] The present application relates to the field of ecological remote sensing, and in particular to a remote sensing prediction method and a remote sensing prediction device for natural grassland forage demand. Background Art

[0002] As a national strategic resource, grassland plays a vital role in maintaining ecological security and promoting national economic development. By implementing a grass-livestock balance management system and rationally regulating grassland utilization intensity, the ecological health of grasslands can be protected. The core of the grass-livestock balance management system lies in accurately grasping the matching relationship between the edible forage yield of grasslands and livestock needs, which requires an accurate understanding of the natural growth of grasslands, the existing amount, and the actual feed intake of livestock. However, when faced with grazing disturbance, traditional methods have difficulty in obtaining near-real-time and accurate data on the existing aboveground biomass and potential aboveground biomass of grasslands, and thus cannot obtain forage demand, which in turn affects the effective supply of artificial forage for grass-livestock balance. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a natural grassland forage demand remote sensing prediction method and remote sensing prediction device, which can effectively perform remote sensing prediction on the natural grassland forage demand, thereby promoting the implementation of the grass-livestock balance management system.

[0004] In order to achieve the above-mentioned purpose, the embodiment of the present application provides a method for remote sensing prediction of natural grassland forage demand, comprising:

[0005] The daily forage intake was obtained based on the existing aboveground biomass of grassland and the potential aboveground biomass of grassland; the existing aboveground biomass of grassland was obtained by inputting the leaf area index data set, normalized difference vegetation index and remote sensing surface reflectance into the existing biomass remote sensing inversion model; the potential aboveground biomass of grassland was obtained by inputting the digital elevation model data, slope terrain factor data, aspect terrain factor data, total precipitation and 2-meter ground temperature data into the potential biomass remote sensing estimation model;

[0006] Based on the daily forage intake, the potential aboveground biomass of grassland in the future preset period and the reasonable utilization rate of natural grassland, the forage demand in the future preset period is obtained; among which, the potential aboveground biomass of grassland in the future preset period is obtained by inputting digital elevation model data, slope terrain factor data, slope aspect terrain factor data, total precipitation in the future preset period and predicted ground 2-meter temperature in the future preset period into the potential biomass remote sensing estimation model.

[0007] Optionally, the forage demand in a future preset period of time obtained based on the daily forage intake, the potential aboveground biomass of the grassland in a future preset period of time, and the reasonable utilization rate of natural grassland specifically includes:

[0008] According to the daily grass intake and the potential aboveground biomass of the grassland in the future preset period, the total grass intake in the future preset period is obtained;

[0009] According to the potential aboveground biomass of grassland and the reasonable utilization rate of natural grassland in the future preset period, the total amount of edible forage in the future preset period is obtained;

[0010] The forage demand is obtained based on the total amount of forage intake and the total amount of edible forage in a preset period in the future.

[0011] As an option, the training process of the existing biomass remote sensing inversion model is:

[0012] Loading a preprocessed first training data set, the first training data set includes first feature data and first label data, wherein the first feature data includes a leaf area index data set, remote sensing surface reflectance and normalized difference vegetation index, and the first label data includes measured values ​​of aboveground biomass;

[0013] The random forest model is trained using the first training data set to obtain an existing biomass remote sensing inversion model, which includes:

[0014] Preset the number of decision trees to be created in the random forest model;

[0015] For each decision tree, a preset number of features are randomly selected from all features to form a first feature subset considered when training the decision tree;

[0016] Using a bootstrap method to extract a first sample subset with replacement from the first training data set with the same size as the first training data set;

[0017] Training a decision tree using the first feature subset and the first sample subset;

[0018] The above process is repeated until a preset number of decision trees are constructed to obtain the existing biomass remote sensing inversion model.

[0019] Optionally, the leaf area index dataset and remote sensing surface reflectance are obtained by image stitching, resampling and reprojection of data collected by a medium resolution imaging spectrometer; the normalized vegetation index is calculated using the reflectance of the red light band and the reflectance of the near infrared band;

[0020] The aboveground biomass measured value is:

[0021]

[0022] In the formula, AGB represents the measured value of aboveground biomass, n represents the total number of plots, and AGB i Represents the dry weight of aboveground biomass of the i-th grassland population in a certain plot.

[0023] As an option, the training process of the potential biomass remote sensing estimation model is:

[0024] Loading the preprocessed second training data set, the second training data set includes second feature data and second label data, wherein the second feature data includes digital elevation model data, slope terrain factor data, aspect terrain factor data, total precipitation and ground 2-meter temperature data, and the second label data includes existing aboveground biomass of unused grassland and hayfield;

[0025] The random forest model is trained using the second training data set to obtain a potential biomass remote sensing estimation model, which includes:

[0026] Preset the number of decision trees to be created in the random forest model;

[0027] For each decision tree, randomly selecting a preset number of features from all features to form a second feature subset considered when training the decision tree;

[0028] Using a bootstrap method to extract a second sample subset with replacement from the second training data set with the same size as the second training data set;

[0029] Training a decision tree using a second feature subset and a second sample subset;

[0030] The above process is repeated until a preset number of decision trees are constructed to obtain a potential biomass remote sensing estimation model.

[0031] As an option, the process of acquiring the second characteristic data is:

[0032] The slope terrain factor data and the aspect terrain factor data are calculated based on the original digital elevation model data. The daily total precipitation is obtained by accumulating the total precipitation data every hour, and the daily average temperature is obtained by averaging the 2-meter ground temperature data every day.

[0033] The slope terrain factor data, aspect terrain factor data, daily total precipitation and daily average temperature were reprojected and resampled to obtain digital elevation model data with a spatial resolution of 500 meters and longitude and latitude projection, slope terrain factor data, aspect terrain factor data, total precipitation and hourly data of 2-meter ground temperature.

[0034] Optionally, the process of obtaining the second tag data is as follows:

[0035] The time series data of the normalized vegetation index were transformed into principal components, and the top five characteristic bands with the highest characteristics in the principal component analysis results were retained;

[0036] Based on the time series data of the first five characteristic bands, a multi-scale segmentation algorithm is used to identify pasture fences and obtain pasture fence boundary data;

[0037] Based on the grassland fence boundary data, the mowing recognition algorithm is used to identify unused grassland and threshed grassland;

[0038] Based on the existing aboveground biomass of grassland output by the existing biomass remote sensing inversion model, the existing aboveground biomass of unused grassland and the existing aboveground biomass of pasture were obtained.

[0039] As an option, the process of obtaining the total precipitation in the future preset period and the predicted value of the ground 2-meter temperature in the future preset period is:

[0040] Perform time interpolation on the weekly average precipitation and the weekly average ground temperature at 2 meters to obtain the predicted values ​​of the daily average temperature and the daily average ground temperature at 2 meters in the future preset period;

[0041] The daily average temperature and the daily average ground temperature at 2 meters in the preset period in the future are scaled, and the initial 0.25° spatial resolution is converted to 500-meter spatial resolution to obtain the 500-meter spatial resolution, the daily total precipitation in the preset period in the future, and the predicted value of the ground temperature at 2 meters.

[0042] The embodiment of the present application also provides a natural grassland forage demand remote sensing prediction device, comprising:

[0043] A calculation module is configured to obtain daily forage intake according to the existing aboveground biomass of the grassland and the potential aboveground biomass of the grassland; wherein the existing aboveground biomass of the grassland is obtained by inputting the leaf area index data set, the normalized difference vegetation index and the remote sensing surface reflectance into the existing biomass remote sensing inversion model; the potential aboveground biomass of the grassland is obtained by inputting the digital elevation model data, the slope terrain factor data, the aspect terrain factor data, the total precipitation and the ground 2-meter temperature data into the potential biomass remote sensing estimation model;

[0044] The prediction module is configured to obtain the forage demand in a future preset period based on the daily forage intake, the potential aboveground biomass of the grassland in a future preset period, and the reasonable utilization rate of natural grassland; wherein the potential aboveground biomass of the grassland in the future preset period is obtained by inputting digital elevation model data, slope terrain factor data, slope aspect terrain factor data, total precipitation in the future preset period, and the predicted value of the ground 2-meter temperature in the future preset period into the potential biomass remote sensing estimation model.

[0045] An embodiment of the present application further provides a storage medium, which carries one or more computer programs, and when one or more of the computer programs are executed by a processor, the steps of any of the above methods are implemented.

[0046] The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application can accurately calculate the daily forage intake of natural grassland and the total amount of edible forage in a preset time period in the future, and can effectively predict the forage demand of grassland.

[0047] The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application can support the rational utilization and management of forage resources, avoid grassland degradation caused by overgrazing, and maximize the production potential of grassland through scientific evaluation of the existing aboveground biomass and potential aboveground biomass of grassland. The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application predicts the forage demand based on the actual conditions such as the leaf area index and normalized vegetation index of the grassland, as well as environmental factors such as terrain, precipitation, and temperature, which can promote the healthy cycle of grassland ecosystems, help maintain the ecological balance of grasslands, reduce soil erosion, and maintain biodiversity. The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application uses advanced remote sensing technology and data analysis models to provide a reliable tool for ranch managers to make data-driven management decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flow chart of a method for remote sensing prediction of natural grassland forage demand provided in an embodiment of the present application;

[0049] Figure 2 A flowchart of step S200 in the method for remote sensing prediction of natural grassland forage demand provided in an embodiment of the present application;

[0050] Figure 3 A remote sensing prediction effect diagram of the forage demand in a certain area (pastoral area and grassland area) in September 2024 provided in an embodiment of the present application;

[0051] Figure 4 A schematic diagram of the structure of a remote sensing prediction device for natural grassland forage demand provided in an embodiment of the present application;

[0052] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] Various schemes and features of the present application are described herein with reference to the accompanying drawings. It should be understood that various modifications may be made to the embodiments of the present application. Therefore, the above description should not be considered as limiting, but only as an example of an embodiment. Other modifications within the scope and spirit of the present application will occur to those skilled in the art.

[0054] The remote sensing prediction method for natural grassland forage demand provided in this application uses remote sensing technology to accurately estimate the existing and potential biomass of grasslands over a large area, and combines meteorological data and historical grazing intensity to achieve near real-time prediction of grassland forage demand within a preset period in the future. The application of the method in this application will provide strong support for the implementation of more flexible and accurate grass-livestock balance management, which will not only help protect and restore grassland ecosystems, but also promote the harmonious development of pastoral social economy.

[0055] The following is a detailed description of the remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application in conjunction with the accompanying drawings. Figure 1 One of the flowcharts of the remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application is as follows: Figure 1 As shown, the method comprises the following steps:

[0056] S100. Calculate the daily forage intake based on the existing aboveground biomass and potential aboveground biomass of the grassland.

[0057] Specifically, daily forage intake = potential aboveground biomass of grassland - existing aboveground biomass of grassland.

[0058] The existing aboveground biomass of grassland is obtained by inputting the leaf area index dataset, remote sensing surface reflectance and NDVI (Normalized Difference Vegetance Index) into the existing biomass remote sensing inversion model.

[0059] The potential aboveground biomass of grassland is obtained by inputting digital elevation model data, slope terrain factor data, aspect terrain factor data, total precipitation and 2-meter ground temperature data into the potential biomass remote sensing estimation model.

[0060] In an embodiment of the present application, the existing biomass remote sensing inversion model is obtained by training the random forest model using the first training data set. The first training data set includes first feature data and first label data, the first feature data includes a leaf area index data set, remote sensing surface reflectance and normalized difference vegetation index, and the first label data includes the measured value of aboveground biomass.

[0061] Specifically, the acquisition process of the leaf area index dataset and remote sensing surface reflectance is as follows:

[0062] Obtain the LAI (Leaf Area Index) product MOD15A3H and reflectance product MOD43A4 of MODIS (Moderate Resolution Imaging Spectroradiometer).

[0063] The LAI product MOD15A3H and the reflectance product MOD43A4 were processed by image stitching, resampling and reprojection to obtain the MODIS LAI data with a spatial resolution of 500 meters and longitude and latitude projection in the target area and the remote sensing surface reflectance in the MODIS b band (i.e., the blue light band to the near-infrared band).

[0064] Among them, MODIS is a multifunctional medium-resolution imaging spectrometer that can provide rich data on the earth's surface and atmosphere, and is widely used in many fields such as vegetation monitoring, surface temperature monitoring, ocean monitoring, atmosphere monitoring, fire monitoring, and ice and snow monitoring. Through multiple levels of data products, MODIS supports a wide range of scientific research and environmental monitoring work.

[0065] Through MODIS, you can get multiple levels of data products, including: Level 0, Level 1, Level 2, Level 3 and Level 4. Among them, Level 0 products represent unprocessed raw data; Level 1 products represent data that have been preliminarily processed for geolocation and radiation correction; Level 2 products represent data products that have been further processed, such as surface reflectivity, surface temperature and vegetation index; Level 3 products represent data products that synthesize Level 2 products into a unified spatiotemporal grid, which are usually used for long-term trend analysis; Level 4 products represent comprehensive products based on models.

[0066] MODIS has 36 spectral bands, covering a wide range of wavelengths from visible light to thermal infrared, with different bands having different spatial resolutions, ranging from 250 meters (visible light band) to 1000 meters (thermal infrared band). MODIS has a scan width of 2330 kilometers, which can cover almost the entire surface of the Earth every day. The orbit design of the satellites Terra and Aqua allows them to cover the same area multiple times a day, providing data with high temporal resolution. Since MODIS data are strictly calibrated and verified, the accuracy and consistency of the data can be ensured.

[0067] It should be noted that spatial resolution determines the actual ground area represented by each pixel in the remote sensing image. High resolution means that each pixel represents a smaller ground area, and low resolution means the opposite. For a 500-meter spatial resolution, each pixel represents a 500-meter*500-meter ground area. In other words, for an image with a 500-meter spatial resolution, each pixel in the image actually represents a 500-meter*500-meter square area.

[0068] Latitude and longitude projection is a method of mapping the three-dimensional coordinates (longitude and latitude) of the earth's surface to a two-dimensional plane. Its characteristic is that the longitude and latitude values ​​of the earth are directly used to represent the geographical location.

[0069] Specifically, NDVI is a commonly used remote sensing index used to assess the greenness and health of surface vegetation. NDVI is calculated using the reflectance of the red light band and the reflectance of the near-infrared band. The calculation process of NDVI is:

[0070]

[0071] In formula (1), NIR represents the reflectivity in the near-infrared band, and Red represents the reflectivity in the red band.

[0072] In the embodiment of the present application, the aboveground biomass can be investigated by setting up sample plots, selecting sample plots and measuring aboveground biomass. According to the grassland field survey data, the aboveground biomass measured value is obtained:

[0073]

[0074] In formula (2), AGB represents the measured value of aboveground biomass, and its unit is g / m 2 , n represents the total number of plots, AGB i It represents the dry weight of aboveground biomass of the i-th grassland population in a certain plot, and its unit is g / m 2 .

[0075] In a specific embodiment, the process of training the random forest model using the first training data set to obtain the existing biomass remote sensing inversion model is:

[0076] The preprocessed first training data set is loaded, wherein the first training data set includes first feature data and first label data, wherein the first feature data includes a leaf area index data set, remote sensing surface reflectance and normalized difference vegetation index, and the first label data includes measured values ​​of aboveground biomass.

[0077] The random forest model is trained using the first training data set to obtain an existing biomass remote sensing inversion model, which includes:

[0078] Preset the number of decision trees to be created in the random forest model;

[0079] For each decision tree, a preset number of features are randomly selected from all features to form a first feature subset considered when training the decision tree;

[0080] Using a bootstrap method to extract a first sample subset with replacement from the first training data set with the same size as the first training data set;

[0081] Training a decision tree using the first feature subset and the first sample subset;

[0082] The above process is repeated until a preset number of decision trees are constructed to obtain the existing biomass remote sensing inversion model.

[0083] During the training process, the first feature data can be divided into a training set and a validation set in a ratio of 7:3. The performance of the existing biomass remote sensing inversion model is evaluated on the validation set, and the model parameters are adjusted according to the validation results to establish the final existing biomass remote sensing inversion model. Specifically, the existing aboveground biomass can be expressed as the equation with the following input parameters:

[0084] AGB c =f c (LAI, NDVI, ρ b ) (3)

[0085] In formula (3), AGB c represents the existing aboveground biomass of grassland, f c () represents the random forest model for inverting the existing aboveground biomass, ρ b Represents the remote sensing surface reflectance on the MODIS b band.

[0086] In the embodiments of the present application, the potential aboveground biomass of grassland refers to the aboveground biomass that can be achieved by natural grassland in an unused state under specific hydrothermal and topographic conditions.

[0087] The potential biomass remote sensing estimation model is obtained by training the random forest model using the second training data set. The second training data set includes second feature data and second label data, wherein the second feature data includes DEM (Digital Elevation Model) data, slope terrain factor data, aspect terrain factor data, total precipitation and ground 2-meter temperature data, and the second label data includes the existing aboveground biomass of unused grassland and pasture.

[0088] Specifically, DEM is a digital model that represents the topography of the earth's surface. It stores the height information of the surface in the form of a grid. The value of each grid unit (pixel) represents the altitude of the location.

[0089] Slope terrain factor data indicates the angle of inclination or gradient of the ground relative to the horizontal plane. It can be expressed in percentage or degree. For example, a place with a slope of 10% means that the height rises by 10 meters for every 100 meters of horizontal advancement; if expressed in degrees, it is approximately 5.71°. Aspect terrain factor data refers to the direction of maximum descent at a point on the surface. It is usually measured clockwise from north, ranging from 0° (due north) to 360° (back to due north again). For example, the eastward slope aspect is 90° and the southward slope aspect is 180°. Both slope terrain factor data and aspect terrain factor data can be calculated from DEM data.

[0090] The daily total precipitation in the target area is obtained by accumulating the hourly total precipitation data of ERA5. ERA5 is the latest global reanalysis product of ECMWF (European Centre for Medium-Range Weather Forecasts), which provides data on the atmosphere, land and ocean from 1977 to the present. Compared with ECMWF's previous reanalysis product ERA-Interim, ERA5 has improved spatial and temporal resolution and contains more observational data, so it can provide more accurate historical climate information.

[0091] The daily average temperature of the target area is calculated by averaging the 2-meter surface temperature data of ERA5 every day.

[0092] The DEM data, slope terrain factor data, aspect terrain factor data, daily total precipitation and daily average temperature in the target area are projected and resampled using ENVI (Environment for Visualizing Images) software to obtain DEM data, slope terrain factor data, aspect terrain factor data, total precipitation and 2-meter ground temperature data with a spatial resolution of 500 meters and longitude and latitude projection in the target area.

[0093] Specifically, the process of obtaining the existing aboveground biomass of unused grassland and pasture is as follows:

[0094] The time series data of the normalized vegetation index were transformed into principal components, and the top five characteristic bands with the highest characteristics in the principal component analysis results were retained;

[0095] Based on the time series data of the first five characteristic bands, a multi-scale segmentation algorithm is used to identify pasture fences and obtain pasture fence boundary data;

[0096] Based on the grassland fence boundary data, the mowing recognition algorithm is used to identify unused grassland and threshed grassland;

[0097] Based on the existing aboveground biomass of grassland obtained from formula (3), the existing aboveground biomass of unused grassland and the existing aboveground biomass of pasture are obtained, which are the training labels of the potential biomass remote sensing estimation model.

[0098] In a specific embodiment, the process of training the random forest model using the second training data set to obtain the potential biomass remote sensing estimation model is:

[0099] Loading the preprocessed second training data set, the second training data set includes second feature data and second label data, wherein the second feature data includes digital elevation model data, slope terrain factor data, aspect terrain factor data, total precipitation and ground 2-meter temperature data, and the second label data includes existing aboveground biomass of unused grassland and hayfield;

[0100] The random forest model is trained using the second training data set to obtain a potential biomass remote sensing estimation model, which includes:

[0101] Preset the number of decision trees to be created in the random forest model;

[0102] For each decision tree, randomly selecting a preset number of features from all features to form a second feature subset considered when training the decision tree;

[0103] Using a bootstrap method to extract a second sample subset with replacement from the second training data set with the same size as the second training data set;

[0104] Training a decision tree using a second feature subset and a second sample subset;

[0105] The above process is repeated until a preset number of decision trees are constructed to obtain a potential biomass remote sensing estimation model.

[0106] During the training process, the DEM data with 500-meter spatial resolution and latitude and longitude projection, slope terrain factor data, aspect terrain factor data, total precipitation and ground 2-meter temperature data were used as model input parameters. The input parameters were divided into training set and validation set in a ratio of 7:3. The potential aboveground biomass of unused grassland and pasture was used as the model output parameter. The performance of the potential biomass remote sensing estimation model was evaluated on the validation set, and the model parameters were adjusted according to the validation results to establish the final potential biomass remote sensing estimation model. Specifically, the potential aboveground biomass can be expressed as the equation of the following input parameters:

[0107] AGB p =f p (DEM,slope,aspect,tp t1…tn ,T2m ti…tn ) (4)

[0108] In formula (4), AGB p represents the potential aboveground biomass of grassland, f p () represents the random forest model for estimating potential aboveground biomass, slope represents the slope terrain factor data, aspect represents the slope terrain factor data, tp t1…tn represents the total precipitation from time t1 to time tn, T2m t1…tnIt represents the air temperature 2 meters above the ground from time t1 to time tn.

[0109] S200, obtaining the forage demand in a future preset period of time according to the daily forage intake, the potential aboveground biomass of the grassland in a future preset period of time, and the reasonable utilization rate of natural grassland.

[0110] Among them, the potential aboveground biomass of grassland in a preset period in the future is obtained by inputting digital elevation model data, slope terrain factor data, aspect terrain factor data, total precipitation in a preset period in the future and predicted ground 2-meter temperature in a preset period in the future into the potential biomass remote sensing estimation model.

[0111] In the embodiment of the present application, the process of obtaining the total precipitation in the future preset period and the predicted value of the ground 2-meter temperature in the future preset period is as follows:

[0112] Temporal interpolation is performed on the weekly average precipitation and the weekly average surface temperature at 2 meters above the ground to obtain the predicted values ​​of the daily average temperature and the daily average surface temperature at 2 meters above the ground within a preset period in the future.

[0113] The daily average temperature and the daily average 2-meter ground temperature forecast values ​​for the future preset period are scaled, and the initial 0.25° spatial resolution is converted to 500-meter spatial resolution to obtain the 500-meter spatial resolution, the daily total precipitation and the 2-meter ground temperature forecast values ​​for the future preset period.

[0114] In a specific embodiment, the weekly average precipitation and the weekly average ground temperature at 2 meters extracted from the meteorological forecast data of ERA5 IFS (Integrated Forecasting System) are temporally interpolated to obtain the daily average temperature and the daily average ground temperature at 2 meters predicted values ​​for the next 30 days.

[0115] The ENVI software was used to scale the daily average temperature and the daily average 2-meter ground temperature forecast for the next 30 days, and the initial 0.25° spatial resolution was converted to 500-meter spatial resolution, and the 500-meter spatial resolution, longitude and latitude projection, daily total precipitation and 2-meter ground temperature forecast for the target area in the next 30 days were obtained.

[0116] In the above step S200 of the embodiment of the present application, Figure 2 As shown in the figure, based on the daily forage intake, the potential aboveground biomass of grassland in the future preset period and the reasonable utilization rate of natural grassland, the forage demand in the future preset period is obtained. The specific process is as follows:

[0117] S210, obtaining the total amount of forage intake in the future preset period according to the daily forage intake and the potential aboveground biomass of the grassland in the future preset period.

[0118] S220. Calculate the total amount of edible forage grass within a preset period of time in the future based on the potential aboveground biomass of grassland and the reasonable utilization rate of natural grassland within a preset period of time in the future.

[0119] S230: Subtract the total amount of forage consumed and the total amount of edible forage in a future preset period of time to obtain the forage demand.

[0120] The above step S200 is described in detail below using a specific embodiment.

[0121] In the embodiment of the present application, in order to describe the consumption of grassland vegetation biomass by grazing activities, the difference between the potential aboveground biomass of grassland and the existing aboveground biomass of grassland is defined as the forage intake, that is,

[0122] GC=AGB p -AGB c (5)

[0123] In formula (5), GC represents the grass intake, AGB p represents the potential aboveground biomass of grassland, AGB c Represents the existing aboveground biomass of grassland.

[0124] The total amount of grass intake GC from time t1 to time t2 t1-t2 It can be estimated by the difference between the potential aboveground biomass and the existing aboveground biomass during this period, that is,

[0125] GC t1-t2 =AGB p_t2 -AGB c_t2 -AGB p_t1 +AGB c_t1 (6)

[0126] In formula (6), GC t1-t2 Indicates the total amount of forage intake during the period from t1 to t2, AGB p_t2 represents the potential aboveground biomass of grassland at t2, AGB c_t2 represents the existing aboveground biomass of grassland at t2, AGB p_t1 represents the potential aboveground biomass of grassland at t1, AGB c_t1 Represents the existing aboveground biomass of grassland at t1.

[0127] The daily feed intake DGC can be expressed as the total amount of grass intake GC during the period from t1 to t2 t1-t2 The ratio of the duration from time t1 to time t2, that is

[0128]

[0129] In the above step S210, assuming that the daily feed intake remains unchanged during the time period t1-t3, the total amount of forage intake GC during the time period t2-t3 is t2-t3 for:

[0130]

[0131] In the embodiment of the present application, without considering the compensatory growth of natural grassland, the existing aboveground biomass AGB of grassland at the future time t3 is c_t3 It can be considered as the existing aboveground biomass AGB of grassland at time t2 c_t2 The increase in aboveground biomass ΔAGB during the period t2-t3 t2_t3 The sum of

[0132] AGB c_t3 =AGB c_t2 +ΔAGB t2_t3 =AGB c_t2 +AGB p_t3 -AGB p_t2 (9)

[0133] Among them, according to the grassland type data of the target area, the reasonable utilization rate p of different grassland types is set in accordance with the standard of the Agricultural Industry Standard of the People's Republic of China "Calculation of Reasonable Carrying Capacity of Natural Grassland" (NY / T 635-2015). For example, the reasonable utilization rate of four-season grazing in typical meadows is 50-55%, and the reasonable utilization rate of four-season grazing in sandy grasslands is 20-30%. Here, the reasonable utilization rate p takes the average of the range of reasonable utilization rates of four-season grazing in this standard.

[0134] In the above step S220, the reasonable utilization rate p of natural grassland and the existing aboveground biomass AGB of grassland at the future time t3 are combined. c_t3 , we can get the total amount of edible forage TF in the time period t2-t3 t2-t3 :

[0135] TF t2-t3 = p·AGB c_t3 =p(AGB p_t3 -AGB p_t2 +AGB c_t2 ) (10)

[0136] In the present embodiment, if the total forage intake GC during the time period from t2 to t3 is t2-t3 Greater than the total amount of edible forage TF during this period t2-t3 In order to ensure the ecological health of natural grasslands and the orderly management of grass-livestock balance, exogenous forage is needed.

[0137] In the above step S230, the predicted exogenous forage demand NF is the total forage intake GC during the period from t2 to t3. t2-t3 Total edible forage TF t2-t3 The difference is

[0138]

[0139] Exemplarily, taking a geographical location as a unit quantity, a forecast graph of forage demand in September 2024 for a certain area (pastoral area and grassland area) to which the geographical location belongs is drawn. Figure 3 The figure shows the distribution of forage demand for the next 30 days (i.e. September 1st to September 30th) predicted in August 2024. Figure 3 It can be seen that the present application can effectively predict the future demand for fodder in vast pastoral and grassland areas.

[0140] The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application can effectively predict the natural grassland forage demand by accurately calculating the daily forage intake of the natural grassland and the total amount of edible forage in a preset time period in the future.

[0141] The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application can ensure the rational use of forage resources, avoid grassland degradation caused by overgrazing, and maximize the production potential of grassland through scientific evaluation of the existing aboveground biomass and potential aboveground biomass of grassland.

[0142] The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application predicts the forage demand based on the actual conditions of the grassland such as leaf area index and normalized vegetation index as well as environmental factors such as terrain, precipitation, and temperature. It can promote the healthy cycle of grassland ecosystems, help maintain the ecological balance of grasslands, reduce soil erosion, and maintain biodiversity.

[0143] The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application uses advanced remote sensing technology and data analysis models to provide a reliable tool for ranch managers to make data-driven management decisions. This can not only improve ranch management efficiency, but also help adapt to the uncertainty brought about by climate change and enhance the flexibility and risk resistance of agricultural production.

[0144] The remote sensing prediction method for natural grassland forage demand provided in the embodiment of the present application can effectively control grazing intensity, help mitigate greenhouse gas emissions, reduce the impact on the surrounding natural environment, and is in line with the concept of green development.

[0145] Based on the same inventive concept, Figure 4As shown, the embodiment of the present application also provides a natural grassland forage demand remote sensing prediction device, comprising:

[0146] The calculation module is configured to obtain the daily forage intake according to the existing aboveground biomass of the grassland and the potential aboveground biomass of the grassland; wherein the existing aboveground biomass of the grassland is obtained by inputting the leaf area index data set, the normalized vegetation index and the remote sensing surface reflectivity into the existing biomass remote sensing inversion model; the potential aboveground biomass of the grassland is obtained by inputting the digital elevation model data, the slope terrain factor data, the slope aspect terrain factor data, the total precipitation and the ground 2-meter temperature data into the potential biomass remote sensing estimation model.

[0147] In an embodiment of the present application, the existing biomass remote sensing inversion model is obtained by training the random forest model using the first training data set. The first training data set includes first feature data and first label data, the first feature data includes a leaf area index data set, remote sensing surface reflectance and normalized difference vegetation index, and the first label data includes the measured value of aboveground biomass.

[0148] In the embodiment of the present application, the potential biomass remote sensing estimation model is obtained by training the random forest model using the second training data set. The second training data set includes second feature data and second label data, wherein the second feature data includes DEM (Digital Elevation Model) data, slope terrain factor data, aspect terrain factor data, total precipitation and ground 2-meter temperature data, and the second label data includes the existing aboveground biomass of unused grassland and hayfield.

[0149] The prediction module is configured to obtain the forage demand in a future preset period based on the daily forage intake, the potential aboveground biomass of the grassland in a future preset period, and the reasonable utilization rate of natural grassland; wherein the potential aboveground biomass of the grassland in the future preset period is obtained by inputting digital elevation model data, slope terrain factor data, slope aspect terrain factor data, total precipitation in the future preset period, and the predicted value of the ground 2-meter temperature in the future preset period into the potential biomass remote sensing estimation model.

[0150] In the embodiment of the present application, the process of obtaining the total precipitation in the future preset period and the predicted value of the ground 2-meter temperature in the future preset period is as follows:

[0151] Temporal interpolation is performed on the weekly average precipitation and the weekly average surface temperature at 2 meters above the ground to obtain the predicted values ​​of the daily average temperature and the daily average surface temperature at 2 meters above the ground within a preset period in the future.

[0152] The daily average temperature and the daily average 2-meter ground temperature forecast values ​​for the future preset period are scaled, and the initial 0.25° spatial resolution is converted to 500-meter spatial resolution to obtain 500-meter spatial resolution, longitude and latitude projection, daily total precipitation and 2-meter ground temperature forecast values ​​for the future preset period.

[0153] Optionally, the prediction module is further configured to:

[0154] According to the daily grass intake and the potential aboveground biomass of the grassland in the future preset period, the total grass intake in the future preset period is obtained;

[0155] The total amount of edible forage grass in a preset future period is calculated based on the potential aboveground biomass of grassland and the reasonable utilization rate of natural grassland in a preset future period;

[0156] The forage demand is obtained by subtracting the total amount of forage consumed and the total amount of edible forage in the future preset period of time.

[0157] Based on the same concept, the embodiment of the present application also provides an electronic device, such as Figure 5 As shown, it includes a processor and a memory, wherein an executable program is stored in the memory, and the processor executes the executable program to perform the steps of the method as described above.

[0158] An embodiment of the present application further provides a storage medium, wherein the storage medium carries one or more computer programs, and when the one or more computer programs are executed by a processor, the steps of the above method are implemented.

[0159] The storage medium in this embodiment may be included in the electronic device / system; or it may exist independently without being assembled into the electronic device / system. The above storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present application is implemented.

[0160] It should be understood that in the embodiments of the present application, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.

[0161] It should also be understood that the memory mentioned in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM).

[0162] It should be noted that when the processor is a general-purpose processor, DSP, ASIC, FPGA or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, the memory (storage module) is integrated in the processor.

[0163] It should be noted that the memory described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0164] It should also be understood that the first, second, third, fourth and various numerical numbers involved in this document are only distinctions made for the convenience of description and are not intended to limit the scope of the present application.

[0165] It should be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0166] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software. The steps of the method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in a processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.

[0167] In various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0168] Those skilled in the art will appreciate that the various illustrative logical blocks (ILBs) and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0169] In the several embodiments provided in the present application, it should be understood that the disclosed remote sensing prediction methods, devices, electronic devices, storage media and chips can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units (or modules) is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0170] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0171] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0172] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integration. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0173] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for remote sensing prediction of natural grassland forage demand, characterized in that: include: The daily forage intake was obtained based on the existing aboveground biomass of grassland and the potential aboveground biomass of grassland; the existing aboveground biomass of grassland was obtained by inputting the leaf area index data set, normalized difference vegetation index and remote sensing surface reflectance into the existing biomass remote sensing inversion model; the potential aboveground biomass of grassland was obtained by inputting the digital elevation model data, slope terrain factor data, aspect terrain factor data, total precipitation and 2-meter ground temperature data into the potential biomass remote sensing estimation model; Based on the daily forage intake, the potential aboveground biomass of grassland in the future preset period and the reasonable utilization rate of natural grassland, the forage demand in the future preset period is obtained; among which, the potential aboveground biomass of grassland in the future preset period is obtained by inputting digital elevation model data, slope terrain factor data, slope aspect terrain factor data, total precipitation in the future preset period and predicted ground 2-meter temperature in the future preset period into the potential biomass remote sensing estimation model.

2. The method according to claim 1, characterized in that: The forage demand in the future preset period obtained based on the daily forage intake, the potential aboveground biomass of the grassland in the future preset period and the reasonable utilization rate of natural grassland specifically includes: According to the daily grass intake and the potential aboveground biomass of the grassland in the future preset period, the total grass intake in the future preset period is obtained; According to the potential aboveground biomass of grassland and the reasonable utilization rate of natural grassland in the future preset period, the total amount of edible forage in the future preset period is obtained; The forage demand is obtained based on the total amount of forage intake and the total amount of edible forage in a preset period in the future.

3. The method according to claim 1, characterized in that The training process of the existing biomass remote sensing inversion model is as follows: Loading a preprocessed first training data set, the first training data set includes first feature data and first label data, wherein the first feature data includes a leaf area index data set, remote sensing surface reflectance and normalized difference vegetation index, and the first label data includes measured values ​​of aboveground biomass; The random forest model is trained using the first training data set to obtain an existing biomass remote sensing inversion model, which includes: Preset the number of decision trees to be created in the random forest model; For each decision tree, a preset number of features are randomly selected from all features to form a first feature subset considered when training the decision tree; Using a bootstrap method to extract a first sample subset with replacement from the first training data set with the same size as the first training data set; Training a decision tree using the first feature subset and the first sample subset; The above process is repeated until a preset number of decision trees are constructed to obtain the existing biomass remote sensing inversion model.

4. The method according to claim 3, characterized in that The leaf area index dataset and remote sensing surface reflectance are obtained by image stitching, resampling and reprojection of data collected by a medium resolution imaging spectrometer; the normalized vegetation index is calculated using remote sensing reflectance in the red light band and remote sensing reflectance in the near infrared band; The aboveground biomass measured value is: In the formula, AGB represents the measured value of aboveground biomass, n represents the total number of plots, and AGB i Represents the dry weight of aboveground biomass of the i-th grassland population in a certain plot.

5. The method according to claim 1, characterized in that The training process of the potential biomass remote sensing estimation model is as follows: Loading the preprocessed second training data set, the second training data set includes second feature data and second label data, wherein the second feature data includes digital elevation model data, slope terrain factor data, aspect terrain factor data, total precipitation and ground 2-meter temperature data, and the second label data includes existing aboveground biomass of unused grassland and hayfield; The random forest model is trained using the second training data set to obtain a potential biomass remote sensing estimation model, which includes: Preset the number of decision trees to be created in the random forest model; For each decision tree, randomly selecting a preset number of features from all features to form a second feature subset considered when training the decision tree; Using a bootstrap method to extract a second sample subset with replacement from the second training data set with the same size as the second training data set; Training a decision tree using a second feature subset and a second sample subset; The above process is repeated until a preset number of decision trees are constructed to obtain a potential biomass remote sensing estimation model.

6. The method according to claim 5, characterized in that The process of obtaining the second characteristic data is as follows: The slope terrain factor data and the aspect terrain factor data are calculated based on the original digital elevation model data. The daily total precipitation is obtained by accumulating the total precipitation data every hour, and the daily average temperature is obtained by averaging the 2-meter ground temperature data every day. The slope terrain factor data, aspect terrain factor data, daily total precipitation and daily average temperature were reprojected and resampled to obtain digital elevation model data with a spatial resolution of 500 meters and longitude and latitude projection, slope terrain factor data, aspect terrain factor data, total precipitation and hourly data of 2-meter ground temperature.

7. The method according to claim 5, characterized in that The process of obtaining the second tag data is as follows: The time series data of the normalized vegetation index were transformed into principal components, and the top five characteristic bands with the highest characteristics in the principal component analysis results were retained; Based on the time series data of the first five characteristic bands, a multi-scale segmentation algorithm is used to identify pasture fences and obtain pasture fence boundary data; Based on the grassland fence boundary data, the mowing recognition algorithm is used to identify unused grassland and threshed grassland; Based on the existing aboveground biomass of grassland output by the existing biomass remote sensing inversion model, the existing aboveground biomass of unused grassland and the existing aboveground biomass of pasture were obtained.

8. The method according to claim 1, characterized in that The acquisition process of the total precipitation in the future preset period and the predicted value of the ground 2-meter temperature in the future preset period is: Perform time interpolation on the weekly average precipitation and the weekly average ground temperature at 2 meters to obtain the predicted values ​​of the daily average temperature and the daily average ground temperature at 2 meters in the future preset period; The daily average temperature and the daily average 2-meter ground temperature forecast values ​​for the future preset period are scaled, and the initial 0.25° spatial resolution is converted to 500-meter spatial resolution to obtain the 500-meter spatial resolution, the daily total precipitation and the 2-meter ground temperature forecast values ​​for the future preset period.

9. A remote sensing prediction device for natural grassland forage demand, characterized in that: include: A calculation module is configured to obtain daily forage intake based on remotely sensed existing aboveground biomass of grassland and potential aboveground biomass of grassland; wherein the existing aboveground biomass of grassland is obtained by inputting leaf area index data set, normalized difference vegetation index and remote sensing surface reflectivity into the existing biomass remote sensing inversion model; the potential aboveground biomass of grassland is obtained by inputting digital elevation model data, slope terrain factor data, aspect terrain factor data, total precipitation and ground 2-meter temperature data into the potential biomass remote sensing estimation model; The prediction module is configured to obtain the forage demand in a future preset period based on the daily forage intake, the potential aboveground biomass of the grassland in a future preset period, and the reasonable utilization rate of natural grassland; wherein the potential aboveground biomass of the grassland in the future preset period is obtained by inputting digital elevation model data, slope terrain factor data, slope aspect terrain factor data, total precipitation in the future preset period, and the predicted value of the ground 2-meter temperature in the future preset period into the potential biomass remote sensing estimation model.

10. A storage medium, characterized in that: The storage medium carries one or more computer programs, and when the one or more computer programs are executed by the processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images