Drought monitoring method combining improved TVDI index and dynamic drought threshold

Abstract

The invention discloses a drought monitoring method combining an improved TVDI index and a dynamic drought threshold. According to the drought monitoring method, an improved temperature vegetation drought index TVDIm and the dynamic drought threshold considering the water demand characteristic of a crop in the growth stage are introduced. By screening out vegetation which does not conform to the TVDI basic theory, the detection accuracy of dry and wet edges of the surface temperature-vegetation index space is enhanced to improve the monitoring accuracy, so that the TVDIm can more accurately reflect the surface humidity and drought conditions. By analyzing the correlation between the TVDIm of the crops in different growth stages and the actual soil humidity and combining the suitable soil humidity range of the crops, the dynamic drought threshold is determined, the water demand characteristic of the crop growth cycle is considered, and drought monitoring is more adaptive and accurate. Through the improved TVDIm and the dynamic drought threshold, the region influenced by drought can be identified more accurately.

Description

Technical Field

[0001] The invention relates to the technical field of drought monitoring, and in particular to a drought monitoring method combining an improved TVDI index with a dynamic drought threshold. Background Art

[0002] The frequent occurrence of droughts poses a major threat to agricultural productivity and food security, which requires accurate drought monitoring and assessment tools. Existing drought monitoring methods have difficulty in effectively coping with the complexity of variable climate and geographical conditions. Although soil moisture is a reliable indicator of agricultural drought, due to limitations in data access, researchers have turned to remote sensing technology for dynamic drought monitoring across temporal and spatial scales. Remote sensing technology provides in-depth insights into vegetation and soil moisture conditions through the combined use of surface temperature and vegetation indices. The Temperature Vegetation Drought Index (TVDI) is highly regarded for its ability to effectively monitor surface moisture and drought without the need for auxiliary data, and has become the main agricultural drought monitoring indicator.

[0003] Although TVDI is widely used because it can effectively monitor surface moisture and drought without auxiliary data, its accuracy in fitting the regional dry / wet edge equation is still challenging, affecting the accuracy of the monitoring results. In order to improve the accuracy, researchers have adopted a variety of improvement methods, including terrain correction of surface temperature, consideration of diurnal surface temperature changes, dry-wet edge fitting combined with the energy balance equation, using the EVI index and the polynomial formula of vegetation index and surface temperature to fit the dry-wet edge, and using the slope extreme value of the two-dimensional spatial relationship between surface temperature and vegetation index in multiple phases of the target pixel for calculation. These methods show that the adoption of diversified strategies and continuous innovation are crucial to improving TVDI drought monitoring. Although soil moisture monitoring data is limited, it is critical to evaluate the accuracy of TVDI. At present, it is mainly used for accuracy evaluation indicators and analyzing the effectiveness of the TVDI index. Previous classifications of TVDI values ​​for drought monitoring have limitations. For example, TVDI values ​​are directly classified into levels, such as areas with TVDI values ​​above 0.6 are classified as drought, or areas with TVDI values ​​above 0.4 are classified as drought. This method does not take into account geographical differences and the water requirements of crops at different growth stages, and the drought areas obtained are subject to great uncertainty.

[0004] It can be seen that the accuracy of TVDI mainly depends on the accuracy of its fitting of the dry and wet edges of the surface temperature and vegetation index space. The traditional TVDI method does not effectively screen out vegetation that does not meet the requirements when fitting the dry and wet edges of the surface temperature and vegetation index, resulting in insufficient fitting accuracy. The assessment of drought areas mainly relies on the classification of TDVI values, but the traditional standard does not take into account geographical variation and crop water requirements, especially the lack of consideration of water demand characteristics at different growth stages. Summary of the invention

[0005] The present invention provides a drought monitoring method combining an improved TVDI index and a dynamic drought threshold value, so as to solve the problem that the accuracy of the existing drought monitoring method needs to be improved.

[0006] The present invention provides a drought monitoring method combining an improved TVDI index and a dynamic drought threshold, comprising: Step 1: Obtain the land surface temperature image data and vegetation index image data of the required period of the study area, the data covering the entire study area and the complete growing season, and pre-process and screen the land surface temperature image data and vegetation index image data; Step 2: Extract the maximum and minimum surface temperature values ​​corresponding to different vegetation index values ​​from the screened data, and establish a surface temperature-vegetation index scatter distribution diagram, with the horizontal axis coordinate being the vegetation index value and the vertical axis coordinate being the maximum and minimum surface temperature values ​​corresponding to the vegetation index value; the maximum surface temperature is called the dry edge of space, and the minimum surface temperature is called the wet edge of space; Step 3: Based on the surface temperature-vegetation index scatter plot, analyze whether the maximum surface temperature and the minimum surface temperature in the scatter space are consistent with the following theory as the vegetation index value increases from small to large: the dry side temperature increases with the increase of vegetation coverage, and the wet side temperature decreases with the increase of vegetation coverage; find the critical value VI of the vegetation index that does not conform to this theory threshold , filter out vegetation index values ​​less than the critical value, and retain the critical value to the maximum vegetation index value VI max The effective vegetation index range is determined as [VI threshold , VI max ]; Step 4: Based on the effective vegetation index range, a spatial relationship diagram between the effective vegetation index and the surface temperature is established, where the horizontal axis coordinate is the vegetation index value, and the vertical axis coordinate is the maximum surface temperature value and the minimum surface temperature value corresponding to the vegetation index value; Step 5, based on the spatial relationship diagram of the effective vegetation index and the surface temperature, the improved dry edge fitting equation and the wet edge fitting equation are calculated; Step 6: Calculate the improved TVDI index based on the improved dry side fitting equation and the wet side fitting equationTVD m ; Step 7: Analysis TVD m Correlation with the measured soil moisture on the surface, select effective TVD m ; Step 8: Based on the information of the main crops in the study area, the starting time of the crop growth stage is divided and the suitable relative soil moisture for the crop is determined; according to the determined starting time of each growth stage, the effective soil moisture for each period of the year is calculated. TVD m Classify into corresponding growth stages; Step nine, determining dynamic drought thresholds for each growth stage; Step 10: According to the dynamic drought thresholds of each growth stage, TVD m , determine the dynamic drought threshold of the corresponding growth stage, and then TVD m The area is classified as drought-free zone, which is greater than or equal to TVD m The areas with dynamic thresholds are classified as drought areas.

[0007] Furthermore, in step one, the vegetation index image data is normalized vegetation index image data or enhanced vegetation index image data; the surface temperature image data and vegetation index image data are preprocessed to ensure that the data are consistent in time and spatial resolution; and data screening requires that the proportion of effective surface temperature pixels in the study area in the images of each period is more than 85% of the study area.

[0008] Furthermore, step five includes: For the improved fitting of dry edge and wet edge, the fitted linear correlation between the vegetation index and the maximum surface temperature and the minimum surface temperature in the effective scatter plot is calculated respectively to obtain the improved dry edge fitting equation and the wet edge fitting equation; Improved dry edge fitting equation: T s max-m =a 1 +b 1 *VI [VIthreshold, VImax] ; T s max-m For the corresponding VI [VIthreshold, VImax] The maximum surface temperature of the value; a 1 、b 1 is the fitting coefficient of the improved dry edge fitting equation; Improved wet edge fitting equation: Ts min-m =a 2 +b 2*VI [VIthreshold, VImax] ; Ts min-m For the corresponding VI [VIthreshold, VImax] The lowest surface temperature of the value; a 2 、b 2 is the fitting coefficient of the improved wet edge fitting equation.

[0009] Furthermore, after step five, the following steps are also included: The improvement is determined by comparing the determination coefficients of the dry side fitting equation and the wet side fitting equation before and after the improvement.

[0010] Further, in step 6, the improved TVDI index is calculated according to the following formula: TVD m : , T s max-m For the corresponding VI [VIthreshold, VImax] The maximum surface temperature of the Ts min-m For the corresponding VI [VIthreshold, VImax] The minimum surface temperature of Ts is the surface temperature of any pixel in the remote sensing image.

[0011] Furthermore, step seven includes: Obtain the measured soil moisture data on the ground and the field water holding capacity data of the corresponding stations to ensure that the date of each sampling data is consistent with the remote sensing calculation. TVD m The relative soil moisture (RSM) of the stations at each period is calculated. TVD m The linear fitting relationship between the values; where RSM = (soil moisture / field capacity) × 100%; TVD m The value is the corresponding pixel of the ground measured data in the same period TVD m value; Estimation of the coefficient of determination using a linear fit relationship TVD m The degree of fit with the ground monitoring soil moisture is judged by the reference table of determination coefficient significance. TVD m The number of paired samples was determined to determine the significant correlation at the level of p < 0.05, and the different periods that met the significant correlation were TVD m Considered valid TVD m .

[0012] Furthermore, step nine includes: Get all the TVD m Scatter plot with measured soil moisture; Determine the appropriate relative soil moisture range TVD m value, and use the percentile method to determine the stage TVD m The maximum value of TVD m The values ​​are sorted, and then the position of the specified percentile is calculated. If the calculated percentile position is not an integer, interpolation is performed between adjacent data points to obtain a more accurate percentile value and use it as TVDI m The upper limit of The upper limit is set as the threshold for possible drought in this stage, and the period above this upper limit may indicate drought. TVD m scope.

[0013] The present invention has the following beneficial effects: a drought monitoring method combining an improved TVDI index and a dynamic drought threshold is provided by introducing an improved temperature vegetation drought index. TVD m A dynamic drought threshold that takes into account the water demand characteristics of crop growth stages is proposed to enhance agricultural drought monitoring and thus more accurately identify drought-affected areas. TVD m , by filtering out TVD Based on the vegetation theory, the detection accuracy of the dry and wet edges of the surface temperature-vegetation index space is enhanced to improve the monitoring accuracy. TVD m It can more accurately reflect the surface moisture and drought conditions. Develop dynamic drought thresholds by analyzing the different growth stages of crops. TVD m The correlation between the actual soil moisture and the soil moisture range suitable for crops is used to determine the dynamic drought threshold. This method takes into account the water demand characteristics of the crop growth cycle, making drought monitoring more adaptable and accurate. TVD m and dynamic drought thresholds, which enable more accurate identification of drought-affected areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0015] tú1A flowchart of a drought monitoring method combining an improved TVDI index and a dynamic drought threshold provided by the present invention; tú2 for TVD , TVD m Comparison of dry and wet edge fitting in space; tú3 For different periods TVD m Correlation coefficients with soil moisture at different soil layers; tú4 For different crops in different growth cycles TVD m Scatter plot against relative soil moisture. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0017] TVD The method has been widely used in drought monitoring, which combines remotely sensed surface temperature and vegetation index to assess soil moisture conditions. The assumption is that changes in surface temperature corresponding to a specific vegetation cover indicate changes in soil moisture levels, and that the temperature of vegetation-covered areas is more sensitive to soil moisture changes than the temperature of bare soil. The theory holds that in the spatial relationship between remotely sensed surface temperature and vegetation index, on the dry side, as vegetation cover increases, surface temperature generally decreases because transpiration from vegetation increases water evaporation and lowers temperature; on the wet side, as vegetation cover increases, surface temperature may increase because transpiration from vegetation may lead to higher temperatures under high humidity conditions. TVD By obtaining the theoretical dry-side temperature and wet-side temperature to calculate the drought index, an effective drought monitoring tool is provided.

[0018] The present invention firstly improves the traditional TVD Improvements have been made so that the improved TVD m It can more accurately reflect the vegetation and water conditions in different seasons and environmental conditions. TVDIn the method, the relationship between surface temperature and vegetation index is used to assess soil moisture conditions. The improved method analyzes this relationship in depth, paying special attention to the temperature changes between the dry and wet edges, and analyzes whether the dry edge temperature increases with the increase of vegetation cover and whether the wet edge temperature decreases with the increase of vegetation cover in the spatial relationship between surface temperature and vegetation index. By analyzing the changing trends of dry and wet edge temperatures, clear turning points are identified. These turning points are used to determine effective vegetation index thresholds, thereby helping to identify conditions that meet the requirements. TVD The vegetation coverage range of the basic theory is screened out, and the vegetation range that does not meet the requirements is screened out. The spatial relationship between the surface temperature and the vegetation index is re-established through the screened effective vegetation index. Based on the new spatial relationship, the dry edge and the wet edge are re-fitted to improve the fitting accuracy of the dry edge and the wet edge, thereby improving TVD m overall accuracy.

[0019] Secondly, a dynamic drought threshold method is proposed to adjust the drought assessment criteria according to the crop growth stage. Traditionally, the same drought assessment criteria are used in different growth stages, ignoring the soil moisture characteristics suitable for regional crops. The dynamic threshold method can better reflect the water demand characteristics of crops at different growth stages. TVD m After linear fitting correlation with measured relative soil moisture, the linear fitting determination coefficient R 2 The significance of the screen out effective TVD m , make sure you select TVD m It can accurately reflect the soil moisture status. Then, the crop growth cycle is determined according to the agricultural information, and the annual period is TVD m Divide into different growth stages, and draw the TVD m The scatter plot with the measured relative soil moisture can be intuitively observed TVD m The relationship between the relative soil moisture and the measured relative soil moisture. TVD m Range, yes TVD m Sort and use the percentile method to determine the TVDI for that stage m The maximum value of is set as the drought threshold for this growth stage, ensuring that the threshold is set based on the actual soil moisture requirements. TVD m The value range is 0~1, with higher TVD m Values ​​generally reflect more severe drought conditions. TVDm The threshold value can be used to obtain the drought area and further divide the severity of drought.

[0020] See also tú1zhìtú4 The present invention provides a drought monitoring method combining an improved TVDI index and a dynamic drought threshold, comprising: Step 1: Obtain the surface temperature image data and vegetation index image data of the required period of the study area, the data covering the entire study area and the complete growing season, and preprocess and screen the surface temperature image data and vegetation index image data.

[0021] Specifically, according to the region or needs, select appropriate satellite data, such as MDOIS, Landsat or Sentinel series. Download the Land Surface Temperature (LST) image and vegetation index image of the required time period from the selected satellite data platform. The vegetation index can be the Normalized Difference Vegetation Index (NDVI) or the Enhanced Vegetation Index (EVI) to ensure coverage of the entire study area and the complete growing season. Perform general routine operations such as preprocessing on the data to ensure that the data has been quality controlled and is consistent in time and spatial resolution. In this embodiment, when screening the data, it is required that the proportion of effective surface temperature pixels in the study area in the images of each period should be more than 85% of the study area to ensure the effectiveness of the established surface temperature-vegetation index space and to ensure that a sufficiently wide range of surface temperature and vegetation index is covered.

[0022] Step 2: Extract the maximum and minimum surface temperature values ​​corresponding to different vegetation index values ​​from the screened data, and establish a surface temperature-vegetation index scatter distribution diagram, with the horizontal axis being the vegetation index value and the vertical axis being the maximum and minimum surface temperature values ​​corresponding to the vegetation index value; the maximum surface temperature is called the spatial dry edge, and the minimum surface temperature is called the spatial wet edge.

[0023] This is tradition TVD General operations when establishing the surface temperature-vegetation index relationship in the method.

[0024] Step 3: Based on the surface temperature-vegetation index scatter plot, analyze whether the maximum surface temperature and the minimum surface temperature in the scatter space are consistent with the following theory as the vegetation index value increases from small to large: the dry side temperature increases with the increase of vegetation coverage, and the wet side temperature decreases with the increase of vegetation coverage; find the critical value VI of the vegetation index that does not conform to this theory threshold, filter out vegetation index values ​​less than the critical value, and retain the critical value to the maximum vegetation index value VI max The effective vegetation index range is determined as [VI threshold , VI max ].

[0025] In order to find the best fitting dry edge and fitting wet edge, the effective vegetation index range is first determined. The maximum surface temperature and the minimum surface temperature in the scattered point space are analyzed as the vegetation index value increases from small to large. It is analyzed whether the dry edge temperature increases with the increase of vegetation coverage and whether the wet edge temperature decreases with the increase of vegetation coverage. This is the basic theory of the TVDI method.

[0026] In this embodiment, taking the EVI index as an example, the critical value of the vegetation index value that does not conform to the basic theory for the dry edge and the wet edge is VI threshold =0.2, that is, vegetation indices less than 0.2 are screened out.

[0027] Step 4: Based on the effective vegetation index range, a spatial relationship diagram between the effective vegetation index and the surface temperature is established, with the horizontal axis being the vegetation index value and the vertical axis being the maximum surface temperature value and the minimum surface temperature value corresponding to the vegetation index value.

[0028] In this embodiment, for the EVI index, the effective spatial relationship vegetation index value range is [0.2, EVI max ], where EVI max is the actual maximum EVI value of the image in each period, which does not exceed 1.

[0029] Step 5: Based on the spatial relationship diagram between the effective vegetation index and the surface temperature, the improved dry edge fitting equation and the wet edge fitting equation are calculated.

[0030] Specifically, for the improved fitting of the dry edge and the wet edge, the fitted linear correlation between the vegetation index and the maximum surface temperature and the minimum surface temperature in the effective scatter plot is calculated and obtained respectively, and the improved dry edge fitting equation and the wet edge fitting equation are obtained; Improved dry edge fitting equation: T s max-m =a 1 +b 1 *VI [VIthreshold, VImax] ; T s max-m For the corresponding VI [VIthreshold, VImax] The maximum surface temperature of the value; a 1 、b 1 is the fitting coefficient of the improved dry edge fitting equation; Improved wet edge fitting equation: Ts min-m =a2 +b 2 *VI [VIthreshold, VImax] ; Ts min-m For the corresponding VI [VIthreshold, VImax] The lowest surface temperature of the value; a 2 、b 2 is the fitting coefficient of the improved wet edge fitting equation.

[0031] Taking the EVI index as an example, its dry edge fitting equation can be expressed as: Ts max-m =a 1 +b 1 *EVI [0.2, EVImax] , the wet edge fitting equation can be expressed as Ts min-m =a 2 +b 2 *EVI [0.2, EVImax] The improvement is judged by comparing the determination coefficients of the dry side fitting equation and the wet side fitting equation before and after the improvement. The larger the determination coefficient value, the better the dry side and wet side fitting effect. The determination coefficient is expressed as R 2 .

[0032] Step 6: Calculate the improved dry edge fitting equation and wet edge fitting equation TVD index TVD m .

[0033] Specifically, refer to TVD The index is calculated by , the traditional calculation fits the dry side and wet side equations: Ts max =a 1 +b 1 *VI, Ts min =a 2 +b 2 *VI. Ts Represents the surface temperature of any pixel in the remote sensing image. Ts max for Ts The highest surface temperature corresponding to the VI value of the pixel, Ts min It is the lowest surface temperature corresponding to the VI value of Ts pixel.

[0034] Similarly, improved TVD index( TVD m, modified Temperature Vegetation DrynessIndex) can be expressed as: , where Tsmax-m , Ts min-m From step 5, Ts is the surface temperature of any pixel in the remote sensing image.

[0035] Step 7: Analysis TVD m Correlation with the measured soil moisture on the surface, select effective TVD m .

[0036] Specifically, the measured soil moisture data on the ground and the field water holding capacity data of the corresponding stations were obtained to ensure that the date of each sampling data was consistent with the remote sensing calculation. TVD m The relative soil moisture (RSM) of the stations at each period is calculated. TVD m The linear fitting relationship between the values; where RSM = (soil moisture / field capacity) × 100%; TVD m The value is the corresponding pixel of the ground measured data in the same period TVD m Value; Evaluated using the coefficient of determination of the linear fit relationship TVD m The degree of fit with the ground monitoring soil moisture is judged by the reference table of determination coefficient significance. TVD m The number of paired samples was determined to determine the significant correlation at the level of p < 0.05, and the different periods that met the significant correlation were TVD m Considered valid TVD m .

[0037] Step 8: Based on the information of the main crops in the study area, the starting time of the crop growth stage is divided and the suitable relative soil moisture for the crop is determined; according to the determined starting time of each growth stage, the effective soil moisture for each period of the year is calculated. TVD m Classify into corresponding growth stages.

[0038] Different crop growth stages have different sensitivities to water stress. The present invention divides crop growth stages and determines the appropriate relative soil moisture range for each growth stage. Different growth stages may include one or more TVD m Image. Valid images may not be included. If valid images are not included, the dynamic drought threshold for this stage will continue to use the threshold of the previous stage.

[0039] Step nine, determine the dynamic drought threshold for each growth stage.

[0040] Specifically, obtain all TVD m Scatter plot of measured soil moisture; determine the appropriate relative soil moisture range corresponding to TVD m value, and use the percentile method to determine the stage TVD m The percentile method includes: first TVD m The values ​​are sorted, and then the specified percentile position is calculated. In this embodiment, 90% is selected. If the calculated percentile position is not an integer, interpolation is performed between adjacent data points to obtain a more accurate percentile value and use it as the TVDI m The upper limit means that the data will not exceed this value in a given percentage. This method can reflect the overall trend of the data while avoiding excessive influence of extreme values ​​on the results. The upper limit is set as the threshold for drought in this stage, and the value greater than the upper limit indicates drought. TVD m Therefore, the dynamic drought thresholds at different growth stages were determined.

[0041] Step 10: According to the dynamic drought thresholds of each growth stage, TVD m , determine the dynamic drought threshold of the corresponding growth stage, and then TVD m The area is classified as drought-free zone, which is greater than or equal to TVD m The areas with dynamic thresholds are classified as drought areas.

[0042] tú2 for TVD , TVD m Comparison of dry and wet edge fitting in space, where the EVI index at different periods is selected TVD , TVD m Dry and wet edge fitting diagram. Taking some periods as an example, tú2 (a) to (f) are February 6, March 12, June 11, August 30, October 1, and December 25. The left pictures in (a) to (f) are TVD Spatial dry and wet edge fitting, the right figure is improved TVD m Spatial dry and wet edge fitting. 2 You can see the improved TVD m The fitting coefficient of the spatial dry and wet edges is significantly improved. tú3It's from different periods TVD I m The correlation coefficient with soil moisture in different soil layers, specifically the correlation coefficient between TVDIm and soil relative humidity in 0-10cm, 10-20cm, and 20-40cm soil layers at different periods of the year. tú4 Different crops in different growth cycles TVD m Scatter plot of relative soil moisture. For easy comparison, the horizontal axis value is 1- TVD m . tú4 In the figure, (a) is wheat and from left to right are the sowing period, wintering period, jointing and booting period, filling period and maturity period. (b) is corn and from left to right are the jointing period to heading period, heading period to milky period, and milky period to maturity period. Table 1 shows the division of crop growth stages and the determined dynamic drought thresholds.

[0043] Table 1. Crop growth stage division and determination of dynamic drought thresholds

[0044] The above-described embodiments of the present invention do not limit the protection scope of the present invention.

Claims

1. A drought monitoring method combining an improved TVDI index and a dynamic drought threshold, characterized in that: include: Step 1: Obtain the land surface temperature image data and vegetation index image data of the required period of the study area, the data covering the entire study area and the complete growing season, and pre-process and screen the land surface temperature image data and vegetation index image data; Step 2: Extract the maximum and minimum surface temperature values ​​corresponding to different vegetation index values ​​from the screened data, and establish a surface temperature-vegetation index scatter distribution diagram, with the horizontal axis coordinate being the vegetation index value and the vertical axis coordinate being the maximum and minimum surface temperature values ​​corresponding to the vegetation index value; the maximum surface temperature is called the dry edge of space, and the minimum surface temperature is called the wet edge of space; Step 3: Based on the surface temperature-vegetation index scatter plot, analyze whether the maximum surface temperature and the minimum surface temperature in the scatter space are consistent with the following theory as the vegetation index value increases from small to large: the dry side temperature increases with the increase of vegetation coverage, and the wet side temperature decreases with the increase of vegetation coverage; find the critical value VI of the vegetation index that does not conform to this theory threshold , filter out vegetation index values ​​less than the critical value, and retain the critical value to the maximum vegetation index value VI max The effective vegetation index range is determined as [VI threshold , VI max ]; Step 4: Based on the effective vegetation index range, a spatial relationship diagram between the effective vegetation index and the surface temperature is established, where the horizontal axis coordinate is the vegetation index value, and the vertical axis coordinate is the maximum surface temperature value and the minimum surface temperature value corresponding to the vegetation index value; Step 5, based on the spatial relationship diagram of the effective vegetation index and the surface temperature, the improved dry edge fitting equation and the wet edge fitting equation are calculated; Step 6: Calculate the improved TVDI index based on the improved dry side fitting equation and the wet side fitting equation TVDI m ; Step 7: Analysis TVDI m Correlation with the measured soil moisture on the surface, select effective TVDI m ; Step 8: Based on the information of the main crops in the study area, the starting time of the crop growth stage is divided and the suitable relative soil moisture for the crop is determined; according to the determined starting time of each growth stage, the effective soil moisture for each period of the year is calculated. TVDI m Classify into corresponding growth stages; Step nine, determining dynamic drought thresholds for each growth stage; Step 10: According to the dynamic drought thresholds of each growth stage, TVDI m , determine the dynamic drought threshold of the corresponding growth stage, and then TVDI m The area is classified as drought-free zone, which is greater than or equal to TVDI m Areas with dynamic thresholds are classified as drought zones.

2. A drought monitoring method combining an improved TVDI index and a dynamic drought threshold as claimed in claim 1, characterized in that: In step one, the vegetation index image data is normalized vegetation index image data or enhanced vegetation index image data; the surface temperature image data and vegetation index image data are preprocessed to ensure that the data are consistent in time and spatial resolution; in data screening, it is required that the proportion of effective surface temperature pixels in the study area in the images of each period is more than 85% of the study area.

3. A drought monitoring method combining improved TVDI index and dynamic drought threshold according to claim 1, characterized in that: Step five includes: For the improved fitting of dry edge and wet edge, the fitted linear correlation between the vegetation index and the maximum surface temperature and the minimum surface temperature in the effective scatter plot is calculated respectively to obtain the improved dry edge fitting equation and the wet edge fitting equation; Improved dry edge fitting equation: T s max-m =a1+b1*VI [VIthreshold, VImax] ; T s max-m For the corresponding VI [VIthreshold, VImax] The highest surface temperature of the value; a1, b1 are the fitting coefficients of the improved dry edge fitting equation; Improved wet edge fitting equation: Ts min-m =a2+b2*VI [VIthreshold, VImax] ; Ts min-m For the corresponding VI [VIthreshold, VImax] The lowest surface temperature of the value; a2 and b2 are the fitting coefficients of the improved wet edge fitting equation.

4. A drought monitoring method combining improved TVDI index and dynamic drought threshold as claimed in claim 3, characterized in that: Step 5 and beyond also include: The improvement is determined by comparing the determination coefficients of the dry side fitting equation and the wet side fitting equation before and after the improvement.

5. A drought monitoring method combining improved TVDI index and dynamic drought threshold as claimed in claim 4, characterized in that: In step 6, the improved TVDI index is calculated according to the following formula: TVDI m : , T s max-m For the corresponding VI [VIthreshold, VImax] The maximum surface temperature of the Ts min-m For the corresponding VI [VIthreshold, VImax] The minimum surface temperature of the Ts is the surface temperature of any pixel in the remote sensing image.

6. A drought monitoring method combining improved TVDI index and dynamic drought threshold as claimed in claim 5, characterized in that: Step seven includes: Obtain the measured soil moisture data on the ground and the field water holding capacity data of the corresponding stations to ensure that the date of each sampling data is consistent with the remote sensing calculation. TVDI m The relative soil moisture (RSM) of the stations at each period is calculated. TVDI m The linear fitting relationship between the values; where RSM = (soil moisture / field capacity) × 100%; TVDI m The value is the corresponding pixel of the ground measured data in the same period TVDI m value; Estimation of the coefficient of determination using the linear fit relationship TVDI m The degree of fit with the ground monitoring soil moisture is judged by the reference table of determination coefficient significance. TVDI m The number of paired samples was determined to determine the significant correlation at the level of p < 0.05, and the different periods that met the significant correlation were TVDI m Considered valid TVDI m .

7. A drought monitoring method combining improved TVDI index and dynamic drought threshold according to claim 1, characterized in that: Step nine includes: Get all the TVDI m Scatter plot with measured soil moisture; Determine the appropriate relative soil moisture range TVDI m value, and use the percentile method to determine the stage TVDI m The maximum value of TVDI m The values ​​are sorted, and then the position of the specified percentile is calculated. If the calculated percentile position is not an integer, interpolation is performed between adjacent data points to obtain a more accurate percentile value and use it as TVDI m The upper limit of The upper limit is set as the threshold for possible drought in this stage, and the period above this upper limit may indicate drought. TVDI m scope.

Images