Forest drought monitoring method, equipment and medium

Abstract

The invention discloses a forest drought monitoring method and device and a medium, and relates to the field of drought detection, and the method comprises the steps: carrying out the normalization processing of a Keetch-Byram drought index in a target forest region and a standardized rainfall index under a preset time scale, and inputting the Keetch-Byram drought index and a normalized vegetation index into a forest drought monitoring model, obtaining a corresponding forest drought index value; the preset duration scale comprises a seasonal scale and an annual scale; matching the forest drought index values based on a preset drought grade threshold to obtain a drought grade corresponding to the target forest region; wherein the forest drought monitoring model is determined based on a plurality of weight coefficients, and the weight coefficients are used for representing the importance of a Keetch-Byram drought index, a normalized vegetation index or a standardized rainfall index under a preset duration scale in forest drought monitoring. Therefore, forest drought monitoring can be realized rapidly and accurately.

Description

Technical Field

[0001] The present application relates to the field of drought monitoring, and in particular to a forest drought monitoring method, equipment and medium. Background Art

[0002] Forestry drought is part of agricultural drought and has comprehensive and multifaceted impacts on forests and trees, such as forestry productivity, vegetation growth, and surface microorganisms. From the perspective of water resources, forestry drought is an unbalanced water shortage situation where water supply cannot meet normal water demand; the drought index can accurately reflect the rapid changes in humidity and has temporal and spatial comparability. Monitoring the drought index helps make decisions to alleviate drought in the early stages. However, at present, there is no unified standard for forestry drought indicators, and most of them are calculated based on a single drought-related index, which has low accuracy. Summary of the invention

[0003] The purpose of this application is to provide a forest drought monitoring method, equipment and medium, which can quickly and accurately realize forest drought monitoring.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a forest drought monitoring method, comprising:

[0006] Obtaining the Keetch-Byram drought index, normalized difference vegetation index and standardized precipitation index at a preset time scale in the target forest area; the preset time scale includes a seasonal scale and an annual scale;

[0007] After normalizing the Keetch-Byram drought index and the standardized precipitation index under the preset time scale, the index and the normalized vegetation index are input into the forest drought monitoring model to obtain the corresponding forest drought index value;

[0008] Based on a preset drought level threshold, the forest drought index value is matched to obtain the drought level corresponding to the target forest area;

[0009] Among them, the forest drought monitoring model is determined based on multiple weight coefficients, and the weight coefficients are used to characterize the importance of the Keetch-Byram drought index, the normalized vegetation index or the standardized precipitation index under a preset time scale in forest drought monitoring.

[0010] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a forest drought monitoring method.

[0011] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements a forest drought monitoring method when executed by a processor.

[0012] According to the specific embodiments provided in this application, this application discloses the following technical effects: This application provides a forest drought monitoring method, equipment and medium, which combines the Keetch-Byram drought index, the normalized vegetation index and the standardized precipitation index under a preset time scale in the target forest area to monitor forest drought, so that the forest drought index value finally obtained is more practical and more accurate. A forest drought monitoring model is also constructed in this application. Through this pre-constructed model, the forest drought index value can be obtained more quickly, and the weights set in the model can adaptively adjust the importance of the Keetch-Byram drought index, the normalized vegetation index and the standardized precipitation index under a preset time scale in the calculation process of the forest drought index value, so that the obtained forest drought index value is more accurate and feasible. In summary, the calculation standard of the forestry drought index value is determined in this application, combined with the drought level threshold, which can quickly and accurately realize forest drought monitoring, and is of great significance in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0014] Figure 1 This is an application environment diagram of a forest drought monitoring method in one embodiment of the present application;

[0015] Figure 2 A schematic diagram of a process flow of a forest drought monitoring method provided in one embodiment of the present application;

[0016] Figure 3 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] This application organically combines the forest drought index value with the meteorological and remote sensing drought index, which can effectively reflect the large-scale drought conditions and secondary effects in the region, and provide scientific reference and technical support for drought monitoring and fire prevention in forest areas.

[0019] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0020] The forest drought monitoring method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, or it can be integrated on the server 104, or it can be placed on the cloud or other servers. The terminal 102 can send the Keetch-Byram drought index, the normalized vegetation index and the standardized precipitation index under the preset time scale in the target forest area to the server 104. After receiving the above three data, the server 104 first performs normalization processing, and then calls the forest drought monitoring model to obtain the corresponding forest drought index value, and finally matches based on the preset drought level threshold to obtain the corresponding drought level. The server 104 can feedback the obtained drought level of the target forest area to the terminal 102. In addition, in some embodiments, the forest drought monitoring method can be implemented by the server 104 or the terminal 102 alone, or the server 104 can directly perform data processing.

[0021] The terminal 102 may be, but is not limited to, various desktop computers, laptop computers, tablet computers, and IoT devices. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers, or may be a cloud server.

[0022] In an exemplary embodiment, Figure 2 As shown, a forest drought monitoring method is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or can be executed by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used for explanation, and the steps include the following steps 201 to 203.

[0023] Step 201, obtain the Keetch-Byram drought index, normalized vegetation index and standardized precipitation index at a preset time scale in the target forest area. In an application example, the forest type in the target forest area can be coniferous forest or deciduous broad-leaved forest, and the coniferous forest and deciduous broad-leaved forest spatially correspond to the cold-temperate and temperate forest zones in the vegetation horizontal-vertical zone map.

[0024] In an application example, the process of determining the standardized precipitation index at a preset time scale in the target forest area is as follows:

[0025] First of all, it should be noted that forestry is less sensitive to drought. Therefore, this application is based on meteorological drought indicators, takes into account the underlying surface type (forest area), and describes forestry drought by calibrating the drought level threshold. Furthermore, the standard precipitation index can effectively reflect the drought conditions in various regions and periods, has statistical consistency, can be calculated on different time scales, has a simple calculation method, is easy to obtain data, and can eliminate differences in the use of precipitation. Therefore, this application obtains the standard precipitation index for forest drought monitoring.

[0026] Secondly, the change of monthly precipitation in a certain period does not obey the normal distribution, but a Gamma distribution. Therefore, when calculating the standardized precipitation index, the precipitation with skewed probability distribution needs to be normalized. The calculation method is as follows:

[0027] Assuming that the precipitation in a certain period is a random variable x, the probability density function of its Γ distribution is:

[0028]

[0029] Among them, β>0, γ>0, β is the scale parameter, γ is the shape parameter, which can be obtained by the maximum likelihood estimation method:

[0030]

[0031] Among them, x i Sample data for historical precipitation data: The multi-year average of rainfall.

[0032] After determining the parameters in the probability density function, for the precipitation x0 in a certain year, the probability of the random variable x being less than x0 can be calculated as:

[0033]

[0034] The probability of an event when precipitation is 0 is estimated by the following formula:

[0035] P(x=0)=m / n.

[0036] Among them, m is the number of samples when the precipitation is 0; n is the total number of samples.

[0037] The Standardized Precipitation Index (SPI) is calculated by calculating the probability of the Γ distribution of precipitation in a certain period, and then normalizing it. Finally, the standardized cumulative frequency distribution of precipitation is used to divide the drought level. The SPI calculation formula based on the Γ fitting function is as follows:

[0038]

[0039] in: G(x) is the precipitation distribution probability associated with the Γ function; S is the positive and negative coefficient of the probability density, when G(x)>0.5, S=1; when G(x)≤0.5, S=-1.

[0040] G(x) is calculated by the probability density integral formula of the Γ distribution function, as follows:

[0041]

[0042] Among them, γ and β are the shape and scale parameters of the Γ distribution function, respectively, which are obtained by maximum likelihood estimation method: c0=2.515517, c1=0.802853, c2=0.010328, d1=1.432788, d2=0.189269, d3=0.001308.

[0043] The Standardized Precipitation Index (SPI) is a drought index that is based on the cumulative frequency distribution of precipitation and reflects the climate characteristics of precipitation in different times and regions. It is a drought index that relies solely on precipitation. The SPI is based on the fact that the shortage of precipitation will affect the changes in surface water, storage water, soil moisture, snow accumulation and flow at a certain time and space scale. It can be provided by the ground observation stations of the meteorological department and is suitable for the calculation of drought conditions in different regions at any time.

[0044] In an application example, two time scales are set, that is, the preset time scales include the season scale and the year scale, for example, the season scale is set to 3 months, and the year scale is set to 24 months. By determining the standardized precipitation index of these two scales, precipitation data at different times can be integrated, so that precipitation conditions within different time ranges can be added to forest drought monitoring, providing a data basis for achieving more accurate predictions.

[0045] In an application example, the process of determining the Keetch-Byram drought index in the target forest area is as follows:

[0046] First of all, it should be explained that the Keetch-Byram Drought Index (KBDI) is a cumulative algorithm for estimating potential fire risk based on weather information, including daily maximum temperature, daily total precipitation and annual rainfall. The index comprehensively considers precipitation, temperature and surface conditions related to drought, reflects the physical mechanism of drought, effectively determines the starting time of drought, and uses a recursive method to reflect the cumulative effect of drought. Furthermore, the Keetch-Byram Drought Index estimates the potential forest fire risk based on the input meteorological parameters by estimating the water loss of the upper soil and litter layer. It is an early warning tool for potential fire risk. The model theory of the Keetch-Byram Drought Index is based on the following assumptions:

[0047] (1) The loss of water from forest land depends on the vegetation cover density in the area. The vegetation density and its transpiration capacity are functions of the average annual rainfall, and the vegetation can make full use of the available water through its own regulation ability.

[0048] (2) The relationship between vegetation and rainfall is approximately an exponential relationship. The water loss rate is a function of the average annual rainfall, and the water loss rate decreases with the decrease in vegetation density, which is also accompanied by a decrease in the average annual rainfall.

[0049] (3) The rate of water loss from the soil is determined by transpiration.

[0050] (4) The loss of soil moisture is exponentially related to time. The lowest moisture content is represented by the wilting point. Therefore, under similar conditions, the rate at which soil moisture decreases from the initial state to the wilting point is proportional to the total amount of moisture obtained by the soil layer in a given time.

[0051] (5) The field water holding capacity of the soil layer where the drought occurred is 20 cm (8 inches). The reason for using 20 cm water holding capacity is that most forest vegetation where forest fires occur evaporates about this much water in the summer, and the exact value is not required in the calculation.

[0052] Secondly, the KBDI is initialized. Keetch and Byram (1968) suggested that 15 cm to 20 cm of precipitation in a week is sufficient for initialization. If the climate zone does not typically meet the initialization criteria, the KBDI is not suitable as a monitoring tool. The calculation process of the KBDI is as follows:

[0053] The calculation formula of KBDI is: t =KBDI t-1 –(P t )+DF t .

[0054] When P t =0cm, TMAX tKBDI at ≤6.78℃ t =KBDI t-1 .

[0055] When P t =0cm, TMAX t >6.78℃,KBDI t =KBDI t-1 +DF t .

[0056] When P t >0cm,∑P t KBDI ≤0.51cm t =KBDI t-1 +DF t .

[0057] When P t >0cm,∑P t >0.51cm, KBDI t =KBDI t-1 -P t +DF t .

[0058] Among them, DF t is the amount of water loss caused by transpiration. Assuming that the vegetation intercepts 5 mm of rainfall, the net rainfall P t Defined as: Net rainfall = 24-hour rainfall – 5 mm. The 5 mm interception is only used on the first day of rainfall. If the rainfall lasts for two days or more, the second day and subsequent rainfall will no longer remove the interception.

[0059] DF t It is an index of daily maximum temperature and annual rainfall, and is calculated as follows:

[0060]

[0061] Among them, T max is the daily maximum temperature (℃); R is the average annual rainfall (cm); KBDI t-1 is the KBDI of day t-1. The above formula uses the average annual precipitation as the effect of vegetation density on transpiration to estimate the relationship between soil moisture loss and litter moisture content. In the formula, 6.78℃ is a limit value. When the maximum temperature is lower than 6.78℃, DF t =0; when the maximum temperature is higher than 6.78℃, DF t It is a power exponential growth; when the precipitation within 48 hours is less than 0.51 cm, the moisture changes of soil and litter layer will not be considered, and KBDI will not decrease. In this application, the KBDI value range is between 0 and 200. The fire risk meaning represented by KBDI is shown in Table 1 below.

[0062] Table 1

[0063]

[0064] KBDI requires fewer meteorological factors to be input, and can be promoted and applied by grassroots fire prevention departments. In practical applications, the existing meteorological station data can be fully utilized to draw a drought index map in a target area in near real time every day, which will help to understand the early fire danger situation in a large area and provide a reference for the rational deployment of fire fighting resources.

[0065] In an application example, the Normalized Difference Vegetation Index (NDVI) is selected as a remote sensing monitoring indicator for forest drought. The forest area distribution is used as the underlying surface, and the drought level threshold applicable to the forest area is used as the classification standard to judge the forest drought level. Specifically, the determination process of the Normalized Difference Vegetation Index in the target forest area is as follows:

[0066]

[0067] Among them, CH1 and CH2 represent the albedo of the 1st and 2nd channels of NOAA / AVHRR, respectively.

[0068] Step 202, after normalizing the Keetch-Byram drought index and the standardized precipitation index under the preset time scale, input them into the forest drought monitoring model together with the normalized vegetation index to obtain the corresponding forest drought index value.

[0069] In an application example, the standardized precipitation index at the preset time scale includes the standardized precipitation sub-value at the seasonal scale and the standardized precipitation sub-value at the annual scale; the following normalization processing, the construction of the forest drought monitoring model, etc. are all based on the standardized precipitation sub-value at the seasonal scale and the standardized precipitation sub-value at the annual scale as examples. If the preset time scale is 3, 4 or more, it can be derived by analogy.

[0070] The Keetch-Byram drought index and the standardized precipitation index at the preset time scale are normalized by using the Min-Max Normalization method to scale the value range of each element to 0 to 1. The specific process is as follows:

[0071] The standardized precipitation sub-values ​​at the seasonal scale are normalized using the following formula:

[0072]

[0073] The standardized precipitation subvalues ​​at the annual scale are normalized using the following formula:

[0074]

[0075] The Keetch-Byram drought index is normalized using the following formula:

[0076]

[0077] Where Y is the original value, Y∈[SPI season ,SPI year ,KBDI]; Xnorm is the normalized Y, Ymin and Ymax are the minimum and maximum values ​​of Y respectively; SPI season is the seasonal standardized precipitation subvalue, SPI year is the standardized precipitation subvalue on an annual scale, and KBDI is the Keetch-Byram Drought Index.

[0078] Among them, the forest drought monitoring model is determined based on multiple weight coefficients, and the weight coefficients are used to characterize the importance of the Keetch-Byram drought index, the normalized vegetation index or the standardized precipitation index under a preset time scale in forest drought monitoring.

[0079] Methods for determining multiple weight coefficients in the forest drought monitoring model include: expert scoring method, statistical analysis method, machine learning algorithm and multi-objective optimization method. Expert scoring method: experts score the importance of each factor, and then determine the weight based on the score; although this method combines the experience of experts, it has the problem of strong subjectivity. Statistical analysis method: through statistical analysis of historical data, determine the contribution of each factor to the drought situation; for example, the correlation between each factor and the actual drought event can be calculated, and the weight can be determined according to the size of the correlation. Machine learning algorithm: using machine learning algorithm to automatically determine the weight is a modern and effective method. Multi-objective optimization method: considering multiple objectives (such as accuracy, sensitivity, etc.), a multi-objective optimization algorithm (such as genetic algorithm) is used to determine the optimal weight combination of each factor.

[0080] The machine learning algorithm is one of linear regression, random forest, decision tree, and gradient boosting tree. Decision tree: Use the decision tree model to analyze the impact of each factor on drought, and determine the weight according to the split point and split number of each factor in the tree. Random forest: Use the random forest model to determine the importance of each factor through the integration method of multiple decision trees, and determine the weight accordingly. Gradient boosting tree (GBDT): Use the GBDT model to determine the weight of each factor through step-by-step optimization.

[0081] Although the models adopted by the above-mentioned multiple machine learning algorithms are different, the process of constructing a forest drought monitoring model based on the machine learning algorithm is roughly the same. Based on this, when the method for determining multiple weight coefficients in the forest drought monitoring model is a machine learning algorithm, the process of constructing the forest drought monitoring model includes:

[0082] (21) Obtain a historical forest data set in a sample forest area; the forest type in the sample forest area is the same as the forest type in the target forest area; the historical forest data set includes a historical Keetch-Byram drought index, a historical normalized difference vegetation index, a historical standardized precipitation index at a preset time scale, and a corresponding historical forest drought index value in different historical periods.

[0083] In an application example, the type of trees in the sample forest area can be coniferous forest or deciduous broad-leaved forest. In order to obtain a wider sample data base, multiple ground observation meteorological stations can be set up in the sample forest area to obtain the corresponding forest historical data set in the sample forest area. In the algorithm call, the forest historical data set can be stored in a CSV file.

[0084] (22) After normalizing the historical Keetch-Byram drought index and the historical standardized precipitation index at a preset time scale in the forest historical data set, the training sample set and the test sample set are determined by combining the normalized vegetation index and the corresponding historical forest drought index value.

[0085] The normalization process in this step is the same as the normalization calculation process described above, which can scale the value range to 0 to 1, and will not be repeated here. It should be noted that the normalization process can ensure that all components are on the same scale, thereby improving the accuracy and scientificity of the forest drought index value (FCDI).

[0086] After the normalization operation is completed, the normalized vegetation index, the normalized historical Keetch-Byram drought index and the historical standardized precipitation index under the preset time scale are used as input variables, and the corresponding historical forest drought index values ​​are used as labels. An input variable and the corresponding label constitute a sample, and then all samples are divided into a training sample set and a test sample set.

[0087] (23) Using the training sample set to train the preset machine learning model, and using the test sample set to test and adjust the preset machine learning model after training, so as to obtain an optimized machine learning model. The test sample set can test or verify the model to evaluate its prediction accuracy; the test sample set can also adjust the model parameters and weights to ensure the robustness and stability of the model.

[0088] (24) Based on the optimized machine learning model, the importance of the Keetch-Byram drought index, the normalized vegetation index, and the standardized precipitation index under a preset time scale in forest drought monitoring is determined, and marked as a first weight coefficient, a second weight coefficient, and a third weight coefficient; wherein, corresponding to the preset time scale including the seasonal scale and the annual scale, the third weight coefficient includes a seasonal weight coefficient and an annual weight coefficient.

[0089] (25) Based on the first weight coefficient, the second weight coefficient and the third weight coefficient, a forest drought monitoring model is constructed.

[0090] In a specific application example, after the first weight coefficient, the second weight coefficient and the third weight coefficient are obtained in the above step (24), the importance scores of the Keetch-Byram drought index, the normalized vegetation index, and the standardized precipitation index under a preset time scale in forest drought monitoring are determined based on expert opinions, and are marked as the first importance score, the second importance score and the third importance score respectively; based on the first importance score, the second importance score and the third importance score, the first weight coefficient, the second weight coefficient and the third weight coefficient are adjusted and updated; based on the updated first weight coefficient, the second weight coefficient and the third weight coefficient, the forest drought monitoring model is updated.

[0091] That is, in practical applications, machine learning algorithms can be combined with expert opinions, which have the advantages of both subjective opinions and objective calculations, thereby ensuring the accuracy of multiple weight coefficients that are ultimately used to construct a forest drought monitoring model.

[0092] After obtaining the final multiple weight coefficients, the function formula of the forest drought monitoring model is:

[0093] FCDI=a×SPI season norm+b×SPI year norm+c×KBDInorm+d×NDVI.

[0094] Among them, FCDI is the forest drought index value, monthly value; SPI season norm is the standardized precipitation value on a seasonal scale after normalization, SPI year norm is the standardized precipitation sub-value on the annual scale after normalization; KBDInorm is the normalized Keetch-Byram drought index, the monthly average; NDVI is the normalized vegetation index, the monthly average; a, b, c, d are all weight coefficients, which are 0.25, 0.35, -0.10, and 0.30 respectively in an application example.

[0095] Step 203, matching the forest drought index value based on a preset drought level threshold to obtain the drought level corresponding to the target forest area.

[0096] In an application example, the drought levels of various single drought indices are comprehensively considered, combined with long sequence analysis data, and compared with the historical drought records of the entire region, the drought level classification of the grassland comprehensive drought index is obtained. Specifically, the drought levels include no drought, light drought, moderate drought, severe drought, and extreme drought; the preset drought level thresholds corresponding to each of the drought levels are: no drought is [0, ∞); light drought is [-0.5, 0); moderate drought is [-1.0, -0.5); severe drought is [-2.0, -1.0); extreme drought is [-∞, -2.0), as shown in Table 2 below.

[0097] Table 2

[0098] Drought level Threshold No drought 0≤FCDI Mild drought -0.5≤FCDI<0 Moderate Drought -1.0≤FCDI<-0.5 Severe drought -2.0≤FCDI<-1.0 Extreme drought FCDI<-2.0

[0099] In summary, this application starts from the physical mechanism of forest drought, comprehensively considers indicators such as meteorological drought, remote sensing drought and forest fire risk drought, obtains more objective and accurate weight coefficients, and then constructs a forest drought monitoring model to calculate the forest drought index value, thereby effectively integrating and compensating for the advantages and disadvantages of each single drought index. This application organically combines remote sensing and meteorological drought indices, which can not only reflect the short-term duration of meteorological drought from point data, but also reflect the regional large-scale drought conditions from aspects such as vegetation and soil moisture and secondary effects of drought. In addition, the distribution trend of the comprehensive drought index and each single drought index on the time scale is significant, which is manifested as the stronger the solar radiation in spring and summer, the more significant the drought risk.

[0100] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the forest drought monitoring method is implemented.

[0101] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0102] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0103] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0104] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0105] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0106] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0107] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A forest drought monitoring method, characterized in that: The forest drought monitoring method comprises: Obtaining the Keetch-Byram drought index, normalized difference vegetation index and standardized precipitation index at a preset time scale in the target forest area; the preset time scale includes a seasonal scale and an annual scale; After normalizing the Keetch-Byram drought index and the standardized precipitation index under the preset time scale, the index and the normalized vegetation index are input into the forest drought monitoring model to obtain the corresponding forest drought index value; Based on a preset drought level threshold, the forest drought index value is matched to obtain the drought level corresponding to the target forest area; Among them, the forest drought monitoring model is determined based on multiple weight coefficients, and the weight coefficients are used to characterize the importance of the Keetch-Byram drought index, the normalized vegetation index or the standardized precipitation index under a preset time scale in forest drought monitoring.

2. The forest drought monitoring method according to claim 1, characterized in that: The method for determining multiple weight coefficients in the forest drought monitoring model includes: expert scoring method, statistical analysis method, machine learning algorithm and multi-objective optimization method.

3. The forest drought monitoring method according to claim 1, characterized in that: When the method for determining the multiple weight coefficients in the forest drought monitoring model is a machine learning algorithm, the construction process of the forest drought monitoring model includes: Acquire a historical data set of trees in a sample forest area; the tree type in the sample forest area is the same as the tree type in the target forest area; the historical data set of trees includes a historical Keetch-Byram drought index, a historical normalized vegetation index, a historical standardized precipitation index at a preset time scale, and corresponding historical forest drought index values ​​in different historical periods; After normalizing the historical Keetch-Byram drought index and the historical standardized precipitation index at a preset time scale in the forest history data set, combining the normalized vegetation index and the corresponding historical forest drought index value, a training sample set and a test sample set are determined; The training sample set is used to train the preset machine learning model, and the test sample set is used to test and adjust the preset machine learning model after the training is completed, so as to obtain an optimized machine learning model; Based on the optimized machine learning model, the importance of the Keetch-Byram drought index, the normalized vegetation index, and the standardized precipitation index under a preset time scale in forest drought monitoring is determined and marked as a first weight coefficient, a second weight coefficient, and a third weight coefficient; wherein, corresponding to the preset time scale including a seasonal scale and an annual scale, the third weight coefficient includes a seasonal weight coefficient and an annual weight coefficient; Based on the first weight coefficient, the second weight coefficient and the third weight coefficient, a forest drought monitoring model is constructed.

4. The forest drought monitoring method according to claim 3, characterized in that: When the method for determining the multiple weight coefficients in the forest drought monitoring model is a machine learning algorithm, the process of constructing the forest drought monitoring model further includes: After obtaining the first weight coefficient, the second weight coefficient and the third weight coefficient, based on expert opinions, the importance scores of the Keetch-Byram drought index, the normalized vegetation index and the standardized precipitation index at a preset time scale in forest drought monitoring are determined, and marked as the first importance score, the second importance score and the third importance score respectively; Based on the first importance score, the second importance score, and the third importance score, adjusting and updating the first weight coefficient, the second weight coefficient, and the third weight coefficient; The forest drought monitoring model is updated based on the updated first weight coefficient, the second weight coefficient and the third weight coefficient.

5. The forest drought monitoring method according to claim 3, characterized in that: The machine learning algorithm is one of linear regression, random forest, decision tree, and gradient boosting tree.

6. The forest drought monitoring method according to claim 1, characterized in that: The standardized precipitation index at the preset time scale includes a standardized precipitation sub-value at a seasonal scale and a standardized precipitation sub-value at an annual scale; The Keetch-Byram drought index and the standardized precipitation index at the preset time scale are normalized, including: The standardized precipitation sub-values ​​at the seasonal scale are normalized using the following formula: The standardized precipitation subvalues ​​at the annual scale are normalized using the following formula: The Keetch-Byram drought index is normalized using the following formula: Where Y is the original value, Y∈[SPI season ,SPI year ,KBDI]; Ynorm is the normalized Y, Ymin and Ymax are the minimum and maximum values ​​of Y respectively; SPI season is the seasonal standardized precipitation subvalue, SPI year is the standardized precipitation subvalue on an annual scale, and KBDI is the Keetch-Byram Drought Index.

7. The forest drought monitoring method according to claim 1, characterized in that: The standardized precipitation index at the preset time scale includes the standardized precipitation sub-value at the seasonal scale and the standardized precipitation sub-value at the annual scale; the function formula of the forest drought monitoring model is: FCDI=a×SPI season norm+b×SPI year norm+c×KBDInorm+d×NDVI; Among them, FCDI is the forest drought index value, SPI season norm is the standardized precipitation value on a seasonal scale after normalization, SPI year norm is the standardized precipitation sub-value on the annual scale after normalization; KBDInorm is the normalized Keetch-Byram drought index; NDVI is the normalized vegetation index; a, b, c, and d are all weight coefficients.

8. The forest drought monitoring method according to claim 1, characterized in that: The drought levels include no drought, mild drought, moderate drought, severe drought and extreme drought; The preset drought level thresholds corresponding to each of the drought levels are: no drought is [0, ∞); light drought is [-0.5, 0); moderate drought is [-1.0, -0.5); severe drought is [-2.0, -1.0); extreme drought is [-∞, -2.0).

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the forest drought monitoring method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the forest drought monitoring method described in any one of claims 1 to 8 is implemented.

Images