A drought early warning method and system based on vegetation phenology

Abstract

This invention discloses a drought early warning method and system based on vegetation phenology, comprising: basic data preparation and preprocessing; establishing a baseline phenological period; constructing a daily-scale phenological weight curve; identifying drought events based on soil moisture index; calculating the annual-scale comprehensive drought stress; calculating the phenological disturbance value for the target year; constructing a phenological cumulative disturbance model and calculating a comprehensive phenological disorder index; simulating and assessing drought scenario risks based on the phenological cumulative disturbance model, generating a phenological disorder risk map; and providing drought early warning and precise guidance based on the risk map. As an assessment and early warning scheme that adapts to the characteristics of climate change and comprehensively considers both cumulative drought stress and phenological sensitivity, this invention can provide a scientific basis and precise guidance for drought management, effectively improving the initiative and adaptability in responding to drought disasters.

Description

Technical Field

[0001] This invention relates to drought monitoring and ecological early warning methods, and more particularly to a drought early warning method and system based on vegetation phenology. Background Technology

[0002] In recent years, the frequency, intensity, and duration of drought events have changed significantly under the background of climate change, with extreme drought events becoming more frequent and widespread. The timing window for drought occurrence is also shifting, challenging traditional phenological patterns based on historical statistics and placing higher demands on drought management. The response of vegetation phenology to drought exhibits significant stage-specific characteristics: drought of the same intensity occurring at different phenological stages has vastly different impacts on vegetation growth. For example, crops are most sensitive to water stress at the end of their growth cycle, while they are less sensitive to drought during dormancy. Existing drought assessment methods primarily focus on the intensity and duration of drought itself, failing to fully consider the differences in vegetation sensitivity to water stress at different growth stages. Furthermore, conventional assessment methods considering a single drought event cannot comprehensively assess the long-term disturbances of drought to vegetation phenology; and traditional drought early warning systems are mostly based on real-time monitoring or simple threshold judgments, lacking forward-looking scenario simulation capabilities. Summary of the Invention

[0003] Purpose of the invention: To address the above problems, this invention proposes a drought early warning method and system based on vegetation phenology, which can more accurately assess the impact of drought events on plants and improve the foresight of existing drought early warning systems.

[0004] Technical solution: The drought early warning method based on vegetation phenology described in this invention includes:

[0005] Obtain vegetation index data and soil moisture data;

[0006] The baseline phenological period of vegetation is obtained from vegetation index data; drought events are identified from soil moisture data.

[0007] A daily-scale phenological weight curve is constructed based on the baseline phenological period. The phenological weight curve is used to characterize the physiological sensitivity of vegetation to water stress at different growth and development stages.

[0008] Based on the phenological weight curve and the identified drought events, calculate the annual-scale comprehensive drought stress; calculate the phenological disturbance value for the target year based on vegetation index data;

[0009] Based on the annual-scale comprehensive drought stress and phenological disturbance values, a cumulative phenological disturbance model is constructed and a comprehensive phenological disorder index is calculated.

[0010] Based on the aforementioned phenological cumulative disturbance model, drought scenario risks are simulated and assessed, and a phenological disorder risk map is generated.

[0011] Drought warnings and guidance are provided based on the aforementioned phenological disorder risk map and comprehensive phenological disorder index.

[0012] The method of obtaining the baseline phenological period of vegetation based on vegetation index data includes: using a dynamic threshold method on the vegetation index data, extracting the start date and end date of the growing season each year, removing extremely dry years, and then calculating the baseline start date and end date of the growing season.

[0013] The construction of a daily-scale phenological weight curve based on the baseline phenological period includes: dividing the year into several phenological stages with the start and end dates of the baseline growing season as nodes, and assigning corresponding weights to each phenological stage based on the differences in vegetation sensitivity to drought in different phenological stages, as phenological sensitivity weights.

[0014] Based on the phenological weight curve and the identified drought events, the annual-scale comprehensive drought stress is calculated, including: calculating the drought intensity weight based on the soil moisture index; calculating the joint weight based on the drought intensity weight and the phenological sensitivity weight to obtain the comprehensive drought stress for each drought event; and summing the comprehensive drought stress according to the drought events to obtain the comprehensive drought stress for the year.

[0015] The drought intensity weight is calculated using a nonlinear function, and the calculation formula is as follows:

[0016] ;

[0017] In the formula, S d SMP is the intensity weight for the d-th drought day. d Let d be the soil moisture index on day d, where ε is a positive number close to 0 and is a very small constant (e.g., 0.01) used to avoid the denominator being zero;

[0018] The formula for calculating the total drought stress for each drought event is as follows:

[0019] ;

[0020] In the formula, ED i Let W be the total drought stress of the i-th drought event. d S represents the phenological sensitivity weight corresponding to the d-th drought day. d Let d be the drought intensity weight for the d-th drought day, and n be the total number of days in the i-th drought event.

[0021] The calculation of the phenological disturbance value of the target year based on vegetation index data includes: extracting the start date and end date of the growing season of the target year, and calculating the growing season start offset ΔSOS, growing season end offset ΔEOS, and growing season length loss LGS in combination with the baseline phenological period.

[0022] Based on the annual comprehensive drought stress and phenological disturbance values, a cumulative phenological disturbance model is constructed and a comprehensive phenological disorder index is calculated. This includes: using nonlinear functions to fit the relationships between the annual comprehensive drought stress and the growing season start offset, growing season end offset, and growing season length loss, respectively. The calculation formula is as follows:

[0023] ;

[0024] ;

[0025] ;

[0026] In the formula, ΔSOS is the growing season start offset, ΔEOS is the growing season end offset, LGS is the growing season length loss, and EDY is the annual-scale comprehensive drought stress. , , , , , These are the parameters of the cumulative phenological disturbance model corresponding to different phenological disturbance values;

[0027] The formula for calculating the comprehensive phenological disorder index is as follows:

[0028] ;

[0029] In the formula, ΔSOS_ano and ΔEOS_ano are the normalized values ​​of the growing season start offset and growing season end offset in the phenological disturbance values ​​of the target year calculated based on vegetation index data.

[0030] This invention provides a drought early warning system based on vegetation phenology, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned drought early warning method based on vegetation phenology.

[0031] The present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described drought early warning method based on vegetation phenology.

[0032] The present invention also provides a computer program product, including a computer program / instruction that, when executed by a processor, implements the aforementioned drought early warning method based on vegetation phenology.

[0033] Beneficial Effects: Compared with existing technologies, this invention has the following beneficial effects: 1. This invention divides the year into five phenological stages by constructing a diurnal phenological sensitivity weight curve, and assigns differentiated weights based on the sensitivity of vegetation to water stress at different growth stages. Compared with existing technologies that use fixed sensitivity coefficients or coarse stage divisions, this invention can more accurately reflect the matching relationship between drought occurrence time and phenological stages, significantly improving the accuracy of drought impact assessment. 2. We found that drought has a cumulative effect on vegetation. The cumulative impact of multiple mild droughts may be no less than that of a single severe drought. Early drought events can change the physiological state of vegetation, affecting its response to subsequent droughts. This invention, by constructing a nonlinear drought intensity weight function and an annual-scale comprehensive drought stress quantity, achieves dual weighting of drought event intensity, duration, and occurrence time, and quantifies the cumulative impact of multiple drought events into equivalent stress days. Compared with existing technologies that only consider a single drought event or the current vegetation state, this invention can comprehensively reflect the long-term disturbance of vegetation phenology by early drought events, which is more in line with the ecological laws of vegetation response to drought. 3. This invention defines a comprehensive phenological disorder index (PDI), which, through normalized growing season start and end offsets, achieves a quantitative comprehensive assessment of the overall degree of phenological cycle disorder. Compared with existing methods that only focus on a single phenological indicator, this invention can more comprehensively characterize the multidimensional impact of drought on vegetation phenology. 4. This invention defines standardized drought events of different intensities and durations, using their start dates as variables to slide across the annual time series, generating a phenological disorder risk map showing how phenological disturbance values ​​change with the start date of drought. Compared with existing early warning methods based on real-time monitoring or simple threshold judgments, this invention can answer the question, "If this drought occurred at different times and with different intensities, what would be the impact?" achieving a leap from passive response to proactive early warning. 5. This invention integrates the phenological disorder risk map with a vegetation geographic information system, classifies four early warning levels based on the comprehensive phenological disorder index, and matches specific drought early warning response measures to each level. Compared with existing technologies that only provide drought indices and lack subsequent guidance, this invention realizes the transformation from "scientific indicators" to "management instructions," providing full-chain decision support for precise drought relief. Attached Figure Description

[0034] Figure 1 This is a flowchart of the drought early warning method based on vegetation phenology described in this invention;

[0035] Figure 2 It is a division of phenological stages for a certain type of vegetation. Detailed Implementation

[0036] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0037] The drought early warning method based on vegetation phenology described in this invention is illustrated in the flowchart below. Figure 1 As shown, it includes:

[0038] Obtain vegetation index data and soil moisture data;

[0039] The baseline phenological period of vegetation is obtained from vegetation index data; drought events are identified from soil moisture data.

[0040] A daily-scale phenological weight curve is constructed based on the baseline phenological period. The phenological weight curve is used to characterize the physiological sensitivity of vegetation to water stress at different growth and development stages.

[0041] Based on the phenological weight curve and the identified drought events, calculate the annual-scale comprehensive drought stress; calculate the phenological disturbance value for the target year based on vegetation index data;

[0042] Based on the annual-scale comprehensive drought stress and phenological disturbance values, a cumulative phenological disturbance model is constructed and a comprehensive phenological disorder index is calculated.

[0043] Based on the aforementioned phenological cumulative disturbance model, drought scenario risks are simulated and assessed, and a phenological disorder risk map is generated.

[0044] Drought warnings and guidance are provided based on the aforementioned phenological disorder risk map and comprehensive phenological disorder index.

[0045] The following details the specific steps of the above method, which are not in any particular order.

[0046] Step 1: Basic data preparation and preprocessing.

[0047] Long-term vegetation index (such as NDVI, EVI, etc.) and soil moisture data were acquired within the study area. Vegetation index data can be obtained using MODIS or similar sensors with a time resolution of 8 or 16 days; this index is used to extract phenological information. Soil moisture data can be obtained from station-measured data or daily root zone soil moisture data assimilated by models such as ERA5. To eliminate the influence of seasonal variations on soil moisture, the raw soil moisture data needs to be uniformly converted to Soil Moisture Percentile (SMP).

[0048] Step 2: Establish baseline phenological periods.

[0049] For the long-term vegetation index data obtained in Step 1, the growing season start date (SOS) and growing season end date (EOS) were extracted annually using a dynamic thresholding method. After removing extremely drought years, the baseline growing season start date (SOS) was calculated. baseand the benchmark growing season end date EOS base .

[0050] Step 3: Construct daily-scale phenological weighted curves.

[0051] Based on the baseline phenological date, a diurnal phenological sensitivity weighting function is established. The weights characterize the physiological sensitivity of vegetation to water stress at different growth and development stages, and are core parameters for subsequent calculations of drought stress. The baseline growing season start date (SOS) is used as the starting point. base ) and the baseline growing season end date (EOS base Using nodes as points, the year is divided into five phenological stages, as shown in Table 1. Figure 2 The growth sequence of a certain vegetation in this embodiment is divided into five phenological stages.

[0052] Table 1. Division of Phenological Stages

[0053]

[0054] Based on the differences in crop sensitivity to drought at different phenological stages, the sensitivity of vegetation to water stress varies significantly at different growth stages. Accordingly, the resetting parameters for each phenological stage are shown in Table 2.

[0055] Table 2 Sensitivity Weights and Physiological Significance of Phenological Stages

[0056]

[0057] This mode is suitable for quick assessments or areas with limited historical data, and users can also fine-tune the weight values ​​based on local experience.

[0058] Step 4: Identify drought events based on soil moisture index.

[0059] Based on daily soil moisture index time series, independent drought events are identified according to the following rules: (1) A drought event is considered to have started when the soil moisture index is below the 40th percentile; a drought event is considered to have ended when the soil moisture index recovers to or exceeds the 40th percentile. (2) There is at least one point in a drought event where the soil moisture index is below the 20th percentile. (3) The drought lasts at least two weeks. For each identified drought event, the drought occurrence timeline during its duration is recorded, including the time and daily drought intensity (soil moisture index SMP for each day).

[0060] Step 5: Calculate the annual-scale composite drought stress.

[0061] A combined drought stress (ED) was calculated by weighting the phenological stage of the drought event and the drought intensity, which was used to assess the equivalent number of stress days of this drought event converted to "standard growing season and standard intensity". The phenological sensitivity weight (W) was used to calculate the overall drought stress. d The drought intensity weight (S) is calculated in step three. d The value is calculated using soil moisture index, and this function ensures that the drier the soil, the higher the drought intensity weight S. d The larger it is. The specific formula is:

[0062] ;

[0063] Where ε is a very small constant greater than 0 (e.g., 0.01) to avoid division by zero errors. This function ensures that the drier the soil, the higher the drought intensity weight S. d The non-linear, rapid growth pattern is more consistent with the physiological response characteristics of vegetation to extreme drought. Therefore, the formula for calculating the total drought stress of the i-th drought event is as follows:

[0064] ;

[0065] Among them, W d S represents the phenological sensitivity weight corresponding to the d-th drought day. d Let d be the intensity weight of the drought day, and n be the total number of days in this drought event. Based on this, the equivalent stress days of all m drought events in the target year are summed to obtain the total drought stress amount (EDY) for that year.

[0066] ;

[0067] Step 6: Calculate the phenological disturbance value for the target year.

[0068] Based on long-series vegetation index data, phenological dates and growing season periods are extracted for each year, including the start date of the growing season (SOS). y ) and the end date of the growing season (EOS) y The annual phenological disturbance values ​​are calculated by combining the baseline phenological period from step two. The specific calculation formula is as follows:

[0069] ;

[0070] ;

[0071] ;

[0072] Wherein, ΔSOS is the offset at the start of the growing season, with a positive value indicating a delay; ΔEOS is the offset at the end of the growing season, with a negative value indicating an earlier end; and LGS is the loss in the length of the growing season, with a positive value indicating a shortened growing season.

[0073] Step 7: Construct a cumulative phenological disturbance model and calculate the disturbance index.

[0074] Using the phenological disturbance value (Y) calculated in step six and the annualized comprehensive drought stress (EDY) calculated in step five, a phenological cumulative disturbance model is constructed. This invention recommends using a nonlinear model to fit the relationship between drought cumulative stress and the growing season start offset, growing season end offset, and growing season length loss, respectively. The calculation formula is as follows:

[0075] ;

[0076] ;

[0077] ;

[0078] In the formula, EDY represents the annual-scale composite drought stress. , , , , , These are the model parameters fitted based on historical data with different perturbation values. Furthermore, to provide an overall evaluation index, the Phenological Disorder Index (PDI) is defined as follows:

[0079] ;

[0080] Where ΔSOS_ano and ΔEOS_ano are the normalized values ​​of ΔSOS and ΔEOS (subtracted from the mean and divided by the standard deviation), respectively. PDI is a dimensionless scalar; the larger the value, the greater the overall disorder of the phenological cycle caused by the cumulative stress of soil drought.

[0081] Step 8: Simulate and assess drought scenario risks based on phenological cumulative disturbance model.

[0082] This step aims to utilize the phenological cumulative disturbance model, validated with historical data and constructed in Step Seven, to conduct prospective scenario simulations to assess the potential disturbance risks of different drought scenarios to phenology. The specific implementation process is as follows:

[0083] ① Define a series of standardized drought events with different drought intensities and durations. Drought intensity is defined by soil moisture index (SMP) thresholds, for example, setting SMP to 0.2, 0.1, 0.05, and 0.02 to represent four levels: moderate drought, severe drought, extreme drought, and exceptional drought, respectively. Drought duration can be set to a fixed value (e.g., 60 days).

[0084] ② For each defined drought scenario, with its start date as the variable, slide it across the annual time series at a fixed step size (e.g., 5 days or 10 days). For each sliding window (i.e., each possible period during which the drought of that level occurs), calculate its corresponding annual integrated drought stress (EDY) and substitute it into the phenological cumulative disturbance model in step seven to calculate the corresponding phenological disturbance prediction value (Y).

[0085] ③ Integrate the simulation results of all scenarios to generate a family of curves or a risk distribution map showing the variation of phenological disturbance values ​​(or the comprehensive phenological disorder index PDI) with the start date of drought. This map can visually reveal when phenologically sensitive critical periods and specific levels of drought will trigger unacceptable phenological disturbances.

[0086] Step Nine: Drought Early Warning and Precision Guidance Based on Risk Maps.

[0087] The "Phenological Disorder Risk Map" generated in step eight is integrated with the vegetation geographic information system. When weather forecasts or real-time monitoring indicate that a certain level of drought is about to occur in a region, the system can automatically call upon the map data to issue an early warning of the main phenological impacts that the drought event may cause, and generate a recommended plan for the optimal irrigation period and irrigation amount, thereby achieving "on-demand irrigation and precise drought resistance." Using the actual phenological disorder composite index (PDI) for the target year calculated in step seven, combined with historical sequences, the warning level is determined, as shown in Table 3:

[0088] Table 3. Early Warning Levels and Comprehensive Response Measures for Phenological Disorders and Drought

[0089]

[0090] The aforementioned PDI thresholds are recommended reference values, and can be dynamically adjusted based on the characteristics of regional historical data in specific applications. Furthermore, the spatiotemporal evolution of the annual PDI can be used as an indicator to assess ecological vulnerability, for long-term water resource allocation and planting structure optimization. Through the above steps, this invention achieves a full-chain service from "drought monitoring - model evaluation - risk prediction - decision support," quantifying abstract drought stress into specific phenological risks and management instructions, significantly improving the initiative, foresight, and scientific rigor of drought management.

[0091] In one embodiment, a drought early warning system based on vegetation phenology is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described drought early warning method based on vegetation phenology.

[0092] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described drought early warning method based on vegetation phenology.

[0093] In one embodiment, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the aforementioned drought early warning method based on vegetation phenology.

[0094] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0095] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0096] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0097] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

Claims

1. A drought early warning method based on vegetation phenology, characterized in that, include: Obtain vegetation index data and soil moisture data; The baseline phenological period of vegetation is obtained based on vegetation index data; Identify drought events based on soil moisture data; A daily-scale phenological weight curve is constructed based on the baseline phenological period. The phenological weight curve is used to characterize the physiological sensitivity of vegetation to water stress at different growth and development stages. Based on the phenological weight curve and the identified drought events, the annual-scale comprehensive drought stress is calculated, including: calculating the drought intensity weight based on soil moisture index; calculating the joint weight based on the drought intensity weight and phenological sensitivity weight to obtain the comprehensive drought stress for each drought event; and summing the comprehensive drought stress for each drought event to obtain the annual-scale comprehensive drought stress. The drought intensity weight is calculated using a nonlinear function, and the calculation formula is as follows: ； In the formula, S d SMP is the intensity weight for the d-th drought day. d Let ε be the soil moisture index on day d, where ε is a positive number to avoid a denominator of zero. The phenological disturbance value for the target year is calculated based on vegetation index data, including: extracting the start date and end date of the growing season for the target year, and combining the baseline phenological period to calculate the growing season start offset ΔSOS, growing season end offset ΔEOS, and growing season length loss LGS. Based on the annual comprehensive drought stress and phenological disturbance values, a cumulative phenological disturbance model is constructed and a comprehensive phenological disorder index is calculated. This includes: using nonlinear functions to fit the relationships between the annual comprehensive drought stress and the growing season start offset, growing season end offset, and growing season length loss, respectively. The calculation formula is as follows: ； ； ； In the formula, ΔSOS is the growing season start offset, ΔEOS is the growing season end offset, LGS is the growing season length loss, and EDY is the annual-scale comprehensive drought stress. , , , , , These are the parameters of the cumulative phenological disturbance model corresponding to different phenological disturbance values; Based on the aforementioned phenological cumulative disturbance model, drought scenario risks are simulated and assessed, and a phenological disorder risk map is generated. Drought warnings and guidance are provided based on the aforementioned phenological disorder risk map and comprehensive phenological disorder index.

2. The drought early warning method based on vegetation phenology according to claim 1, characterized in that: The method of obtaining the baseline phenological period of vegetation based on vegetation index data includes: using a dynamic threshold method on the vegetation index data, extracting the start date and end date of the growing season each year, removing extremely dry years, and then calculating the baseline start date and end date of the growing season.

3. The drought early warning method based on vegetation phenology according to claim 1, characterized in that: The construction of a daily-scale phenological weight curve based on the baseline phenological period includes: dividing the year into several phenological stages with the start and end dates of the baseline growing season as nodes, and assigning corresponding weights to each phenological stage based on the differences in vegetation sensitivity to drought in different phenological stages, as phenological sensitivity weights.

4. The drought early warning method based on vegetation phenology according to claim 1, characterized in that: The formula for calculating the total drought stress for each drought event is as follows: ； In the formula, ED i Let W be the total drought stress of the i-th drought event. d S represents the phenological sensitivity weight corresponding to the d-th drought day. d Let d be the drought intensity weight for the d-th drought day, and n be the total number of days in the i-th drought event.

5. The drought early warning method based on vegetation phenology according to claim 1, characterized in that: The formula for calculating the comprehensive phenological disorder index is as follows: ； In the formula, ΔSOS_ano and ΔEOS_ano are the normalized values ​​of the growing season start offset and growing season end offset in the phenological disturbance values ​​of the target year calculated based on vegetation index data.

6. A drought early warning system based on vegetation phenology, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements a drought early warning method based on vegetation phenology as described in any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements a drought early warning method based on vegetation phenology as described in any one of claims 1 to 5.

8. A computer program product, comprising a computer program and / or instructions, characterized in that, When the computer program and / or instructions are executed by the processor, they implement a drought early warning method based on vegetation phenology as described in any one of claims 1 to 5.

Images