A method and system for identifying groundwater-dependent vegetation

Abstract

The present application relates to the technical field of ecological remote sensing monitoring, in particular to a groundwater-dependent vegetation identification method and system, the method obtains rainfall observation data, soil effective water holding capacity parameters and vegetation effective root layer depth parameters of a target area, combines daily reference evapotranspiration data to construct a soil water balance model to calculate the critical day since the effective rainfall event occurs; within the time window after the critical day, whether the vegetation canopy maintains a water non-stress state is determined according to the difference between the canopy temperature and the temperature of the same period obtained by thermal infrared remote sensing inversion; when it is determined that the vegetation canopy still maintains a water non-stress state after the critical day, the groundwater level monitoring data of the topographic unit where the target pixel is located is obtained, the recognition result is verified according to the diurnal fluctuation characteristics of the groundwater level monitoring data, and the groundwater-dependent vegetation identification result is obtained. The accuracy and reliability of groundwater-dependent vegetation identification are improved.

Description

Technical Field

[0001] This invention relates to the field of ecological remote sensing monitoring technology, specifically a method and system for identifying groundwater-dependent vegetation. Background Technology

[0002] Groundwater-dependent vegetation is widely distributed in oasis-desert transition zones in arid and semi-arid regions. This type of vegetation absorbs shallow groundwater through its deep-rooted systems to maintain its own water supply, playing an irreplaceable role in windbreak and sand fixation, inhibiting desertification expansion, and maintaining the stability of oasis ecological barriers. Accurately identifying groundwater-dependent vegetation and clarifying its spatial distribution is of significant practical importance for the rational allocation of regional water resources and the formulation of ecological protection and restoration measures for desert-oasis transition zones. With the development of remote sensing inversion technology, indirectly inferring vegetation water sources using remote sensing parameters such as evapotranspiration and surface temperature has become an important technical approach for large-scale, low-cost identification of groundwater-dependent vegetation.

[0003] However, in extremely arid inland river basins, precipitation events are characterized by low frequency, small magnitude, and strong intermittency. Soil moisture often undergoes a dynamic process from abundant to gradually depleted after a single rainfall replenishment, exhibiting a clear distinction between the effective period of rainfall replenishment and the period of replenishment depletion. During the effective period of replenishment, vegetation can rely on the temporary water supply formed by rainfall infiltration into the soil to maintain normal transpiration. At this time, even if remote sensing parameters such as vegetation canopy temperature and evapotranspiration level are in a non-stressed state, it does not necessarily indicate that the moisture originates from groundwater. Only when soil moisture is depleted and the replenishment depletion period begins, if the vegetation can still maintain a non-stressed state of moisture, does it indicate the presence of groundwater replenishment.

[0004] Therefore, there is an urgent need for a groundwater-dependent vegetation identification method that can reasonably distinguish between the effective period and the depletion period of precipitation recharge, avoid the deviation of remote sensing evapotranspiration anomaly judgment caused by confusion between the two, and thus improve the accuracy and reliability of groundwater-dependent vegetation identification. Summary of the Invention

[0005] (1) Technical problems to be solved The purpose of this invention is to provide a groundwater-dependent vegetation identification method and system to solve the problem of inaccurate identification of vegetation groundwater dependence in the oasis-desert transition zone of extremely arid inland river basins, where the remote sensing evapotranspiration anomaly criteria are not distinguished between the effective period and depletion period of precipitation recharge, resulting in inaccurate identification of vegetation groundwater dependence.

[0006] (2) Technical solution To achieve the above objectives, in one aspect, the present invention provides a method for identifying groundwater-dependent vegetation, the method comprising: S1. Obtain rainfall observation data, soil effective water holding capacity parameters, and vegetation effective root layer depth parameters for the target area. Combine daily reference evapotranspiration data to construct a soil water balance model and calculate the critical day since the most recent effective rainfall event for each pixel.

[0007] S2. Within the time window after the critical day, obtain the difference between the canopy temperature retrieved by thermal infrared remote sensing and the air temperature at the same time, and determine whether the vegetation canopy maintains a water-free state based on the difference.

[0008] S3. When it is determined that the vegetation canopy still maintains a non-water-stressed state after the critical day, the groundwater level monitoring data of the topographic unit where the target pixel is located is obtained, and the identification result is verified according to the diurnal fluctuation characteristics of the groundwater level monitoring data to obtain the groundwater-dependent vegetation identification result.

[0009] Furthermore, the method for constructing a soil water balance model and calculating the critical day since the most recent effective rainfall event on a pixel-by-pixel basis includes: Extract the occurrence date and rainfall amount of rainfall events that meet the preset effective rainfall determination conditions from rainfall observation data; obtain the soil effective water holding capacity parameter and vegetation effective root layer depth parameter of the pixel, and calculate the maximum effective water storage capacity of the soil root layer of the pixel.

[0010] The reference evapotranspiration values ​​for each day from the date of the rainfall event are obtained from the daily reference evapotranspiration data. The actual daily water consumption conversion factor is determined by combining the vegetation cover characteristics of the pixel, and the cumulative daily water consumption is calculated.

[0011] The daily cumulative water consumption is compared with the maximum effective water storage capacity. When the daily cumulative water consumption reaches the maximum effective water storage capacity at a preset ratio, the corresponding date is determined as the critical day for this rainfall event for that pixel.

[0012] Furthermore, the method for obtaining the effective soil water holding capacity parameter and the effective root layer depth parameter of the pixel, and calculating the maximum effective water storage capacity of the soil root layer of the pixel includes: Based on the remote sensing inversion results of soil texture, the field water holding capacity and wilting coefficient of the pixel are obtained, and the effective water holding capacity per unit depth of soil is calculated.

[0013] Based on the vegetation type classification results of the pixel, the empirical root layer depth range of the corresponding vegetation type is obtained. The empirical root layer depth range is then corrected by combining the groundwater depth data of the pixel to obtain the effective root layer depth of the pixel.

[0014] Multiply the effective water holding capacity of the soil per unit depth by the effective root layer depth to obtain the maximum effective water storage capacity of the soil root layer of that pixel.

[0015] Furthermore, the method for determining the daily actual water consumption conversion factor by combining the vegetation cover characteristics of the pixel and calculating the daily cumulative water consumption includes: Obtain the daily vegetation cover remote sensing inversion results for the pixel, determine the crop coefficient of the pixel based on the vegetation cover, and multiply the reference evapotranspiration value by the crop coefficient to obtain the daily potential water consumption of the pixel.

[0016] Obtain the estimated soil moisture content of the pixel for the previous day, determine the water stress reduction coefficient based on the ratio of the estimated soil moisture content to the effective soil water holding capacity parameter, and multiply the daily potential water consumption by the water stress reduction coefficient to obtain the daily actual water consumption of the pixel.

[0017] The actual daily water consumption from the date of the rainfall event is added up to obtain the cumulative daily water consumption.

[0018] Furthermore, the method for obtaining the difference between the canopy temperature retrieved by thermal infrared remote sensing and the air temperature during the same period within a time window after the critical day, and determining whether the vegetation canopy maintains a water-free state based on the difference, includes: The canopy temperature of the pixel obtained by thermal infrared remote sensing inversion at each observation time after the critical day, and the near-surface air temperature observation value at the corresponding time, are used to calculate the canopy-air temperature difference at each time step.

[0019] Determine the stable range of canopy-temperature difference corresponding to the non-stressed water condition based on the vegetation type to which the pixel belongs.

[0020] Determine whether the canopy-temperature difference at each observation time after the critical day continues to fall within a stable range. If it continues to fall within the stable range, then determine that the vegetation canopy of that pixel maintains a non-water-stressed state after the critical day.

[0021] Furthermore, the method for determining the stable range of the canopy-temperature difference corresponding to the non-stress state of water based on the vegetation type to which the pixel belongs includes: The canopy-temperature difference values ​​of the vegetation type to which the pixel belongs were collected at multiple sample times during historical periods, and were confirmed to be in a non-water-stressed state, to form a reference sample set.

[0022] Calculate the mean and standard deviation of the canopy-temperature difference values ​​in the reference sample set. Use the sum of the mean and the standard deviation of a preset multiple as the upper limit of the stable range, and use the difference between the mean and the standard deviation of the preset multiple as the lower limit of the stable range.

[0023] Obtain near-surface wind speed and air humidity data for the observation day, and correct the upper and lower limits of the stability range based on the near-surface wind speed and air humidity data to obtain the corrected stability range.

[0024] Furthermore, the method for obtaining groundwater level monitoring data of the topographic unit where the target pixel is located, and verifying the identification results based on the diurnal fluctuation characteristics of the groundwater level monitoring data to obtain groundwater-dependent vegetation identification results includes: Determine whether the topographic unit where the target pixel is located has the conditions for groundwater level monitoring. If it does, obtain hourly groundwater level observation data of groundwater level monitoring points within the topographic unit for several consecutive days.

[0025] The highest and lowest daily groundwater levels are extracted from hourly groundwater level observation data, and the daily diurnal fluctuation range of groundwater levels is calculated.

[0026] Determine whether the daily diurnal fluctuation range of groundwater level conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration. If it does, confirm that the pixel is a groundwater-dependent vegetation. If it does not, maintain the judgment result obtained by the critical day and the canopy-temperature difference joint criterion and mark it as pending review.

[0027] Furthermore, the method for determining whether the daily diurnal fluctuation range of groundwater level conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration includes: Obtain the water yield parameters of the aquifer where the groundwater level monitoring point is located, and calculate the daily groundwater evapotranspiration consumption corresponding to the pixel based on the daily diurnal fluctuation range of the groundwater level and the water yield parameters.

[0028] The daily groundwater evapotranspiration loss is compared with the daily potential water loss during the period after the critical day when the water stress state is determined, and the deviation ratio between the two is calculated.

[0029] When the deviation ratio is less than the preset consistency threshold, it is determined that the daily fluctuation range of groundwater level conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration.

[0030] Furthermore, the method also includes: For a pixel in a topographic unit that does not have groundwater level monitoring conditions, the topographic similarity parameters between the pixel and neighboring pixels that do have groundwater level monitoring conditions are obtained. The topographic similarity parameters include elevation difference, groundwater depth difference, and river distance difference.

[0031] When the landform similarity parameters meet the preset similarity conditions, the critical day and canopy-temperature difference joint criterion parameters corresponding to the verified groundwater-dependent vegetation determination results of neighboring pixels are migrated to that pixel as auxiliary criteria for groundwater-dependent vegetation identification in that pixel.

[0032] The recognition result of the pixel is corrected based on the joint criterion parameters after migration, and the corrected groundwater-dependent vegetation recognition result is output.

[0033] On the other hand, based on the same inventive concept, the present invention also provides a groundwater-dependent vegetation identification system, the system comprising: The soil water balance and critical day calculation module is used to acquire rainfall observation data, soil effective water holding capacity parameters, and vegetation effective root layer depth parameters of the target area. It combines daily reference evapotranspiration data to construct a soil water balance model and calculates the critical day since the most recent effective rainfall event on a pixel-by-pixel basis.

[0034] The canopy temperature difference and vegetation moisture status determination module is used to obtain the difference between the canopy temperature retrieved by thermal infrared remote sensing and the air temperature at the same time within a time window after the critical day, and to determine whether the vegetation canopy maintains a non-stressful state of moisture based on the difference.

[0035] The groundwater-dependent vegetation identification and analysis module is used to acquire groundwater level monitoring data of the topographic unit where the target pixel is located when it is determined that the vegetation canopy still maintains a water-free state after the critical day. The identification results are verified based on the diurnal fluctuation characteristics of the groundwater level monitoring data to obtain the groundwater-dependent vegetation identification results.

[0036] (3) Beneficial effects Compared with the prior art, the beneficial effects of the present invention are: This invention achieves accurate differentiation between the effective period and the depletion period of precipitation replenishment by constructing a soil water balance model and calculating the critical day pixel by pixel, thus avoiding misjudgment of remote sensing evapotranspiration anomalies caused by confusion between the two. It determines the non-stress state of water by inverting the canopy-temperature difference through thermal infrared remote sensing and combining it with a stable range determined by vegetation type. Furthermore, it dynamically corrects the criterion range by introducing near-surface wind speed and air humidity, enabling the judgment criteria to adapt to actual differences under different vegetation types and meteorological conditions. By introducing the diurnal fluctuation characteristics of groundwater level monitoring data for cross-validation, it compares the groundwater evapotranspiration water consumption with the actual water consumption during the canopy-temperature difference determination period, allowing the remote sensing inversion results to be independently verified by ground observation data. This significantly improves the accuracy and reliability of identifying groundwater-dependent vegetation in the oasis-desert transition zone of extremely arid inland river basins. Attached Figure Description

[0037] Figure 1 This is a flowchart of a groundwater-dependent vegetation identification method according to the present invention.

[0038] Figure 2 This is a schematic diagram of the module composition of a groundwater-dependent vegetation identification system according to the present invention. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] Example 1: As Figure 1 As shown in the figure, this embodiment provides a method for identifying groundwater-dependent vegetation, the method comprising: S1. Acquire rainfall observation data, soil effective water holding capacity parameters, and vegetation effective root layer depth parameters for the target area. Combine this with daily reference evapotranspiration data to construct a soil water balance model. The method for calculating the critical day since the most recent effective rainfall event on a pixel-by-pixel basis includes: The method for extracting the occurrence date and rainfall amount of rainfall events that meet the preset effective rainfall determination criteria from rainfall observation data; obtaining the soil effective water holding capacity parameter and vegetation effective root layer depth parameter of the pixel; and calculating the maximum effective water storage capacity of the soil root layer of the pixel includes: Based on the remote sensing inversion results of soil texture, the field water holding capacity and wilting coefficient of the pixel are obtained, and the effective water holding capacity per unit depth of soil is calculated.

[0041] Based on the vegetation type classification results of the pixel, the empirical root layer depth range of the corresponding vegetation type is obtained. The empirical root layer depth range is then corrected by combining the groundwater depth data of the pixel to obtain the effective root layer depth of the pixel.

[0042] Multiply the effective water holding capacity of the soil per unit depth by the effective root layer depth to obtain the maximum effective water storage capacity of the soil root layer of that pixel.

[0043] For example, the identification process of a specific Populus euphratica forest pixel (hereinafter referred to as the pixel) within an oasis-desert transition zone is used as an example. The spatial resolution of the pixel is 30m×30m, the groundwater depth is approximately 4.2m, and the vegetation type of the pixel is identified as Populus euphratica. After extracting rainfall observation data from the meteorological station or rainfall interpolation grid data corresponding to the pixel, it is necessary to first determine which rainfall events are "effective rainfall events" that can actually replenish soil moisture in the root zone, rather than "ineffective rainfall" that is lost due to surface evaporation or interception by sparse vegetation canopy and cannot infiltrate into the root zone. In this embodiment, the preset criteria for determining effective rainfall are: a single-day rainfall of not less than 10 mm, or a cumulative rainfall of not less than 15 mm over three consecutive days. This threshold is determined based on the empirical relationship between the infiltration rate and surface evaporation rate of sandy loam in the region—rainfall below this level is largely consumed by surface evaporation or vegetation interception before reaching the root zone, and is insufficient to cause substantial changes in the soil moisture content of the root zone; therefore, it is not included in the statistics. Taking this pixel as an example, if a rainfall event is observed on July 3rd of a certain year, with a rainfall of 18 mm, meeting the condition of a single-day rainfall of not less than 10 mm, it is determined to be an effective rainfall event, and this date is determined as the starting date of this soil water balance calculation (denoted as t=0).

[0044] After determining the date of the effective rainfall event, it is necessary to calculate the maximum effective water storage capacity of the root zone of the soil in that pixel. First, based on the soil texture remote sensing inversion results (in this embodiment, the surface soil of this pixel has a sand content of approximately 70% and a clay content of approximately 8% through joint inversion of multispectral and radar data, and is determined to be sandy loam), the field capacity of this pixel is estimated by combining the Saxton-Rawls soil transformation function. Withering coefficient In this embodiment Take 0.18 cm³ / cm³, Taking 0.10 cm³ / cm³ as an example, both are dimensionless volumetric water content, representing the effective water holding capacity of the soil per unit depth for this pixel. We obtain the following from the difference between the two: Substitute the values This value is a dimensionless volume fraction, representing the proportion of effective water that can be released per unit depth of soil between field capacity and wilting coefficient. Converted to commonly used hydrological units, it means that each meter of soil depth can provide 80 mm of effective water storage. ).

[0045] Subsequently, the empirical root layer depth range corresponding to Populus euphratica needs to be obtained by combining the vegetation type classification results of the pixel. In this embodiment, the empirical range of the total root system depth of Populus euphratica is 2-6m. This range usually covers two parts: the shallow fibrous root system responsible for water absorption and the deep taproot system that penetrates into the soil layer and can reach the groundwater level. Since the soil water balance model constructed by this method aims to characterize the dynamic process of "rainfall infiltration - daily consumption of soil moisture", this process is mainly dominated by the water response of the shallow fibrous root system. If the groundwater depth of the pixel is greater than the lower limit of the empirical root layer depth range (2m in this embodiment), it indicates that when estimated according to the lower limit of the empirical root layer depth, the taproot system cannot ensure stable contact with the groundwater level. If the total root system depth covering the deep taproot is still used in the water balance calculation, the effective water storage capacity that rainfall replenishment can maintain will be overestimated. Therefore, the effective root layer depth used for water balance calculation needs to be corrected to the main distribution depth of the shallow fibrous root system. In this embodiment, the groundwater depth of this pixel is 4.2m, which is greater than the empirical lower limit of root layer depth of 2m. Therefore, it is corrected according to the above rules. The main distribution depth of shallow fibrous roots can be obtained through root excavation surveys or published research data on the root distribution of similar vegetation. In this embodiment, based on the root distribution survey data of Populus euphratica in Ejina Oasis, it is determined that the shallow fibrous root system of this pixel is concentrated in the range of 0-0.8m below the ground surface. Therefore, the effective root layer depth is taken. It is 0.8m (i.e., 800mm). The maximum effective water storage capacity of the soil root zone for this pixel is... It is obtained by multiplying the effective water holding capacity per unit depth of soil by the effective root layer depth: In the formula This is a dimensionless volume fraction. Substitute the values ​​in mm. This value represents the maximum effective water storage capacity available for vegetation consumption after the root layer of the pixel replenishes soil moisture to field capacity during a single effective rainfall event.

[0046] The method for calculating daily cumulative water consumption by obtaining daily reference evapotranspiration values ​​from the date of the rainfall event from daily reference evapotranspiration data, and combining these with the vegetation cover characteristics of the pixel to determine the daily actual water consumption conversion factor, includes: Obtain the daily vegetation cover remote sensing inversion results for the pixel, determine the crop coefficient of the pixel based on the vegetation cover, and multiply the reference evapotranspiration value by the crop coefficient to obtain the daily potential water consumption of the pixel.

[0047] For example, after obtaining the maximum effective water storage capacity, the daily reference evapotranspiration values ​​from July 3rd need to be obtained from the daily reference evapotranspiration data. (In this embodiment, the area is in summer) The value is approximately 5.5 mm / d, which is consistent with the common observation level of potential evapotranspiration in summer in extremely arid areas. The daily actual water consumption conversion factor is determined based on the vegetation cover characteristics of the pixel, and the daily cumulative water consumption is calculated. Specifically, the daily vegetation cover remote sensing inversion results for the pixel are first obtained. In this embodiment, a pixel-based binary model based on the normalized vegetation index is used to calculate the vegetation cover. That is, the relative position of the measured normalized vegetation index of the pixel with the normalized vegetation index of the pixels with bare soil and pixels with pure vegetation cover is used to characterize it: ;in The measured normalized vegetation index (0.35) for this pixel. The NDVI of the bare soil pixel is set to 0.05. For pixels with pure vegetation cover, the NDVI (taken as 0.60) is calculated. All three are dimensionless indices. Substituting them into the calculation yields... The value is set to 0.55. The crop coefficient for this pixel is determined based on vegetation cover, i.e., a linear interpolation is performed between the bare soil crop coefficient and the full-coverage crop coefficient according to vegetation cover: ;in The crop coefficient under bare soil conditions is 0.15. The crop coefficient under full vegetation cover is given (1.20 for Populus euphratica forest; this value is based on the empirical range of 1.10–1.25 for similar arid region tree crop coefficients). Both are dimensionless empirical coefficients. Substituting them into the equation yields… The daily potential water consumption of a pixel is obtained by multiplying the reference evapotranspiration value by the crop coefficient: Substituting, we get This value represents the daily potential (water requirement) water consumption of the pixel under conditions of sufficient soil moisture and no water stress. It is an important reference benchmark for subsequent judgment on whether the vegetation maintains a non-stressed state.

[0048] Obtain the estimated soil moisture content of the pixel for the previous day, determine the water stress reduction coefficient based on the ratio of the estimated soil moisture content to the effective soil water holding capacity parameter, and multiply the daily potential water consumption by the water stress reduction coefficient to obtain the daily actual water consumption of the pixel.

[0049] The actual daily water consumption from the date of the rainfall event is added up to obtain the cumulative daily water consumption.

[0050] The daily cumulative water consumption is compared with the maximum effective water storage capacity. When the daily cumulative water consumption reaches the maximum effective water storage capacity at a preset ratio, the corresponding date is determined as the critical day for this rainfall event for that pixel.

[0051] For example, simultaneously obtain the estimated soil moisture content of that pixel from the previous day. (This value is derived daily from the soil water balance model itself, i.e., calculated from the remaining water storage of the previous day.) The water stress reduction coefficient is determined based on the ratio of the estimated soil moisture content to the effective soil water holding capacity parameter. Because both the numerator and denominator are multiplied by the effective root layer depth. Then, this ratio is numerically equivalent to the remaining water storage from the previous day. With maximum effective water storage The ratio, that is Both have the same physical meaning, representing the proportion of remaining usable water in the soil root zone to the maximum effective water storage capacity, with a value ranging from 0 to 1. A value closer to 1 indicates more abundant soil moisture and less suppression of vegetation water consumption. Multiplying the daily potential water consumption by the water stress reduction factor yields the daily actual water consumption of that pixel: ; The remaining water storage volume will be updated daily using the following formula: .

[0052] It should be noted that, after substituting the expression for daily actual water consumption into the formula for updating remaining water storage, we can obtain... This recursive relation is in Under the simplified condition of approximate constancy, the common ratio is... The proportional decay process has a closed-form expression: .

[0053] It should be noted that in actual implementation Daily changes, vegetation coverage It may also be updated daily, therefore It is not strictly constant; actual calculations should be performed iteratively daily. The above closed-form expression is only valid when... In a simplified scenario where the soil is approximately constant in the short term, the results of both methods for quickly estimating the critical day are usually quite close. Taking this pixel as an example, after the rainfall on July 3rd, the soil root zone is considered to have replenished to field capacity, i.e. After that, take Substitute the values ​​into the closed-form expression above for estimation. Then, compare the daily cumulative water consumption with the maximum effective water storage. For comparison, considering that the consumption process mathematically exhibits an asymptotic decay characteristic, theoretically the remaining water storage approaches zero, but it is difficult to precisely reach zero within a finite number of days. Therefore, this embodiment presupposes that when the daily cumulative water consumption reaches 95% of the maximum effective water storage (i.e., the remaining water storage drops to...), the remaining water storage will decrease to... When the soil moisture in the root zone of a pixel is essentially depleted (as shown below), the corresponding date is determined as the critical day for that pixel's current rainfall event. By reducing the remaining water storage to 5% of the maximum effective water storage, substituting this into the above closed-loop expression and solving for the number of days, we can obtain the expression for the number of days corresponding to the critical day: Substitute the values, , , ,have to Rounded up to 47 days. The 47th day from July 3rd (day 0) is August 19th (July 4th to July 31st, a total of 28 days, plus August 1st to August 19th, a total of 19 days, totaling 47 days). This is the critical day for this pixel. Substituting this number back into the closed-form expression and verifying, we obtain the remaining water storage on that day as... It is basically consistent with the 5% threshold (3.2mm), and the verification is consistent.

[0054] S2. Within a time window after the critical day, the method for obtaining the difference between the canopy temperature retrieved by thermal infrared remote sensing and the air temperature during the same period, and determining whether the vegetation canopy maintains a water-free state based on the difference, includes: The canopy temperature of the pixel obtained by thermal infrared remote sensing inversion at each observation time after the critical day, and the near-surface air temperature observation value at the corresponding time, are used to calculate the canopy-air temperature difference at each time step.

[0055] For example, the canopy temperature of the pixel is obtained from thermal infrared remote sensing inversion at each observation time after the critical day (in this embodiment, daily transit thermal infrared remote sensing data is used, supplemented by synchronous observations from ground meteorological stations). and the corresponding near-surface temperature observation values ​​at that time. Calculate the canopy-temperature difference at each time step: Because vegetation continuously transpires under non-water stress conditions, the canopy cools through transpiration, and its temperature is usually lower than the ambient air temperature at the same time. It is generally a negative value, and the larger the absolute value, the stronger the evaporative cooling effect.

[0056] Methods for determining the stable range of canopy-temperature difference under non-water stress conditions based on the vegetation type of a pixel include: The canopy-temperature difference values ​​of the vegetation type to which the pixel belongs were collected at multiple sample times during historical periods, and were confirmed to be in a non-water-stressed state, to form a reference sample set.

[0057] Calculate the mean and standard deviation of the canopy-temperature difference values ​​in the reference sample set. Use the sum of the mean and the standard deviation of a preset multiple as the upper limit of the stable range, and use the difference between the mean and the standard deviation of the preset multiple as the lower limit of the stable range.

[0058] Obtain near-surface wind speed and air humidity data for the observation day, and correct the upper and lower limits of the stability range based on the near-surface wind speed and air humidity data to obtain the corrected stability range.

[0059] Determine whether the canopy-temperature difference at each observation time after the critical day continues to fall within a stable range. If it continues to fall within the stable range, then determine that the vegetation canopy of that pixel maintains a non-water-stressed state after the critical day.

[0060] For example, to determine what the pixel has observed To determine whether a pixel is within a reasonable fluctuation range corresponding to a non-stress state, the stable range of the canopy-temperature difference corresponding to a water-free state must first be determined based on the vegetation type to which the pixel belongs. This involves collecting data from multiple sample times of Populus euphratica plants within a historical period that are confirmed to be in a water-free state (e.g., pixels near groundwater level monitoring points and historically verified as groundwater-dependent). Numerical values ​​form a reference sample set, and the mean of this sample set is calculated. with standard deviation In this embodiment , Take the preset multiple (Corresponding to approximately 95% confidence interval), the upper and lower limits of the stability range are obtained: Substituting the values, we get the initial stable range as follows: Near-surface wind speed and air humidity have a systematic impact on this stability range: increased wind speed enhances convective heat exchange between the canopy and the atmosphere, causing the canopy temperature to approach the air temperature, thereby reducing... The absolute value (i.e.) (Shifting towards zero); while decreased air humidity increases the demand for atmospheric evaporation, enhancing the transpiration cooling effect of vegetation, making... The absolute value increases (i.e.) (Shifting towards a more negative direction). Based on this, the near-surface wind speed for the day of observation is obtained. With air humidity The data is corrected for the stability range by adding the wind speed deviation term and the humidity deviation term. : In the formula , For reference wind speed and reference humidity (in this embodiment, 2 m / s and 50% are taken respectively, which are the average wind speed and average relative humidity of this landform unit over many years), in the formula... , This is an empirical correction coefficient, whose value can be obtained by performing a linear regression fit between the canopy-temperature difference values ​​at multiple sample times during the historical period of the vegetation type to which the pixel belongs, where the vegetation has been confirmed to be in a non-water-stressed state, and the corresponding near-surface wind speed and air humidity observation values. The wind speed and humidity coefficients obtained from the regression are then used as... , In this embodiment, regression fitting is performed based on historical observation data of the region to obtain... = , When the wind speed is higher than the reference value, the correction term shifts the range towards zero; when the humidity is lower than the reference value, the correction term shifts the range towards a more negative direction. Both directions are consistent with the aforementioned physical mechanism. The measured near-surface wind speed for this pixel on August 19th was 1.0 m / s (lower than the reference wind speed, indicating calm conditions in this transition zone), and the air humidity was 30% (significantly lower than the reference humidity, indicating typical hot and dry weather). Substituting these values ​​into the correction expression yields: ; Because low wind speed weakens convective cooling and low humidity enhances transpiration cooling, both factors correct in the same direction, causing the stability range to shift more negatively. The corrected stability range is: ; Observational data for five consecutive days following the critical date (August 19 to August 23) were collected and measured. They are respectively , , , , All fall within the corrected stability range Within the range and without exceeding the boundary, it is determined that the vegetation canopy of this pixel maintained a non-stressed state of water after the critical day—that is, even when the soil root layer water was basically exhausted, the vegetation could still maintain normal transpiration and heat dissipation, initially indicating that there is a water supply source other than soil moisture.

[0061] S3. When it is determined that the vegetation canopy still maintains a water-free state after the critical day, the groundwater level monitoring data of the topographic unit where the target pixel is located is obtained, and the identification result is verified based on the diurnal fluctuation characteristics of the groundwater level monitoring data. The method for obtaining the groundwater-dependent vegetation identification result includes: Determine whether the topographic unit where the target pixel is located has the conditions for groundwater level monitoring. If it does, obtain hourly groundwater level observation data of groundwater level monitoring points within the topographic unit for several consecutive days.

[0062] The highest and lowest daily groundwater levels are extracted from hourly groundwater level observation data, and the daily diurnal fluctuation range of groundwater levels is calculated.

[0063] Methods for determining whether the daily diurnal fluctuation range of groundwater level conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration include: Obtain the water yield parameters of the aquifer where the groundwater level monitoring point is located, and calculate the daily groundwater evapotranspiration consumption corresponding to the pixel based on the daily diurnal fluctuation range of the groundwater level and the water yield parameters.

[0064] The daily groundwater evapotranspiration loss is compared with the daily potential water loss during the period after the critical day when the water stress state is determined, and the deviation ratio between the two is calculated.

[0065] When the deviation ratio is less than the preset consistency threshold, it is determined that the daily fluctuation range of groundwater level conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration.

[0066] If the criteria are met, the pixel is confirmed as groundwater-dependent vegetation. If the criteria are not met, the determination result obtained by the critical day and the canopy-temperature difference criterion is maintained and marked as pending review.

[0067] For example, when it is determined that the vegetation canopy maintains a non-stressed water condition after the critical day, groundwater level monitoring data of the topographic unit where the target pixel is located is obtained, and the identification result is verified based on the diurnal fluctuation characteristics of the groundwater level monitoring data. First, it is determined whether the topographic unit where the pixel is located has groundwater level monitoring conditions. Specifically, this can be determined by overlaying the pixel coordinates with the spatial distribution layer of groundwater level monitoring wells in the region. In this embodiment, there is a groundwater level monitoring well in the topographic unit where the pixel is located, and the monitoring conditions are met. Therefore, the hourly groundwater level observation data of the monitoring well for several consecutive days is obtained. Based on the hourly groundwater level observation data, the highest and lowest values ​​of the daily groundwater level are extracted, the daily diurnal fluctuation amplitude of the groundwater level is calculated, and it is determined whether the fluctuation amplitude conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration. Specifically, the daily groundwater evapotranspiration water consumption corresponding to the pixel is calculated using the diurnal fluctuation method of groundwater level (White method). This method multiplies the water supply parameter by the depth of the comprehensive change in groundwater level calculated over 24 hours to obtain the water consumption in terms of water depth: In the formula The specific yield parameter of the aquifer where the groundwater level monitoring point is located (in this embodiment, the specific yield of the sandy unconfined aquifer is determined based on previous pumping test data). (Common values ​​for sandy aquifers range from 0.10 to 0.30). The average recovery rate of groundwater level during the period of weak evaporation in the early morning (0:00 to 4:00 in this embodiment) (in this embodiment) This means that the groundwater level rose by an average of 1 mm / h during that period. This value reflects the natural rise in water level caused by lateral recharge of the aquifer under conditions of no evaporation at night. The net drop in groundwater level on that day (in this example) This means that during the day, vegetation transpiration caused a net drop of 4 mm in the groundwater level. and Both figures are in mm. The sum of the two represents the overall change in groundwater level depth over 24 hours. Multiplying this by the specific yield yields the groundwater evapotranspiration loss measured by depth, with consistent dimensions. Substituting the values ​​into the equation... .

[0068] The daily groundwater evapotranspiration loss is compared with the daily potential water loss for the period following the critical day when the state is determined to be free from water stress. The daily potential water loss used as the comparison benchmark refers to the actual water demand level that the pixel should have under conditions unrestricted by soil moisture. Using the daily potential water loss as the comparison benchmark accurately reflects whether groundwater recharge is sufficient to support the pixel in maintaining a normal evapotranspiration level. The deviation ratio between the two is obtained by dividing the absolute value of the difference by the comparison benchmark. ; Potential water consumption in the formula Substituting into The deviation ratio is 5%. In this embodiment, the preset consistency threshold is 20% (this threshold is determined by comprehensively considering the influence of factors such as the estimation error of the nighttime water level rise rate and the heterogeneity of the aquifer in the White method itself, and there is usually an estimation error of about ±10% to 20%). Since 5% is less than 20%, it is determined that the daily diurnal fluctuation range of the groundwater level conforms to the regular diurnal variation characteristics of the groundwater level induced by vegetation transpiration. That is, the diurnal decline and nighttime rise of the groundwater level are highly consistent with the amount of water required for the pixel to maintain non-stress transpiration. Thus, it is confirmed that the continuous transpiration water consumption of the pixel after the soil root layer water is depleted is indeed supported by groundwater replenishment. Finally, the pixel is determined to be groundwater-dependent vegetation. If the daily fluctuation of groundwater level does not conform to the above-mentioned regular diurnal variation characteristics (such as the deviation ratio exceeding the preset consistency threshold), the judgment result obtained by the critical day and the canopy-temperature difference joint criterion will be maintained, and the pixel will be marked as pending review, so as to avoid misjudging non-stressed states caused by non-groundwater sources (such as lateral interflow recharge and other factors) as groundwater dependence.

[0069] For a pixel in a topographic unit that does not have groundwater level monitoring conditions, the topographic similarity parameters between the pixel and neighboring pixels that do have groundwater level monitoring conditions are obtained. The topographic similarity parameters include elevation difference, groundwater depth difference, and river distance difference.

[0070] When the landform similarity parameters meet the preset similarity conditions, the critical day and canopy-temperature difference joint criterion parameters corresponding to the verified groundwater-dependent vegetation determination results of neighboring pixels are migrated to that pixel as auxiliary criteria for groundwater-dependent vegetation identification in that pixel.

[0071] The recognition result of the pixel is corrected based on the joint criterion parameters after migration, and the corrected groundwater-dependent vegetation recognition result is output.

[0072] For example, for pixels in the target area that lack groundwater level monitoring conditions (denoted as pixel B to be determined), this method further includes a parameter migration-assisted determination process: obtaining geomorphic similarity parameters between pixel B and neighboring pixels that have groundwater level monitoring conditions and have been verified (denoted as pixel A). These geomorphic similarity parameters include elevation difference, groundwater depth difference, and river distance difference. In this embodiment, by performing semivariogram analysis on the elevation and groundwater depth data of known monitoring points within the geomorphic unit, the spatial autocorrelation range of the relevant geomorphic parameters in the area is determined to be approximately 500–800 m. Based on this, the preset similarity conditions are: elevation difference not exceeding 1 m, groundwater depth difference not exceeding 0.3 m, and river distance difference not exceeding 200 m. Taking pixel B as an example, it is about 600m away from the verified neighboring pixel A (within the range of spatial autocorrelation). The difference between the two is 0.4m in elevation, 0.2m in groundwater depth, and 120m in river distance, all of which meet the preset similarity conditions. Therefore, the critical day estimation method (i.e., the relevant parameters of the aforementioned soil water balance model) and the joint criterion parameters of the canopy-temperature difference (i.e., the aforementioned corrected stable range boundary) corresponding to the verified groundwater-dependent vegetation identification result of pixel A are transferred to pixel B as auxiliary criteria for the identification of groundwater-dependent vegetation in pixel B. Combined with the thermal infrared remote sensing and soil water balance data that pixel B can obtain, its identification result is corrected, and finally the corrected groundwater-dependent vegetation identification result is output.

[0073] Example 2: Based on the same inventive concept, such as Figure 2 As shown, this embodiment also provides a groundwater-dependent vegetation identification system, the system comprising: The soil water balance and critical day calculation module is used to acquire rainfall observation data, soil effective water holding capacity parameters, and vegetation effective root layer depth parameters of the target area. It combines daily reference evapotranspiration data to construct a soil water balance model and calculates the critical day since the most recent effective rainfall event on a pixel-by-pixel basis.

[0074] The canopy temperature difference and vegetation moisture status determination module is used to obtain the difference between the canopy temperature retrieved by thermal infrared remote sensing and the air temperature at the same time within a time window after the critical day, and to determine whether the vegetation canopy maintains a non-stressful state of moisture based on the difference.

[0075] The groundwater-dependent vegetation identification and analysis module is used to acquire groundwater level monitoring data of the topographic unit where the target pixel is located when it is determined that the vegetation canopy still maintains a water-free state after the critical day. The identification results are verified based on the diurnal fluctuation characteristics of the groundwater level monitoring data to obtain the groundwater-dependent vegetation identification results.

[0076] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0077] Finally, it should be noted that although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for identifying groundwater-dependent vegetation, characterized in that, The method includes: Acquire rainfall observation data, soil effective water holding capacity parameters, and vegetation effective root layer depth parameters for the target area. Combine daily reference evapotranspiration data to construct a soil water balance model and calculate the critical day since the most recent effective rainfall event for each pixel. Within a time window after the critical day, the difference between the canopy temperature retrieved by thermal infrared remote sensing and the air temperature at the same time is obtained, and the difference is used to determine whether the vegetation canopy maintains a water-free state. When it is determined that the vegetation canopy still maintains a non-water-stressed state after the critical day, the groundwater level monitoring data of the topographic unit where the target pixel is located is obtained, and the identification result is verified based on the diurnal fluctuation characteristics of the groundwater level monitoring data to obtain the groundwater-dependent vegetation identification result.

2. The groundwater-dependent vegetation identification method according to claim 1, characterized in that, The method for constructing a soil water balance model and calculating the critical day since the most recent effective rainfall event on a pixel-by-pixel basis includes: Extract the occurrence date and rainfall amount of rainfall events that meet the preset effective rainfall determination conditions from rainfall observation data; obtain the soil effective water holding capacity parameter and vegetation effective root layer depth parameter of the pixel, and calculate the maximum effective water storage capacity of the soil root layer of the pixel; The reference evapotranspiration values ​​for each day from the date of the rainfall event are obtained from the daily reference evapotranspiration data. The actual daily water consumption conversion factor is determined by combining the vegetation cover characteristics of the pixel, and the cumulative daily water consumption is calculated. The daily cumulative water consumption is compared with the maximum effective water storage capacity. When the daily cumulative water consumption reaches the maximum effective water storage capacity at a preset ratio, the corresponding date is determined as the critical day for this rainfall event for that pixel.

3. The groundwater-dependent vegetation identification method according to claim 2, characterized in that, The method for obtaining the effective soil water holding capacity parameter and the effective root layer depth parameter of the pixel, and calculating the maximum effective water storage capacity of the soil root layer of the pixel includes: Based on the soil texture remote sensing inversion results, the field water holding capacity and wilting coefficient of the pixel are obtained, and the effective soil water holding capacity per unit depth is calculated. Based on the vegetation type classification results of the pixel, the empirical root layer depth range of the corresponding vegetation type is obtained. The empirical root layer depth range is then corrected by combining the groundwater depth data of the pixel to obtain the effective root layer depth of the pixel. Multiply the effective water holding capacity of the soil per unit depth by the effective root layer depth to obtain the maximum effective water storage capacity of the soil root layer of that pixel.

4. The groundwater-dependent vegetation identification method according to claim 2, characterized in that, The method for determining the daily actual water consumption conversion factor by combining the vegetation cover characteristics of the pixel and calculating the daily cumulative water consumption includes: Obtain the daily vegetation cover remote sensing inversion results for the pixel, determine the crop coefficient of the pixel based on the vegetation cover, and multiply the reference evapotranspiration value by the crop coefficient to obtain the daily potential water consumption of the pixel. Obtain the estimated soil moisture content of the pixel the previous day, determine the water stress reduction coefficient based on the ratio of the estimated soil moisture content to the effective soil water holding capacity parameter, and multiply the daily potential water consumption by the water stress reduction coefficient to obtain the daily actual water consumption of the pixel. The actual daily water consumption from the date of the rainfall event is added up to obtain the cumulative daily water consumption.

5. The groundwater-dependent vegetation identification method according to claim 1, characterized in that, The method for determining whether the vegetation canopy maintains a water-free state based on the difference between the canopy temperature retrieved by thermal infrared remote sensing and the air temperature during the same period within a time window after the critical day includes: The canopy temperature of the pixel obtained by thermal infrared remote sensing inversion at each observation time after the critical day, and the near-surface air temperature observation value at the corresponding time, are obtained to calculate the canopy-air temperature difference at each time step. Determine the stable range of canopy-temperature difference corresponding to the non-water stress state based on the vegetation type to which the pixel belongs; Determine whether the canopy-temperature difference at each observation time after the critical day continues to fall within a stable range. If it continues to fall within the stable range, then determine that the vegetation canopy of that pixel maintains a non-water-stressed state after the critical day.

6. The groundwater-dependent vegetation identification method according to claim 5, characterized in that, The method for determining the stable range of canopy-temperature difference corresponding to a water-free stress state based on the vegetation type to which the pixel belongs includes: Collect the canopy-temperature difference values ​​of the vegetation type to which the pixel belongs during a historical period and at multiple sample times when it is confirmed to be in a non-water-stressed state, and form a reference sample set; Calculate the mean and standard deviation of the canopy-temperature difference values ​​in the reference sample set. Use the sum of the mean and the standard deviation of a preset multiple as the upper limit of the stable range, and use the difference between the mean and the standard deviation of the preset multiple as the lower limit of the stable range. Obtain near-surface wind speed and air humidity data for the observation day, and correct the upper and lower limits of the stability range based on the near-surface wind speed and air humidity data to obtain the corrected stability range.

7. The groundwater-dependent vegetation identification method according to claim 1, characterized in that, The method for obtaining groundwater level monitoring data of the topographic unit where the target pixel is located, and verifying the identification results based on the diurnal fluctuation characteristics of the groundwater level monitoring data to obtain groundwater-dependent vegetation identification results includes: Determine whether the topographic unit where the target pixel is located has the conditions for groundwater level monitoring. If it does, obtain the hourly groundwater level observation data of the groundwater level monitoring point within the topographic unit for several consecutive days. The highest and lowest values ​​of the daily groundwater level are extracted from the hourly groundwater level observation data, and the daily diurnal fluctuation range of the groundwater level is calculated. Determine whether the daily diurnal fluctuation range of groundwater level conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration. If it does, confirm that the pixel is a groundwater-dependent vegetation. If it does not, maintain the judgment result obtained by the critical day and the canopy-temperature difference joint criterion and mark it as pending review.

8. The groundwater-dependent vegetation identification method according to claim 7, characterized in that, The methods for determining whether the daily diurnal fluctuation range of groundwater level conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration include: Obtain the specific yield parameters of the aquifer where the groundwater level monitoring point is located, and calculate the daily groundwater evapotranspiration consumption corresponding to the pixel based on the daily diurnal fluctuation range of the groundwater level and the specific yield parameters. The daily groundwater evapotranspiration loss is compared with the daily potential water loss during the period after the critical day when the water stress state is determined, and the deviation ratio between the two is calculated. When the deviation ratio is less than the preset consistency threshold, it is determined that the daily fluctuation range of groundwater level conforms to the regular diurnal variation characteristics of groundwater level induced by vegetation transpiration.

9. The groundwater-dependent vegetation identification method according to claim 7, characterized in that, The method further includes: For a pixel in a topographic unit that does not have groundwater level monitoring conditions, obtain the topographic similarity parameters between the pixel and neighboring pixels that do have groundwater level monitoring conditions. The topographic similarity parameters include elevation difference, groundwater depth difference, and river distance difference. When the landform similarity parameters meet the preset similarity conditions, the critical day and canopy-temperature difference joint criterion parameters corresponding to the verified groundwater-dependent vegetation determination results of neighboring pixels are migrated to that pixel as an auxiliary criterion for groundwater-dependent vegetation identification in that pixel. The recognition result of the pixel is corrected based on the joint criterion parameters after migration, and the corrected groundwater-dependent vegetation recognition result is output.

10. A groundwater-dependent vegetation identification system, characterized in that, The system includes: The soil water balance and critical day calculation module is used to acquire rainfall observation data, soil effective water holding capacity parameters and vegetation effective root layer depth parameters of the target area, and construct a soil water balance model by combining daily reference evapotranspiration data, and calculate the critical day since the most recent effective rainfall event on a pixel-by-pixel basis. The canopy temperature difference and vegetation moisture status determination module is used to obtain the difference between the canopy temperature retrieved by thermal infrared remote sensing and the air temperature at the same time within a time window after the critical day, and to determine whether the vegetation canopy maintains a moisture non-stress state based on the difference. The groundwater-dependent vegetation identification and analysis module is used to acquire groundwater level monitoring data of the topographic unit where the target pixel is located when it is determined that the vegetation canopy still maintains a water-free state after the critical day. The identification results are verified based on the diurnal fluctuation characteristics of the groundwater level monitoring data to obtain the groundwater-dependent vegetation identification results.

Images