A quantitative assessment method for the resilience of wetland forest-grassland ecosystems under drought stress

Abstract

The present invention provides a method for quantitatively assessing the resilience of wetland forest and grassland ecosystems under drought stress. The method comprises: data collection, construction of a daily-scale ecosystem service function index (ESS), ecological drought identification, and construction of a recovery resilience curve. The temporal resolution of the index is refined to the daily level, enhancing the capture of short-term droughts on a sub-monthly scale and accurately depicting the dynamic process of ecosystem recovery under drought stress. Drought events are jointly identified using a standardized ecological water deficit, improving the accuracy of identifying drought events that cause significant differences in ecosystem service functions. A quantile regression model is used to quantify the ecosystem resilience of different vegetation types, taking into account the data characteristics of the different recovery time distributions of ecosystems under the same drought intensity. This overcomes the shortcomings of traditional mean regression models, which cannot fully describe the complex characteristics of drought, and provides theoretical support for the sustainable development of ecosystems under extreme climate conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of ecological restoration and ecological assessment, and in particular relates to a quantitative assessment method for the recovery of wetland forest and grass ecosystems under drought stress. Background Art

[0002] Ecological drought refers to an abnormal water cycle event caused by a region's available water resources failing to meet ecological water needs. This abnormality can lead to tree mortality, vegetation degradation, and biodiversity loss, further compromising ecosystem stability and causing disruptions in ecosystem services. Ecological resilience, defined as the ability of an ecosystem to recover from a drought-induced disturbance to a state of equilibrium, is a crucial tool for maintaining sustainable resource and environmental development under extreme climatic conditions. In recent years, under the influence of extreme climatic conditions, the stress of drought on ecosystems has become increasingly severe. Research on ecological resilience is crucial for maintaining sustainable ecosystem development under these conditions.

[0003] Current research on ecosystem resilience largely focuses on constructing indicator evaluation systems, employing methods such as linear weighting and fuzzy hierarchical analysis to integrate subjectively selected factors and indicators to generate resilience evaluation indices. While these studies can fully assess the state of ecosystems after disturbances, their correlation with the disturbance factors is insufficient. This is particularly true for complex climate phenomena such as drought, where varying levels of drought can cause varying degrees of ecosystem disturbance. Considering the dynamics of ecosystem resilience as a function of drought intensity can provide valuable theoretical support for drought preparedness and post-disaster management.

[0004] Existing studies on ecosystem resilience to drought disturbances primarily focus on vegetation phenotypic characteristics, such as annual aboveground net primary productivity (ANPP), enhanced vegetation index (EVI), and normalized difference vegetation index (NDVI). However, this research perspective overlooks the impact of drought stress on the overall ecosystem service function and fails to fully reveal the complex characteristics of drought. Dodd et al. (2023) selected 38 indicators, covering both aboveground and belowground levels, to comprehensively assess ecosystem function and its changes under extreme climate conditions. Similarly, Au et al. (2023) used forest carbon pools and carbon fluxes as indicators of ecosystem service functions and employed a carbon model to analyze the impact of drought on forest ecosystem resilience. Lu and Yan (2023) focused on the carbon cycle, assessing the overall ecosystem service value using indicators such as GPP, NE, and RE. They also explored the relationship between these indicators and vapor pressure deficit, revealing the ecosystem response mechanism to drought. Although these studies provide important support for theoretical development, focusing on a single ecosystem function cannot fully reflect the impact of drought on ecosystem services. Furthermore, not all drought events lead to abnormal ecosystem service values. Traditional hydrometeorological indices (such as the SPI and SPEI) often fail to accurately identify those causing significant abnormalities in ecosystem services, hindering research on ecosystem responses to drought stress. Therefore, developing a method that can accurately identify ecological drought and strengthen the causal relationship between drought and ecosystem abnormalities is of great significance to drought research.

[0005] Under current climate conditions, the frequency of short-duration drought events (typically lasting only two to four weeks) is increasing. These short-duration droughts have a significant impact on plant growth and carbon cycling, making them a key factor in assessing ecosystem recovery. Existing indicators for monitoring ecosystem function or identifying drought events are mostly based on annual or monthly timescales, which cannot capture the rapid onset and dynamic changes of drought. Therefore, improving the temporal resolution of assessments is crucial for a deeper understanding of ecosystem resilience under drought stress.

[0006] Most current methods used to quantify the resilience of ecosystems under drought disturbances use mean regression analysis. However, ecosystem recovery is a complex and dynamic process, affected by multiple factors such as biodiversity, extreme climate, and human activities. Even under the same drought severity, ecosystem recovery times often show significant differences. The mean regression method simplifies different recovery times into average values, ignoring the data distribution characteristics reflected by different recovery times under the same drought severity. This will lead to biased assessments of ecosystem resilience and fail to fully demonstrate its complexity. In contrast, the quantile regression model retains the complete distribution information of the response variable by fitting the recovery times of different percentiles under the same drought severity, revealing the heterogeneity of ecosystems under the same drought pressure, effectively making up for the shortcomings of mean regression, and more comprehensively capturing the characteristics of the data. Summary of the Invention

[0007] In response to the problems existing in the prior art, the purpose of the present invention is to provide a quantitative assessment method for the resilience of wetland forest and grassland ecosystems under drought stress. The method establishes a daily-scale ecosystem service function comprehensive index (ESS), combines the ecosystem's water production value, carbon storage value, and habitat quality value to conduct a holistic evaluation of the ecosystem service function from the perspectives of supply function and regulation function, and proposes an ecological drought joint identification method combined with the standardized ecological water deficit index (SEWDI). The Bayesian non-parametric quantile regression model is used to construct an ecosystem recovery elasticity curve to quantify the ecosystem recovery elasticity of woodlands and grasslands.

[0008] The purpose of the present invention is achieved through the following technical solutions:

[0009] The present invention provides a method for quantitatively assessing the resilience of wetland forest and grassland ecosystems under drought stress, comprising the following steps:

[0010] Step 1, data collection:

[0011] Collect topographic data, meteorological data, soil data and vegetation data within a certain period of time in the study area, and perform data cleaning and preprocessing.

[0012] Step 2: Construction of daily ecosystem service function index ESS:

[0013] S21, based on the data collected in step 1, use the InVEST model to evaluate the ecosystem service function of the study area on an annual scale, specifically including calculating the annual water yield, carbon storage and habitat quality in the study area;

[0014] S22, combined with daily-scale meteorological data, downscales and standardizes the annual-scale ecosystem service function assessment results obtained in S21, and further constructs the daily-scale ecosystem service function index ESS through linear weighting.

[0015] Step 3, ecological drought identification:

[0016] The daily-scale ecosystem service function index ESS constructed in step 2 and the daily-scale standardized ecological water deficit index SEWDI were used to identify abnormal ecological drought in the study area, and abnormal ESS events and SEWDI events were obtained respectively.

[0017] ESS abnormal events and SEWDI abnormal events are jointly identified to determine the ecological drought events that cause significant abnormalities in ecosystem service values.

[0018] Step 4: Construction of recovery elasticity curve:

[0019] The drought intensity and drought recovery time of each abnormal ecological drought event identified in step 3 were calculated, and the Bayesian nonparametric quantile regression model was used to construct the ecosystem recovery elasticity curves of different vegetation types, thereby achieving a quantitative assessment of the ecosystem resilience in the study area.

[0020] The beneficial effects of the present invention compared to the prior art are:

[0021] In the quantitative assessment method for the resilience of wetland forest and grassland ecosystems under drought stress described in the present invention, the ecosystem water production value, carbon storage value and habitat quality value are combined to comprehensively evaluate the ecological resilience from the perspective of ecosystem service function; the temporal resolution of indices such as ESS is accurate to the day, which enhances the capture of short-term droughts below the monthly scale and achieves accurate characterization of the dynamic process of ecosystem recovery under drought stress; drought events are jointly identified in combination with standardized ecological water deficit, which improves the accuracy of identifying drought events that cause significant differences in ecosystem service functions; the quantile regression model is used instead of the mean regression to more fully describe the complex characteristics of drought, realize the quantification of the resilience of different vegetation ecosystems, and provide important reference significance for the sustainable development of ecosystems under extreme climatic conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention will be further described below with reference to the accompanying drawings and examples:

[0023] Figure 1 A flowchart of the quantitative assessment method for the resilience of wetland forest and grassland ecosystems under drought stress is provided;

[0024] Figure 2 The geographical location and topographic distribution of Baiyangdian Lake, an application scenario of the embodiment;

[0025] Figure 3 This is the Baiyangdian ecosystem service distribution map for the application scenario of the embodiment; where (a) is the water yield distribution, (b) is the habitat quality distribution, (c) is the carbon storage distribution, and (d) is the ESS index distribution;

[0026] Figure 4 This is a flow chart of using the ESS index to identify ecosystem service value anomalies in the embodiment;

[0027] Figure 5 A typical ecological drought event identified for the example;

[0028] Figure 6 This is an ecosystem resilience curve constructed for the embodiment; wherein (a) to (f) correspond to low-coverage grassland, medium-coverage grassland, high-coverage grassland, sparse woodland, shrubland and wooded land, respectively. DETAILED DESCRIPTION

[0029] To make the purpose, technical solutions and advantages of the present invention more clear, the following will further describe the disclosed embodiments of the present invention in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0030] Example 1

[0031] like Figure 1 As shown, this embodiment takes the Baiyangdian Basin as an example, provides a quantitative assessment method for the resilience of wetland forest and grassland ecosystems under drought stress, constructs the ecosystem elasticity recovery curve of the Baiyangdian Basin forest and grassland, and conducts example simulation.

[0032] Baiyangdian (38°44′~38°59′N, 115°45′~116°06′E) is the largest freshwater wetland in the North China Plain. It plays a vital role in providing ecosystem services such as water resource regulation, carbon storage, water purification, and biodiversity protection.

[0033] Figure 2 Showing the geographical location and topographical distribution of Baiyangdian. Baiyangdian is located at the border of Baoding and Cangzhou, near the confluence of the Yongding and Hutuo River alluvial fans, and adjacent to the Taihang Mountains. It covers a total area of ​​366 square kilometers. As part of the Daqing River in the Haihe River Basin, Baiyangdian experiences a warm temperate, semi-humid continental monsoon climate. The region experiences dry and windy springs and cold winters, with an average annual temperature of 12.2°C and an average annual precipitation of 529.7 mm. Its vast water area and rich aquatic biodiversity are crucial for protecting the ecological environment, supporting biodiversity, promoting local economic development, and creating ecological value.

[0034] The method comprises the following steps:

[0035] Step 1, data collection:

[0036] The topographic data, meteorological data, soil data and vegetation data of the Baiyangdian Basin from 1982 to 2019 were collected, and the data were cleaned and preprocessed.

[0037] Step 2: Construction of daily ecosystem service function index ESS:

[0038] S21. Based on the data collected in step 1, use the InVEST model to conduct an annual-scale ecosystem service function assessment for the study area. Specifically, use the InVEST model to calculate the annual water yield, carbon storage, and habitat quality of the Baiyangdian Lake Basin from 1982 to 2019; use daily precipitation, the Normalized Difference Vegetation Index (NDVI), and the Leaf Area Index (LAI) as the allocation criteria for annual water yield, carbon storage, and habitat quality, respectively, to calculate the daily water yield, carbon storage, and habitat quality of Baiyangdian Lake from 1982 to 2019.

[0039] S22, combined with daily meteorological data, downscales and standardizes the annual ecosystem service function assessment results obtained in S21, and further constructs the daily ecosystem service function index ESS through linear weighting. Specifically:

[0040] The daily water yield, carbon storage and habitat quality series from 1982 to 2019 were Z-score standardized to eliminate the dimensional differences among the three.

[0041] Water production is classified as ecosystem provision service, carbon storage as regulating service, and habitat quality as supporting service. The weights of water production value, carbon storage value, and habitat quality value are calculated based on the ecosystem service value equivalent factors of the three. The above-mentioned standardized daily-scale dimensionless water production, carbon storage, and habitat quality are linearly weighted to obtain the daily-scale ecosystem service function index ESS.

[0042] Table 1 shows the weights corresponding to water production, carbon storage, and habitat quality after conversion according to the ecosystem service value equivalent factor.

[0043] Table 1

[0044]

[0045] Figure 3 The spatial distribution of the average daily water yield, carbon storage, habitat quality and ESS index in the Baiyangdian Basin from 1982 to 2019 is shown.

[0046] Step 3, ecological drought identification:

[0047] The daily-scale ecosystem service function index ESS constructed in step 2 and the daily-scale standardized ecological water deficit index SEWDI were used to identify abnormal ecological drought in the study area, and abnormal ESS events and SEWDI events were obtained respectively.

[0048] like Figure 4 As shown in the figure, the specific method for using ESS to identify anomalies includes extracting the cumulative values ​​for each year between 1982 and 2019 to form a time series, normalizing them, and calculating the standard deviation (SD). -0.5SD is used as the ESS measurement standard for the selected day. If the ESS on a given day is lower than the corresponding -0.5SD for that day, the ecosystem service function is considered to be abnormal. If the ESS on a given day is higher than -0.5SD and does not fall below -0.5SD again within the next 30 days, the ecosystem service function is considered to have returned to normal.

[0049] SEWDI is used for anomaly identification. Specifically, the three-threshold run theory is adopted with 0.5, -0.5, and -1.5 as thresholds to identify the ecological drought events in Baiyangdian from 1982 to 2019 using the standardized ecological water deficit index (SEWDI).

[0050] ESS abnormal events and SEWDI abnormal events are jointly identified and judged. If at least half of the time in an abnormal period identified by ESS is also identified as abnormal by SEWDI, the abnormal period is retained as an ecological drought event that causes ecosystem abnormality.

[0051] Figure 5 This is a typical drought event identified in this example. The SEWDI began to become abnormal on March 19, 1995, and returned to normal levels on May 18, 1995. After the drought event persisted for a period of time, ecosystem services began to be affected, indicating a time lag between the ESS anomaly and the SEWDI anomaly. The ESS began to become abnormal on April 24, 1995, and returned to normal on May 29, 1995. During this period, the SEWDI was below normal for more than half of the time (from April 24, 1995 to May 18, 1995). Therefore, this ecological drought event was retained, and the entire ESS anomaly period (from April 24, 1995 to May 29, 1995) was considered the recovery period of the drought.

[0052] Step 4: Construction of recovery elasticity curve:

[0053] The drought intensity and drought recovery time of each abnormal ecological drought event identified in step 3 were calculated, and the Bayesian nonparametric quantile regression model was used to construct the ecosystem recovery elasticity curves of different vegetation types, thereby achieving a quantitative assessment of the ecosystem resilience in the study area.

[0054] Specifically, the total time from the onset of ESS anomaly to its return to normal was taken as the drought recovery time, and the absolute value of the SEWDI cumulative value during the drought recovery time was taken as the drought intensity. Bayesian nonlinear quantile regression was used to fit the two for different vegetation cover types.

[0055] The cubic polynomial of the B-spline function is used as the basis function to fit the model. In order to improve the accuracy and efficiency of quantile estimation, the original general test function is replaced by the ELF (extended log-f) loss function.

[0056] The expression is:

[0057]

[0058] Where x is the drought intensity, y is the response variable, i.e., drought recovery time, μ τ (x) is the conditional distribution of the response variable under the τ quantile, τ is 0.1, 0.5, 0.9, b n (x) is the spline basis function for constructing drought recovery time distribution, E is the condition of given covariate x, Regarding the expected value of the response variable y, μ is all possible quantile values, K n are basis function coefficients, is the loss function of quantile estimation, where z is the difference between the predicted value and the observed value, λ is the smoothing parameter, and σ is the scale parameter, all of which affect the complexity of the model and the risk of overfitting.

[0059] Based on the ELF loss function, the Bayesian algorithm is used to update and calibrate the model and solve the spline basis function coefficients. The specific process is as follows:

[0060] (1) The integrated KL divergence (IKL) is used as the calibration loss function, and the 1 / σ0 when the loss function is minimized by the Brent algorithm is used as the baseline learning rate.

[0061]

[0062] Where, It is the posterior variance obtained based on the covariance matrix of the ELF loss function gradient, is the posterior variance based on the model prediction value, It is a positive parameter, here it is 1 / 2; Q is the number of samples, and q is the index value of the traversal sample.

[0063] (2) The optimal scaling loss bandwidth is calculated based on the asymptotic mean square error (MSE) minimization principle of the conditional variance and ELF loss function, and the smoothing parameter λ and the learning rate σ(x) related to the independent variable are determined based on this.

[0064]

[0065] σ(x)=λσ0

[0066] Where, is the optimal scaling loss bandwidth, f z is the probability density function of the error term, f′ z f z The first derivative of F z is the cumulative distribution function of the error term, F z -1 (τ) is the τ quantile of the error term, κ(x) is the conditional variance, and d is the basis function coefficient K n dimension, Q is the number of samples, and 1 / σ0 is the baseline learning rate.

[0067] (3) Use Laplace approximation to calculate the ELF marginal likelihood function and perform numerical optimization to select the smoothing parameter γ.

[0068]

[0069] Where γ and λ are smoothing parameters, σ(x) is the learning rate, and G{γ,σ(x),λ} is the optimal solution of the regression coefficient obtained by internal iteration. is the deviation function of the ELF density function, It is a formula The minimum point of is the saturated log-likelihood function, X T WX is the weighted outer product matrix of the design matrix X, S λ is the precision matrix of the prior distribution, M p For S λ The dimension of the null space.

[0070] (4) The objective function is optimized using the penalized iteratively weighted least squares (PIRLS) algorithm to obtain the maximum a posteriori estimate of the spline basis function coefficients, which can determine the relationship between drought intensity and recovery time.

[0071]

[0072] In the formula, γ, λ are smoothing parameters, σ(x) is the learning rate, K is the basis function coefficient, is the i-th bias component of the likelihood function based on the ELF density, P is the number of precision matrices used to construct the precision matrix, Sp is a positive definite matrix, used to construct the precision matrix, is the deviation of the likelihood function.

[0073] In summary, the resilience characteristics of the Baiyangdian forest and grassland ecosystem are:

[0074] Figure 6 The following are the ecosystem resilience curves for different vegetation types in Baiyangdian Lake. (a) to (f) correspond to low-coverage grassland, medium-coverage grassland, high-coverage grassland, sparse woodland, shrubland, and wooded land, respectively. The horizontal axis represents the independent variable, drought intensity, and the vertical axis represents the response variable, drought recovery time. The drought resilience curves were fitted using quantiles of 0.1, 0.5, and 0.9. This resilience curve reflects the time required for the system to return to normal under a given ecological drought intensity. Different quantiles reflect the ecosystem's resilience to drought under different conditions. The area enclosed by the 0.9 quantile resilience curve and the horizontal axis represents the ecosystem's resilience to drought. The smaller the area, the greater the resilience.

[0075] Resilience curves for all quantiles show that recovery time increases with increasing drought intensity. However, under extreme drought conditions, the 0.9 quantile curves for some vegetation types exhibit anomalies. The 0.9 quantile curves for low-cover grassland, medium-cover grassland, and sparse woodland exhibit a stable or even decreasing trend. This suggests that once drought intensity exceeds a certain threshold, the resilience of these ecosystems remains stable or slightly increases. If drought intensity continues to rise, the ecosystems of these vegetation types will be damaged and unable to recover to their pre-disturbance state. In contrast, the resilience of high-cover grassland, shrubland, and woodland significantly decreases under extreme drought conditions, and the resilience of these ecosystems decreases significantly with increasing drought intensity. In the Baiyangdian Lake Basin, under low and moderate drought intensities, the resilience of forestland is greater than that of grassland (forestland > high-cover grassland > shrubland > medium-cover grassland > sparse woodland >). However, as drought intensity increases, the resilience of forestland and grassland ecosystems exhibits opposite trends, with grassland being more resilient than forestland under high drought intensities.

[0076] Finally, it should be noted that the above is only used to illustrate the technical solution of the present invention and is not limiting. Although the present invention is described in detail with reference to the preferred arrangement scheme, ordinary technicians in this field should understand that the technical solution of the present invention (such as the use of various formulas, the sequence of steps, etc.) can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for quantitatively assessing the resilience of wetland forest and grassland ecosystems under drought stress, characterized in that: The method comprises the following steps: Step 1, data collection: Collect topographic data, meteorological data, soil data and vegetation data within a certain period of time in the study area, and perform data cleaning and preprocessing; Step 2: Construction of daily ecosystem service function index ESS: S21, based on the data collected in step 1, use the InVEST model to evaluate the ecosystem service function of the study area on an annual scale, specifically including calculating the annual water yield, carbon storage and habitat quality in the study area; S22, downscale and standardize the annual ecosystem service function assessment results obtained in S21 by combining daily meteorological data, and further construct the daily ecosystem service function index ESS through linear weighting; Step 3, ecological drought identification: The daily-scale ecosystem service function index ESS constructed in step 2 and the daily-scale standardized ecological water deficit index SEWDI were used to identify abnormal ecological drought in the study area, and abnormal ESS events and SEWDI events were obtained respectively. Jointly identify ESS anomaly events and SEWDI anomaly events to determine ecological drought events that cause significant anomalies in ecosystem service value; Step 4: Construction of recovery elasticity curve: The drought intensity and drought recovery time of each abnormal ecological drought event identified in step 3 were calculated, and the Bayesian nonparametric quantile regression model was used to construct the ecosystem recovery elasticity curves of different vegetation types, thereby achieving a quantitative assessment of the ecosystem resilience in the study area.

2. The method according to claim 1, characterized in that In step 2 ( S22 ), daily rainfall, normalized difference vegetation index, and leaf area index are used as downscaling basis for annual water production, carbon storage, and habitat quality, respectively.

3. The method according to claim 1, characterized in that In step 2, S22, Z-score standardization is used to eliminate dimensional differences.

4. The method according to claim 1, wherein The specific methods for using ESS to identify anomalies in step 3 include: A date was selected and the cumulative ESS value for the 30 days preceding that date was calculated using a 30-day time scale. The 30-day cumulative values ​​of all years corresponding to that date were used as the time series, which were normalized and the standard deviation (SD) was calculated. -0.5 times the SD was used as the criterion for judging whether the daily ESS was abnormal. If the daily ESS was lower than the corresponding -0.5SD of the day, the ecosystem service function was considered abnormal. If the ESS was higher than -0.5SD and did not fall below -0.5SD in the next 30 days, the ecosystem service function was considered to have returned to normal.

5. The method according to claim 1, wherein In step 3, SEWDI is used to identify anomalies, specifically: The SEWDI index is used to identify abnormal phenomena using the three-threshold run theory with -0.5, -1.5, and 0.5 as standards.

6. The method according to any one of claims 1, 4 or 5, characterized in that In step 3, ESS abnormal events and SEWDI abnormal events are jointly identified, specifically: ESS abnormal events and SEWDI abnormal events are jointly judged. If at least half of the time in an abnormal period identified by ESS is also identified as abnormal by SEWDI, the abnormal period is retained as an ecological drought event that causes abnormal ecosystems.

7. The method according to claim 1, characterized in that In step 4, the drought intensity is the cumulative value of the absolute value of SEWDI during the drought recovery time, and the drought recovery time is the total time from the beginning of the abnormality of ESS to the recovery of ESS in step 3.

8. The method according to claim 1, characterized in that In step 4, the generalized additive model with ELF as the loss function is used to fit the drought recovery time and drought intensity using the Bayesian algorithm to construct the ecosystem recovery resilience curve, which is expressed as: Where, is the drought intensity, Drought recovery time, for The conditional distribution of the response variable under quantiles, Take 0.1, 0.5, 0.9, To construct the spline basis function of drought recovery time distribution, is the basis function coefficient, E is the condition under which the covariate x is given, is the expected value of the response variable y, For all possible quantile values, is the loss function of quantile estimation, where is the difference between the predicted value and the observed value, is the smoothing parameter, is the scale parameter.

Images