A parameterization method for anthropogenic dust emission for different underlying surfaces

Abstract

The present invention discloses a parameterization method for anthropogenic dust emission for different underlying surfaces, and the method is as follows: Determine the relationship among the anthropogenic dust emission flux, population density, and critical friction velocity for dust emission of different underlying surfaces in the sample area to obtain a fitting equation; Determine the critical friction velocity for dust emission of different underlying surfaces in the area to be measured, and optimize the fitting equation with the critical friction velocity for dust emission of different underlying surfaces in the area to be measured to obtain an optimized equation; Construct the dynamic surface types in the area to be measured; Substitute the dynamic surface types in the area to be measured and the optimized equation into different dust emission parameterization schemes for dust simulation to obtain simulation results. The beneficial effects of the present invention are: Solve the problems of non-uniformity of the critical friction velocity for dust emission in the previous anthropogenic dust emission parameterization schemes, neglect of the influence of dynamic surface types on anthropogenic dust emission, neglect of the influence of direct human activities on dust emission, and lack of long-term and accurate observational constraints on anthropogenic dust.

Description

Technical Field

[0001] The present invention relates to the field of numerical simulation of sand and dust, and particularly to a parameterization method for anthropogenic sand-dust emission for different underlying surfaces. Background Art

[0002] Sand-dust emission refers to the spatial displacement process of soil particles under the action of wind, which is a key link in the sand-dust cycle and determines the intensity, spatio-temporal distribution characteristics, transportation and sedimentation processes of sand and dust. The quantitative and accurate description of the emission process is the basis for realizing sand-dust simulation and forecasting (Zhu Hao and Zhang Hongsheng, 2011).

[0003] Since the 1980s, the rapid construction and development of sand-dust emission parameterization schemes at home and abroad (Tegen and Fung, 1994 (TF scheme); Marticorena and Bergametti, 1995 (MB scheme); Shao, 2001, 2004 (Shao scheme); Wang et al., 2000; Ginoux et al., 2001 (GOCART scheme); Zhao et al., 2003; Huang Meiyuan, 1998; Sun Jianhua, 2003; Lei Hang et al., 2005; Zhao Linna, 2007; Uno et al., 2006; Zhao et al., 2010) have greatly deepened people's understanding of the emission mechanism and the impact of sand and dust on the environment and climate.

[0004] In current weather and climate models, most of the sand-dust emission parameterization schemes only consider the sand-dust storm process of strong wind-induced sand-dust emission under natural factors, ignoring the anthropogenic sand-dust emission process caused by direct human activities or land use / cover changes through human activities, resulting in great uncertainty in the simulated sand-dust emission flux of numerical models.

[0005] In addition, for different sand-dust emission schemes, after considering human activities, the quantitative results of anthropogenic sand-dust emission also vary greatly. The uncertainty of current anthropogenic sand-dust emission parameterization schemes mainly comes from four aspects: (1) The non-uniformity problem of the critical sand-dust emission friction velocity in existing anthropogenic sand-dust emission schemes has not been well solved. (2) The impact of dynamic surface types on anthropogenic sand-dust emission is ignored. (3) The impact of direct human activities on sand-dust emission is not considered in the sand-dust emission parameterization scheme. (4) There is a lack of long-term and accurate observational constraints on anthropogenic sand and dust. Summary of the Invention

[0006] In order to address the technical problems of the existing anthropogenic sand-dust emission parameterization schemes, such as insufficient differential treatment of different sand-dust emission underlying surfaces, ignoring the dynamic impact of anthropogenic sand sources and human activities, and immature observational identification techniques, the present invention provides a parameterization method for anthropogenic sand-dust emission for different underlying surfaces, and the method includes the following steps:

[0007] S1: Determine the relationship between the anthropogenic dust emission flux, population density, and critical friction velocity for dust emission over different underlying surfaces in the sample area, and obtain the fitting equation;

[0008] S2: Determine the critical friction velocity for dust emission over different underlying surfaces in the area to be measured, and optimize the fitting equation using the critical friction velocity for dust emission over different underlying surfaces in the area to be measured to obtain the optimized equation;

[0009] S3: Construct the dynamic land surface types in the area to be measured;

[0010] S4: Substitute the dynamic land surface types in the area to be measured and the optimized equation into different dust emission parameterization schemes for dust simulation to obtain the simulation results.

[0011] The beneficial effects provided by the present invention are as follows: Based on the WRF-Chem (Weather Research and Forecasting model with chemistry) model and combined with a variety of advanced research methods, considering the non-uniformity of the critical friction velocity for dust emission over different land surface types, quantitatively estimating the impact of human activities on land cover and land use, and considering the impact of direct human activities on dust emission, thereby developing and constructing an anthropogenic dust emission parameterization scheme for the East Asian region; solving the problems of non-uniformity of the critical friction velocity for dust emission in the previous anthropogenic dust emission parameterization schemes, neglecting the impact of dynamic land surface types on anthropogenic dust emission, neglecting the impact of direct human activities on dust emission, and lacking long-term and accurate observational constraints on anthropogenic dust. Brief Description of the Drawings

[0012] Figure 1 is the flowchart of the method of the present invention; Detailed Embodiments

[0013] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described below in conjunction with the drawings.

[0014] Please refer to Figure 1 , Figure 1 is the flowchart of the method of the present invention. The present invention provides an anthropogenic dust emission parameterization method for different underlying surfaces, including the following steps:

[0015] S1: Determine the relationship between the anthropogenic dust emission flux, population density, and critical friction velocity for dust emission over different underlying surfaces in the sample area, and obtain the fitting equation;

[0016] Specifically, the present application first uses a surface dust release observation instrument to determine the critical friction velocity threshold by statistically analyzing the variation laws of the critical friction velocity for dust emission corresponding to different land surface types (such as farmland, pasture, and sparse grassland) through test results;

[0017] Furthermore, wind tunnel tests were carried out in the sample area; four typical land surfaces, namely farmland, pasture, sparse grassland, and city, were simulated in the multi-functional environmental wind tunnel in the sample area to study the critical sand initiation friction velocity of different land surfaces and its variation law under different anthropogenic disturbances, draw the corresponding variation map of the critical sand initiation friction velocity, and compare it with the field observation results to evaluate the accuracy of the determination of the critical sand initiation threshold.

[0018] When the critical sand initiation threshold is accurate, in order to quantitatively estimate the sand dust release caused by direct anthropogenic activities, the relationship between anthropogenic activities of different underlying surfaces with different intensities, friction velocity, and sand dust release was further simulated in the wind tunnel.

[0019] Using population density to characterize the intensity of anthropogenic activities, the empirical formula for sand dust release caused by direct anthropogenic activities is:

[0020] E A =E(1 + f a (N, u*)) (1)

[0021] f a (N, u*) = CN α u* β (2)

[0022] Among them, E A represents the sand initiation flux after different intensities of anthropogenic activities, E represents the sand initiation flux of the underlying surface without anthropogenic activities, f a (N, u*) is the empirical term for sand initiation caused by anthropogenic activities and is a function of population density N and critical sand initiation friction velocity u*; C is the empirical parameter for sand initiation by anthropogenic activities; α and β are fitting parameters.

[0023] The fitting process is as follows: By conducting multiple sets of sand initiation experiments caused by different intensities of anthropogenic activities and sand initiation experiments on the same underlying surface without anthropogenic activities, the critical sand initiation friction velocity and sand dust initiation under different intensities of anthropogenic activities are obtained. Through the least squares method to fit the regression equation, the values of C, α, and β are finally determined, and the empirical relationship between anthropogenic sand initiation amount, population density, and friction velocity under different underlying surfaces is obtained.

[0024] S2: Determine the critical sand initiation friction velocity of different underlying surfaces in the area to be measured, and optimize the fitting equation with the critical sand initiation friction velocity of different underlying surfaces in the area to be measured to obtain the optimized equation;

[0025] Specifically, conducting experiments in the sample area can only give the local critical sand initiation threshold, while the numerical model requires the spatio-temporal distribution characteristics of the critical sand initiation threshold of the entire simulated area.

[0026] In the present invention, a semi-empirical parameterization scheme is used to obtain the seasonal variation characteristics of the critical sand initiation threshold in the entire East Asian region (the aforementioned area to be measured).

[0027] First, based on the AVHRR satellite, the Normalized Difference Vegetation Index (NDVI) of the East Asian region was obtained. According to the methods of Gutman and Ignatov (1998) and Shao et al. (1996), surface roughness parameters such as vegetation coverage Av and vegetation roughness density λv were first calculated, and then the effective geometric height h and total roughness density λ were obtained.

[0028] In order to introduce the dynamic changes of vegetation into the sand - emission scheme based on the physical mechanism of sand - emission, the empirical relationship (Xi et al., 2015) of surface roughness z0, average geometric height h, and total roughness density λ was used to obtain z0.

[0029] In anthropogenic underlying surfaces such as farmland and pasture where bare soil and sparse vegetation are mixed (0.1 < NDVI ≤ 0.3), the seasonal variation characteristics of z0 were further calculated by integrating the bidirectional reflectance distribution product of the Polarization and Directionality of Earth's Reflectances (POLDER - 1) (Marticorena et al., 2004; Laurent et al., 2005).

[0030] According to parameters such as soil moisture in the land surface module of the numerical model, the critical sand - emission friction velocity of different anthropogenic underlying surfaces in the entire East Asian region was obtained.

[0031] Finally, the parameterization scheme of the critical sand - emission friction velocity was improved and perfected by combining the observation results.

[0032] The calculation process of the relevant parameters in step S2 is as follows:

[0033] Based on the AVHRR satellite, the Normalized Difference Vegetation Index (NDVI) of the East Asian region was obtained to calculate the surface roughness parameters; the surface roughness parameters include vegetation coverage A v and vegetation roughness density λv;

[0034]

[0035] λ V =-0.35ln(1 - A V )

[0036] NDVI is the observed real - time NDVI value. NDVI V and NDVI S represent the NDVI values of dense vegetation and bare soil respectively. Take NDVI V as 0.93; NDVI S as 0.06.

[0037] Obtain the effective geometric height h and total roughness density λ of the ground surface based on the surface roughness parameter;

[0038] Refer to Table 1 below.

[0039] Table 1 Relationship Table of Land, Crop Type and Height

[0040]

[0041] Where: λ = λ B + λ V , h = (h B × λ B + h V × λ V ) / λ;

[0042] λ B Non-vegetation roughness; h V Weighted height of non-vegetation; h B Weighted height of vegetation;

[0043] Obtain the surface roughness z0 based on the empirical relationship of the surface roughness z0, average geometric height h and total roughness density λ;

[0044]

[0045] Fuse the bidirectional reflectance distribution product in the artificial underlying surface and calculate the seasonal variation characteristics of the surface roughness z0;

[0046]

[0047]

[0048] Where the two parameters K1 and K0 are obtained by fusing the POLDER-1 bidirectional reflectance distribution product; a = 4.859×10 -3

[0049] cm, b is a dimensionless parameter and is 0.052.

[0050] Obtain the critical sand-moving friction velocity of different underlying surfaces in the area to be measured according to the seasonal variation characteristics of the surface roughness z0 and the soil moisture in the land surface module of the numerical model.

[0051]

[0052]

[0053]

[0054] Where, d is the diameter of the particle; f w and f rThey are the correction terms of the soil moisture of the underlying surface and the roughness of the non-eroded surface respectively. The ratio of δ particles to air density; ρ represents air density; A N represents the cohesive force between particles. According to experience, it is generally 0.0123; g is the acceleration due to gravity. w r is a function of the clay content of the soil on the underlying surface of farmland under different tillage patterns, and w is the soil moisture content. w r is a function of the soil clay content, w r = 0.0014×(clay %)² + 0.17×(clay %), and w is the soil moisture in the top 0 - 2 cm soil layer.

[0055] S3: Construct the dynamic surface type of the area to be measured;

[0056] Step S3 is specifically as follows:

[0057] Take the spatial distribution of potential vegetation in a certain historical year of the area to be measured as the starting point of the dynamic surface type, and continuously iterate and transform the weight factor to obtain a dynamic surface type data set;

[0058] Use satellite surface type observation constraints to correct the dynamic surface type data set to obtain the constructed dynamic surface type of the area to be measured;

[0059] As an embodiment, the present invention takes the spatial distribution of potential vegetation in the East Asian region in 1999 as the starting point of the dynamic surface database, and obtains the dynamic surface type data sets for each year from 2000 to 2010 by continuously iterating the "transformation weight factor". And under the observation constraints of the MOIDS satellite surface type data, by adjusting the calculation rule of the "priority transformation factor", the difference between the calculation result and the satellite data is reduced, and the calculation accuracy of the dynamic surface type is improved;

[0060] It should be noted that the relevant concept explanations of the transformation weight factor are as follows:

[0061] In order to quantitatively estimate the impact of human activities on land cover and land use, the "transformation weight factor" (Meiyappan and Jain, 2012) (the numerical range is 0 - 1.0) is defined to characterize the transformation weight between 4 main artificial land surfaces and 23 other land surface types (excluding water bodies), reflecting the change and percentage of the natural land surface type on the grid area caused by human activities.

[0062] If a certain anthropogenic land surface type A within the grid area continues to expand compared to the previous year (such as farmland reclamation), the remaining natural land surface types within the grid are converted to A according to the "conversion weight factor"; if a certain anthropogenic land surface type A decreases in area compared to the previous year (such as abandoned farmland), the potential natural vegetation at this grid point (representing the native vegetation unaffected by human interference) (Ramankutty and Foley, 1999) is used to replace this type.

[0063] S4: Bring the dynamic land surface types of the area to be measured and the optimized equation into different sand emission parameterization schemes for dust simulation to obtain simulation results.

[0064] Specifically, step S4 is as follows:

[0065] Adopt three common sand emission parameterization schemes, namely MB, Shao, and GOCART, in the WRF-Chem model. Introduce the improved critical sand emission friction velocity, dynamic land surface types, and the newly constructed direct anthropogenic activity empirical term into the sand emission scheme to explore the uncertainty of different sand emission parameterization schemes in simulating anthropogenic dust in East Asia.

[0066] Set six groups of experiments, namely: MB, Shao, Shao_anthro, GOCART, GOCART_dyn, GOCART_dyn_anthro.

[0067] Among them, MB, Shao, and GOCART respectively represent different sand emission parameterization schemes, dyn represents using the newly constructed dynamic land surface types in the sand emission scheme, and anthro represents adding the empirical term of sand emission caused by different direct anthropogenic activities on the underlying surface based on the original sand emission scheme.

[0068] Based on the six groups of numerical simulation experiments,

[0069] ① Based on satellite inversion and ground observation results, evaluate the simulation performance of the anthropogenic sand emission parameterization scheme and analyze the sources of uncertainty;

[0070] ② Compare the seasonal variation differences in the spatial distribution of the surface roughness correction function and the critical sand emission friction velocity among the MB, Shao, and Shao_anthro schemes in the East Asian region;

[0071] ③ Compare the GOCART and GOCART_dyn schemes and discuss the impact of the dynamic land surface on the dynamic changes of anthropogenic sand sources and the seasonal and interannual changes of sand emission flux;

[0072] ④ Compare the uncertainty of the GOCART_dyn and GOCART_dyn_anthro schemes in simulating sand emission caused by direct anthropogenic activities.

[0073] Generally, regarding the logical relationship among steps S1 to S4, the following is supplemented:

[0074] Step S1: Based on wind tunnel experiments, for different underlying surfaces, considering the artificial sand emission flux, population density, and critical sand-emitting friction velocity, a fitting equation is constructed.

[0075] Step S2 is to optimize the critical sand-emitting friction velocity in S1. There are significant differences in the critical friction velocities of different underlying surfaces. Modify the parameterized simulation of this part in the model. Furthermore, optimize the spatio-temporal distribution characteristics of the critical friction velocity threshold of the entire region in the model.

[0076] Step S3 is to correct the underlying surface type in S1. Using satellite analysis, a more accurate underlying surface dataset is constructed.

[0077] Step S4 is to use the optimized scheme obtained from the above process for simulation verification.

[0078] Generally speaking, the beneficial effects of the present invention are as follows: Based on the WRF-Chem (Weather Research and Forecasting model with chemistry) model and combined with various advanced research methods, considering the non-uniformity of the critical sand-emitting friction velocity of different surface types, quantitatively estimating the impact of human activities on land cover and land use, and considering the impact of direct human activities on sand emission, so as to develop and construct a parameterization scheme for artificial sand emission in the East Asian region. It mainly solves the problems of non-uniformity of the critical sand-emitting friction velocity in the previous artificial dust emission parameterization scheme, neglecting the impact of dynamic surface types on artificial sand emission, neglecting the impact of direct human activities on sand emission, and lacking long-term and accurate observational constraints on artificial dust.

[0079] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A parameterization method for anthropogenic dust emission over different underlying surfaces, characterized in that: It includes the following steps: S1: Determine the relationship between the anthropogenic sand emission flux, population density, and critical sand - initiating friction velocity of different underlying surfaces in the sample area, and obtain the fitting equation; The fitting equation described in step S1 is: E A = E(1 + f a (N, u*)) (1) f a (N,u*) = CN α u* β (2) Among them, E A represents the sand emission flux after different human activity intensities, E represents the sand emission flux of the underlying surface without human activities, and f a (N, u*) is an empirical term for sand emission caused by human activities and is a function of population density N and critical sand emission friction velocity u*; C is an empirical parameter for sand emission caused by human activities; α and β are fitting parameters; S2: Determine the critical sand - initiating friction velocity of different underlying surfaces in the area to be measured. Optimize the fitting equation with the critical sand - initiating friction velocity of different underlying surfaces in the area to be measured to obtain the optimized equation; S3: Construct the dynamic land surface type of the area to be measured; S4: Substitute the dynamic land surface type of the area to be measured and the optimized equation into different sand - emission parameterization schemes for dust simulation to obtain the simulation results.

2. The parameterization method for anthropogenic dust emission over different underlying surfaces according to claim 1, characterized in that: The specific process of determining the critical sand - initiating friction velocity of different underlying surfaces in step S2 is: Based on the normalized difference vegetation index (NDVI) obtained from AVHRR satellites in the East Asian region, the surface roughness parameters are calculated; the surface roughness parameters include vegetation coverage A v and the rough density λv of vegetation; based on the surface roughness parameters, the effective geometric height h of the surface and the total roughness density λ are obtained; According to the empirical relationship between the surface roughness z0, the effective geometric height h of the surface, and the total roughness density λ, and the effective geometric height h of the surface and the total roughness density λ, inversely deduce the surface roughness z0; According to the seasonal variation characteristics of the surface roughness z0 and the soil moisture in the land surface module of the numerical model, obtain the critical sand - initiating friction velocity of different underlying surfaces in the area to be measured.

3. The parameterization method for anthropogenic dust emission over different underlying surfaces according to claim 1, characterized in that: Step S3 is specifically: Take the spatial distribution of potential vegetation in a certain historical year in the area to be measured as the starting point of the dynamic land surface type, and continuously iterate the conversion weight factor to obtain the dynamic land surface type data set; Use satellite land surface type observations to constrain and correct the dynamic land surface type data set to obtain the constructed dynamic land surface type of the area to be measured.

4. The parameterization method for anthropogenic dust emission over different underlying surfaces according to claim 1, characterized in that: The different sand - emission parameterization schemes include: the MB scheme, the Shao scheme, and the GOCART scheme.

5. The parameterization method for anthropogenic dust emission over different underlying surfaces according to claim 4, characterized in that: The dust simulation described in step S4 specifically includes six groups of simulations, namely: MB, Shao, Shao_anthro, GOCART, GOCART_dyn, GOCART_dyn_anthro; Among them, MB, Shao, and GOCART respectively represent different sand - emission parameterization schemes, dyn represents using the newly constructed dynamic land surface type in the sand - emission scheme, and anthro represents adding an empirical term of sand emission caused by direct human activities on different underlying surfaces on the basis of the original sand - emission scheme.

Images