Land surface evapotranspiration estimation method considering soil moisture, vegetation type and growth status

Abstract

This invention proposes a method for estimating land surface evapotranspiration that considers soil moisture, vegetation type, and vegetation growth. It comprises six steps: Step 1, expressions for the influence of soil moisture, vegetation growth, and vegetation type on land surface evapotranspiration; Step 2, preparation of basic data; Step 3, calculation of reference evapotranspiration; Step 4, maximum evapotranspiration coefficient (K) for different vegetation types. CM The process involves: Step 5: Estimating the distribution of land surface evapotranspiration; Step 6: Calculating the error in the estimation of land surface evapotranspiration. This invention can quantify the impact of soil moisture, vegetation growth, and vegetation type on land surface evapotranspiration, and can estimate the evapotranspiration water consumption under different vegetation types and growth conditions. It is suitable for water resource surveys and monitoring.

Description

Technical Field

[0001] This invention relates to a method for estimating land surface evapotranspiration that takes into account soil moisture, vegetation type and growth status. It can solve the problem that land surface evapotranspiration is difficult to quantify due to the influence of three factors: soil moisture, vegetation growth status and vegetation type. It also significantly improves the accuracy and reliability of land surface evapotranspiration estimation, thereby better serving water resource management. This invention belongs to the field of hydrological remote sensing. Background Technology

[0002] Evapotranspiration is the process by which water on Earth's land (non-water surface) changes from a liquid to a gaseous state and enters the atmosphere. It includes evaporation, where water adhering to the land surface and the surface of vegetation leaves, as well as water contained in the soil, directly vaporizes and enters the atmosphere; and transpiration, where water in the soil enters the plant through the root system and vaporizes through the stomata of the leaves. Evapotranspiration is a crucial part of the terrestrial water cycle; globally, over 60% of precipitation returns to the atmosphere via evapotranspiration, and this proportion reaches over 90% in (semi-)arid regions. Accurately estimating land evapotranspiration is an important topic in hydrological research and a key aspect of water resource management.

[0003] However, due to the complex influence of land surface evapotranspiration on land surface conditions and atmospheric conditions, accurate estimation of land surface evapotranspiration is extremely difficult. The main land surface conditions affecting land surface evapotranspiration include soil moisture, vegetation growth, and vegetation type. Currently, the lack of a concise expression that quantifies how soil moisture, vegetation growth, and vegetation type affect land surface evapotranspiration, and the corresponding estimation methods, is a major reason hindering the improvement of the accuracy and reliability of land surface evapotranspiration estimation.

[0004] This invention proposes a ternary linear land surface evapotranspiration model that considers soil moisture, vegetation growth, and vegetation type, as well as an evapotranspiration estimation method based on watershed water balance, which can significantly improve the accuracy and reliability of watershed-scale land surface evapotranspiration estimation. Summary of the Invention

[0005] 1. Objective: The objective of this invention is to propose a method for estimating land surface evapotranspiration that considers soil moisture, vegetation type, and growth status, quantitatively expressing the impact of soil moisture, vegetation growth status, and vegetation type on land surface evapotranspiration, and an evapotranspiration estimation method based on watershed water balance. This method can estimate the evapotranspiration water consumption under different vegetation types and growth status conditions, significantly improving the accuracy and reliability of land surface evapotranspiration estimation at the watershed scale, and supporting watershed water and soil resource allocation.

[0006] 2. Technical problems to be solved

[0007] Soil moisture, vegetation growth, and vegetation type are the three major land surface state parameters that affect land surface evapotranspiration. The main technical problems to be solved are: (1) quantifying the impact of soil moisture, vegetation growth, and vegetation type on land surface evapotranspiration; (2) solving the evapotranspiration coefficient of different vegetation types based on watershed water balance and estimating land surface evapotranspiration.

[0008] 3. Technical Solution

[0009] This invention addresses the aforementioned technical problems and proposes corresponding solutions. The overall solution is attached. Figure 1 As shown. This invention is a ternary linear land surface evapotranspiration model considering soil moisture, vegetation growth, and vegetation type, and a method for estimating evapotranspiration based on watershed water balance. The specific steps of this method are as follows:

[0010] Step 1: Construct expressions for the effects of soil moisture, vegetation growth, and vegetation type on land surface evapotranspiration;

[0011] The quantitative formula for the impact of soil moisture, vegetation growth, and vegetation type on land surface evapotranspiration is as follows:

[0012] aET=[(K CM -1)×NDVI+1]×θ R ×ET0 (1)

[0013] In the above formula, aET is the actual evapotranspiration of the land surface (mm); K CM θ represents the evapotranspiration coefficient of different vegetation types at their optimal growth stage, i.e., the maximum evapotranspiration coefficient (dimensionless); NDVI is the vegetation index (dimensionless), used to represent vegetation vigor (the higher the value, the better the growth); R ET0 represents the relative soil moisture content (dimensionless); ET0 represents the reference evapotranspiration, which, according to the Food and Agriculture Organization of the United Nations (FAO), is the evapotranspiration (mm) of a grassland completely covered by grass with a height of 12 cm.

[0014] The NDVI vegetation index expression formula in formula (1) is as follows:

[0015]

[0016] In the above formula, R Nir R represents the surface reflectance (dimensionless) in the near-infrared band (generally around 860nm). Green The surface reflectance (dimensionless) is measured in the green band (generally around 550nm).

[0017] The formula for expressing the relative soil moisture content in formula (1) is as follows:

[0018]

[0019] In the above formula, θ R denoted as relative soil moisture content (dimensionless); W represents soil moisture content (%). m W represents the soil moisture content at wilting time (%). f Soil saturated moisture content (%).

[0020] The expression for the reference evaporation ET0 in formula (1) is shown in step three.

[0021] Step Two: Preparation of Basic Data;

[0022] The basic data includes seven categories: watershed boundaries, meteorology, land cover type, watershed outlet runoff, soil moisture content, vegetation index, and changes in watershed water storage. Meteorological data includes precipitation, net surface radiation flux, relative humidity, air temperature, air pressure, and wind speed.

[0023] Step 3: Refer to the calculation of evapotranspiration;

[0024] The calculation formula is:

[0025]

[0026] Where ET0 is the reference evaporation (mm / day), Δ is the slope of the vapor pressure curve (kPa / ℃), and R... n Net surface radiation flux (MJ / m²) 2 / day), G is soil heat flux (MJ / m²) 2 / day), G is generally taken as R n 5%, γ is the wet / dry surface constant (kPa / ℃), T is the air temperature (℃) at a height of 2 meters above the ground, u2 is the wind speed (m / s) at a height of 2 meters above the ground, e s e is the saturated vapor pressure (kPa). a The vapor pressure is (kPa).

[0027] The formula for calculating the slope of the vapor pressure curve is:

[0028]

[0029] The formula for calculating the wet / dry constant is:

[0030] γ = 0.665 × 10 -3 P (6)

[0031] Where P is the air pressure (kPa).

[0032] The formula for calculating saturated vapor pressure is:

[0033]

[0034] The formula for calculating vapor pressure is:

[0035]

[0036] RH represents relative humidity.

[0037] Step 4: Solve for the maximum evapotranspiration coefficient (K) for different vegetation types CM );

[0038] The total evapotranspiration of the watershed is calculated using the following water balance formula:

[0039] aET WB =Pre-ΔW-Q (9)

[0040] Among them, aET WB For the water balance formula, calculate the total evapotranspiration (mm), Pre is the watershed precipitation (mm), Q is the watershed outlet runoff (mm), and Δ W Changes in water storage in the basin (mm).

[0041] Based on land cover type, K is used to classify different crop types in farmland and different vegetation types in non-farmland. CM The range of variation is set to 0-1, for the water surface K CM Set to 1. For different vegetation types such as grassland, deciduous forest, coniferous forest, and broadleaf forest, and different crops such as corn, wheat, and vegetables, K is selected from the land cover type data. CM By varying the step size from 0 to 1 in a step size of 0.01, an exhaustive method was used to set different vegetation and crop K values. CM The coefficient data set combines meteorological data such as precipitation, net surface radiation flux, relative humidity, temperature, air pressure, and wind speed, as well as K... CM Substitute the coefficient data into formulas (1)-(8) to calculate K. CM The evapotranspiration of the basin is collected from different coefficient data combinations in the coefficient data set. The "K" value that is closest to the total evapotranspiration of the basin calculated by formula (9) is then selected. CM "Coefficient data combination".

[0042] Step 5: Estimate the distribution of land surface evapotranspiration;

[0043] Meteorological data such as precipitation, net surface radiation flux, relative humidity, temperature, air pressure, and wind speed are substituted into formulas (4)-(8) to estimate the reference evapotranspiration of the watershed. Soil moisture content is substituted into formula (3) to calculate the relative soil moisture content.

[0044] The estimated watershed reference evapotranspiration, calculated soil moisture content, collected vegetation index data, and the K values ​​obtained from step four for different vegetation types such as grassland, deciduous forest, coniferous forest, and broadleaf forest, as well as different crops such as corn, wheat, and vegetables, are used. CM Substituting into formula (1), the distribution of land surface evapotranspiration in the watershed is estimated.

[0045] Step Six: Calculate the estimation error of land surface evapotranspiration;

[0046] The land surface evapotranspiration distribution data estimated in step five are statistically analyzed according to the watershed boundary to obtain the total land surface evapotranspiration within the watershed.

[0047] The following formula is used to calculate the estimation error of land surface evapotranspiration:

[0048] Rerror = (aET - aET) WB ) / aET WB (10)

[0049] Where Rerror is the error (dimensionless), aET WB The total evapotranspiration (mm) of the basin is calculated using formula (9) in step four. aET is the total land surface evapotranspiration (mm) obtained by statistically analyzing the land surface evapotranspiration distribution data estimated in step five according to the basin boundary.

[0050] 4. Advantages and Efficacy

[0051] This invention relates to a method for estimating land surface evapotranspiration that takes into account soil moisture, vegetation type and growth. Its advantages are: (1) It quantitatively expresses the influence of three main factors, namely soil moisture, vegetation growth and vegetation type, on land surface evapotranspiration, and considers more comprehensive factors; (2) It solves the evapotranspiration coefficient of different vegetation types based on watershed water balance, and then estimates land surface evapotranspiration, which can improve the estimation accuracy at the watershed scale; (3) It can estimate land surface evapotranspiration under different vegetation growth and vegetation types. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating the implementation of a land surface evapotranspiration estimation method that considers soil moisture, vegetation type, and growth status, according to the present invention.

[0053] Figure 2 This is a map showing the evapotranspiration distribution in the Tuligen River Basin in 2006.

[0054] Figure 3 A graph showing the changes in evapotranspiration over many years for forest land, grassland, and cultivated land in the Tuligen River Basin (2006-2017). Detailed Implementation

[0055] To better illustrate a method for estimating land surface evapotranspiration that considers soil moisture, vegetation type, and growth status, a land surface evapotranspiration estimation example was implemented in the Tuligen River Basin in western Duolun County, Inner Mongolia Autonomous Region, from 2006 to 2017. The specific steps are as follows:

[0056] Step 1: Construct expressions for the effects of soil moisture, vegetation growth, and vegetation type on land surface evapotranspiration;

[0057] Same as step one in "3. Technical Solution".

[0058] Step Two: Basic Data Preparation;

[0059] The basic data included watershed boundaries, monthly runoff, meteorological data, soil moisture content, and vegetation indices from 2006 to 2017. Meteorological data included precipitation, net surface radiation flux, relative humidity, air temperature, air pressure, and wind speed. Net surface radiation flux, soil moisture content, and changes in watershed water storage were obtained from FLDAS land surface assimilation data (https: / / ldas.gsfc.nasa.gov / fldas), while land cover type and vegetation indices were obtained from MODIS vegetation index products (https: / / modis.gsfc.nasa).

[0060] For the remaining meteorological data, such as precipitation, relative humidity, temperature, air pressure, and wind speed, the data for the entire basin is obtained by interpolation of the station monitoring data.

[0061] Step 3: Refer to the calculation of evapotranspiration;

[0062] Meteorological data such as precipitation, net surface radiation flux, relative humidity, temperature, air pressure, and wind speed are substituted into formulas (4)-(8) to estimate the reference evapotranspiration of the watershed. Soil moisture content is substituted into formula (3) to calculate the relative soil moisture content.

[0063] The estimated watershed reference evapotranspiration, calculated soil moisture content, collected vegetation index data, and the KCM of different vegetation types (grassland, deciduous forest, coniferous forest, broad-leaved forest, etc.) and different crops (corn, wheat, vegetables, etc.) obtained from step four were substituted into formula (1) to estimate the distribution of land surface evapotranspiration in the watershed. The statistically average reference evapotranspiration of the entire watershed over many years is 736.2 mm.

[0064] Step 4: Solve for the maximum evapotranspiration coefficient (K) for different vegetation types CM );

[0065] Based on land cover type, K is used to classify different crop types in farmland and different vegetation types in non-farmland. CM The range of variation is set to 0-1, for the water surface K CMSet to 1. For different vegetation types such as grassland, deciduous forest, coniferous forest, and broadleaf forest, and different crops such as corn, wheat, and vegetables, K is selected from the land cover type data. CM By varying the step size from 0 to 1 in a step size of 0.01, an exhaustive method was used to set different vegetation and crop K values. CM The coefficient data set combines meteorological data such as precipitation, net surface radiation flux, relative humidity, temperature, air pressure, and wind speed, as well as K... CM Substitute the coefficient data into formulas (1)-(8) to calculate K. CM The evapotranspiration of the basin is collected from different coefficient data combinations in the coefficient data set. The "K" value that is closest to the total evapotranspiration of the basin calculated by formula (9) is then selected. CM "Coefficient data combination".

[0066] Based on data from 2006 to 2017, calculate K for cultivated land, grassland, forest land, and soil. CM The values ​​are 1.8, 1.8, 2.0, and 1.1 respectively.

[0067] Step 5: Estimate the distribution of land surface evapotranspiration;

[0068] Meteorological data such as precipitation, net surface radiation flux, relative humidity, temperature, air pressure, and wind speed are substituted into formulas (4)-(8) to estimate the reference evapotranspiration of the watershed. Soil moisture content is substituted into formula (3) to calculate the relative soil moisture content.

[0069] The estimated watershed reference evapotranspiration, calculated soil moisture content, collected vegetation index data, and the K values ​​obtained from step four for different vegetation types such as grassland, deciduous forest, coniferous forest, and broadleaf forest, as well as different crops such as corn, wheat, and vegetables, are used. CM Substituting into formula (1), the distribution of land surface evapotranspiration in the watershed is estimated.

[0070] Based on the estimated distribution of land surface evapotranspiration in the watershed (see appendix) Figure 2 High evapotranspiration areas are located in the eastern upper reaches of the basin near the river channel and the southwestern lower reaches. The former is characterized by high evapotranspiration coefficients due to forest cover and high soil moisture near the river channel, while the latter is located at the basin outlet, with low elevation and high soil moisture. Low evapotranspiration areas are located in the northern part of the basin and the southern slopes outside the basin. Statistics show that the average annual net water consumption due to evapotranspiration in the basin for forest, farmland, and grassland is approximately 328, 308, and 302 mm, respectively. Figure 3 The average annual runoff-to-precipitation ratio is approximately 6%, 8%, and 9%.

[0071] The basin is located in a (semi)arid region, with high net water consumption due to evapotranspiration, accounting for more than 91% of the total precipitation. Forest land has the highest water consumption, followed by farmland, and grassland has the lowest.

[0072] Step Six: Calculate the estimation error of land surface evapotranspiration;

[0073] The land surface evapotranspiration distribution data estimated in step five are statistically analyzed according to the watershed boundary to obtain the total land surface evapotranspiration within the watershed.

[0074] The following formula is used to calculate the estimation error of land surface evapotranspiration:

[0075] Rerror = (aET - aET) WB ) / aET WB (10)

[0076] Where Rerror is the error (dimensionless), aET WB The total evapotranspiration (mm) of the basin is calculated using formula (9) in step four. aET is the total land surface evapotranspiration (mm) obtained by statistically analyzing the land surface evapotranspiration distribution data estimated in step five according to the basin boundary.

[0077] The relative error of the actual evapotranspiration and water balance calculation for the entire basin from 2006 to 2017, based on the coefficients obtained in step four, is 1.1%.

Claims

1. A method for estimating land surface evapotranspiration that considers soil moisture, vegetation type, and growth status, characterized in that, The specific steps are as follows: Step 1: Construct expressions for the effects of soil moisture, vegetation growth, and vegetation type on land surface evapotranspiration; The quantitative formula for the impact of soil moisture, vegetation growth, and vegetation type on land surface evapotranspiration is as follows: aET=[(K CM -1)×NDVI+1]×θ R ×ET0 (1) In the above formula, aET represents the actual evapotranspiration from the land surface; K CM νa represents the evapotranspiration coefficient of different vegetation types at their optimal growth stage, i.e., the maximum evapotranspiration coefficient; NDVI is the vegetation index used to represent vegetation growth; θ R ET0 represents the relative soil moisture content; ET0 represents the reference evapotranspiration. Step Two: Preparation of Basic Data; The basic data includes seven categories: watershed boundaries, meteorology, land cover type, watershed outlet runoff, soil moisture content, vegetation index, and watershed water storage changes; meteorological data includes precipitation, net surface radiation flux, relative humidity, air temperature, air pressure, and wind speed. Step 3: Refer to the calculation of evapotranspiration; The calculation formula is: Where ET0 is the reference evaporation rate, in mm / day; Δ is the slope of the vapor pressure curve, in kPa / ℃; R n Net surface radiation flux, unit: MJ / m 2 / day; G is soil heat flux, unit: MJ / m 2 / day, G takes R n 5%; γ is the wet surface constant, unit: kPa / ℃; T is the air temperature at a height of 2 meters above the ground, unit: ℃; u2 is the wind speed at a height of 2 meters above the ground, unit: m / s; e s Saturated vapor pressure, unit: kPa; e a Vapor pressure; Unit: kPa; Step 4: Solve for the maximum evapotranspiration coefficient K for different vegetation types CM ; The total evapotranspiration of the watershed is calculated using the following water balance formula: aET WB =Pre-ΔW-Q (3) Among them, aET WB For the water balance formula, calculate the total evapotranspiration of the basin, where Pre is the basin precipitation, Q is the basin outlet runoff, and ΔW is the change in basin water storage. Step 5: Estimate the distribution of land surface evapotranspiration; Substitute the data of precipitation, net surface radiation flux, relative humidity, temperature, air pressure, and wind speed into formula (2) to estimate the reference evapotranspiration of the watershed; and calculate the relative soil moisture content. The estimated watershed reference evapotranspiration, calculated soil moisture content, collected vegetation index data, and K values ​​obtained from step four for different vegetation types (grassland, deciduous forest, coniferous forest, broadleaf forest) and different crops (corn, wheat, and vegetables) are combined. CM Substituting into formula (1), the distribution of land surface evapotranspiration in the watershed can be estimated; Step Six: Calculate the estimation error of land surface evapotranspiration; The land surface evapotranspiration distribution data estimated in step five are statistically analyzed according to the watershed boundary to obtain the total land surface evapotranspiration within the watershed. The following formula is used to calculate the estimation error of land surface evapotranspiration: Error=(aET-aET WB ) / aET WB (4) Where Rerror is the error, aET WB The total evapotranspiration of the basin is calculated using formula (3). aET is the total land surface evapotranspiration within the basin obtained by statistically analyzing the land surface evapotranspiration distribution data estimated in step five according to the basin boundary.

2. The method for estimating land surface evapotranspiration considering soil moisture, vegetation type, and growth status according to claim 1, characterized in that: The formula for expressing the NDVI vegetation index in step one is: In the above formula, R Nir R represents the surface reflectance in the near-infrared band. Green This represents the surface reflectance in the green band.

3. The method for estimating land surface evapotranspiration considering soil moisture, vegetation type, and growth status according to claim 1, characterized in that: The formula for expressing the relative soil moisture content in step one is: In the above formula, θ R W represents the relative soil moisture content; W represents the soil moisture content. m W represents the soil moisture content at which the soil wilts. f This represents the saturated water content of the soil.

4. The method for estimating land surface evapotranspiration considering soil moisture, vegetation type, and growth status according to claim 1, characterized in that: In step three, the formula for calculating the slope of the vapor pressure curve is:

5. The method for estimating land surface evapotranspiration considering soil moisture, vegetation type, and growth status according to claim 1, characterized in that: In step three, the formula for calculating the wet / dry constant is: γ=0.665×10 -3 P (8) Where P is air pressure, in kPa.

6. The method for estimating land surface evapotranspiration considering soil moisture, vegetation type, and growth status according to claim 1, characterized in that: In step three, the formula for calculating saturated vapor pressure is:

7. The method for estimating land surface evapotranspiration considering soil moisture, vegetation type, and growth status according to claim 1, characterized in that: In step three, the formula for calculating vapor pressure is: RH represents relative humidity.

8. The method for estimating land surface evapotranspiration considering soil moisture, vegetation type, and growth status according to claim 1, characterized in that: In step four, based on land cover type, K is classified into different crop types in farmland and different vegetation types in non-farmland. CM The range of variation is set to 0-1, for the water surface K CM Set to 1; for different vegetation types (grassland, deciduous forest, coniferous forest, broadleaf forest) and different crops (corn, wheat, vegetables) in the land cover type data, K... CM By varying the step size from 0 to 1 in a step size of 0.01, an exhaustive method was used to set different vegetation and crop K values. CM The coefficient data set combines meteorological data such as precipitation, net surface radiation flux, relative humidity, temperature, air pressure, and wind speed, as well as K... CM Substitute the coefficient data into formulas (1)-(3) to calculate K. CM The evapotranspiration of the basin is calculated using different coefficient data combinations in the coefficient data set; and the K value that is closest to the total evapotranspiration of the basin calculated by formula (3) is selected. CM Combination of coefficient data.

Images