CN105913149A - Method for evaluating daytime average evapotranspiration according to multi-temporal remote sensing data and meteorological data - Google Patents

Abstract

The invention discloses a method for evaluating daytime average evapotranspiration according to multi-temporal remote sensing data and meteorological data. The method comprises the following steps of calculating a remote sensing ellipse parameter through a land surface temperature (LST) and a net surface shortwave radiation (NSSR), calculating evapotranspiration model coefficients n0-n5 by means of ALEX model analog data; and estimating a daytime average evapotranspiration ETd by means of the remote sensing ellipse parameter and the evapotranspiration model coefficients n0-n5. According to the method of the invention, a multi-temporal earth observation advantage of a stationary meteorological satellite is sufficiently utilized. Through multi-temporal satellite data and meteorological data, the daytime average evapotranspiration is directly obtained. The method settles a problem that the daytime average evapotranspiration can be obtained through inversing instantaneous evapotranspiration and performing time scale expansion in prior art, and furthermore realizes direct estimation for the daytime average evapotranspiration based on FY-2 stationary meteorological satellite data.

Description

A kind of method combining multi-temporal remote sensing data and meteorological data estimation average evapotranspiration on daytime

Technical field

The application belongs to quantitative remote sensing field, specifically, relates to a kind of associating multi-temporal remote sensing data Method with meteorological data estimation average evapotranspiration on daytime.

Background technology

Evapotranspiration is the important component part of water circulation and Land surface energy budget, at Terrestrial soil Earth-vegetation-Atmosphere System serves the effect of key.The more important thing is, evapotranspiration is also simultaneously Key parameter in the fields such as regional scale agricultural, weather, the hydrology and ecology, in region or even Global Agriculture production, climate change and hydrographic water resource research play an important role.Can To say, evapotranspiration is closely bound up with the productive life of the mankind.

Existing evapotranspiration acquisition methods is broadly divided into traditional earth's surface observation procedure and remote sensing is anti- Drill method.Wherein, traditional Remote sensing sends out observation technology such as lysimeter, ripple ratio energy is put down Weighing apparatus device, eddy correlation system etc. are normally based on single-point, the measurement of little yardstick, it is difficult to promote To regional scale, it is impossible to meet the current forward position research demand to regional scale evapotranspiration data.

The research developing into evapotranspiration of remote sensing technology provides a kind of new opportunity.But, when When front remote sensing evapotranspiration model utilizes polar-orbiting satellite single mostly, phase data carries out evapotranspiration Estimation, the most at most can only obtain 1-2 effectively observation, it is impossible to directly obtains day yardstick and puts down The data of this most more urgent needs of equal evapotranspiration；Can only directly obtain wink Time evapotranspiration data, and in the hydrology and water resources management are applied, the evapotranspiring and transmit of day yardstick Breath more has practical significance.

Summary of the invention

In view of this, the application can only directly obtain wink from remotely-sensed data present in prior art Time evapotranspiration problem, it is provided that a kind of associating multi-temporal remote sensing data and meteorological data estimation are put down daytime All methods of evapotranspiration, provide effective for fields such as regional scale agricultural, weather, the hydrology and ecologies Parameter.

In order to solve above-mentioned technical problem, this application discloses a kind of associating multi-temporal remote sensing data gentle The method of image data estimation average evapotranspiration on daytime, does not writes.

Compared with prior art, the application can obtain and include techniques below effect:

1) compared to polar-orbiting satellite data, GMS is to any one in its overlay area Individual given pixel all has fixing observation angle, and one day can provide 48-96 moment Multispectral data (China's wind and cloud GMS temporal resolution 30 minutes, Europe the Secondary GMS temporal resolution 15 minutes).The GMS this high time The feature of resolution earth observation, can not only can provide earth's surface, region to join as polar-orbiting satellite Number, but also the Land Surface Parameters diurnal variation information that polar-orbiting satellite is not provided that can be obtained.Connection Close GMS multidate observation data and meteorological data directly obtains and averagely evapotranspires daytime Send out, it is possible to be effectively the area researches such as regional scale agricultural, weather, the hydrology and ecology and answer With providing real-time, reliable and accurate key parameter, there is highly important using value.

2) present invention makes full use of the advantage of GMS multidate every day earth observation, By multidate satellite data and meteorological data, directly obtain average evapotranspiration on daytime, solve The first instantaneous evapotranspiration of inverting time scale extension must be then carried out just present in prior art The problem that can obtain day yardstick evapotranspiration, it is achieved that based on FY-2 GMS data The directly estimation of the day average evapotranspiration of yardstick.

Certainly, the arbitrary product implementing the application must be not necessarily required to reach all the above simultaneously Technique effect.

Accompanying drawing explanation

Accompanying drawing described herein is used for providing further understanding of the present application, constitutes the one of the application Part, the schematic description and description of the application is used for explaining the application, is not intended that this Shen Improper restriction please.In the accompanying drawings:

Fig. 1 is that the application combines multi-temporal remote sensing data and meteorological data estimation average evapotranspiration on daytime The techniqueflow chart of method；

Fig. 2 is the surface temperature (K) during the 10:00 on the 28th August in 2010 of the application inverting；

Fig. 3 is the surface temperature (K) during the 11:00 on the 28th August in 2010 of the application inverting；

Fig. 4 is the surface temperature (K) during the 12:00 on the 28th August in 2010 of the application inverting；

Fig. 5 is the surface temperature (K) during the 13:00 on the 28th August in 2010 of the application inverting；

Fig. 6 is the surface temperature (K) during the 14:00 on the 28th August in 2010 of the application inverting；

Fig. 7 is the surface temperature (K) during the 15:00 on the 28th August in 2010 of the application inverting；

Fig. 8 is the surface temperature (K) during the 16:00 on the 28th August in 2010 of the application inverting；

Fig. 9 is the surface temperature (K) during the 17:00 on the 28th August in 2010 of the application inverting；

Figure 10 be the application estimate 10:00 on the 28th August in 2010 time earth's surface net short-wave radiation (W/m²)；

Figure 11 be the application estimate 11:00 on the 28th August in 2010 time earth's surface net short-wave radiation (W/m²)；

Figure 12 be the application estimate 12:00 on the 28th August in 2010 time earth's surface net short-wave radiation (W/m²)；

Figure 13 be the application estimate 13:00 on the 28th August in 2010 time earth's surface net short-wave radiation (W/m²)；

Figure 14 be the application estimate 14:00 on the 28th August in 2010 time earth's surface net short-wave radiation (W/m²)；

Figure 15 be the application estimate 15:00 on the 28th August in 2010 time earth's surface net short-wave radiation (W/m²)；

Figure 16 be the application estimate 16:00 on the 28th August in 2010 time earth's surface net short-wave radiation (W/m²)；

Figure 17 be the application estimate 17:00 on the 28th August in 2010 time earth's surface net short-wave radiation (W/m²)；

Figure 18 is per day evapotranspiration August 28 in 2010 (mm) that the application utilizes this method to estimate.

Detailed description of the invention

Presently filed embodiment is described in detail, thereby to this Shen below in conjunction with drawings and Examples The most how application technology means solve technical problem and reach the process that realizes of technology effect and can fully manage Solve and implement according to this.

The present invention provides a kind of associating multi-temporal remote sensing data and meteorological data estimation average evapotranspiration on daytime Method, as it is shown in figure 1, comprise the following steps:

1) Remote sensing parameters is calculated by surface temperature LST and earth's surface net short-wave radiation NSSR；

The data source of the present invention is from China's wind and cloud series GMS (FY-2), the meter of parameter Calculation flow process is as follows:

1.1) Surface Temperature Retrieval

Extract the bright of wind and cloud two Detection Using Thermal Infrared Channel data IR1 of series GMS FY-2 and IR2 Degree temperature, utilizes Split-window algorithm inverting surface temperature:

$LST=a_0+(a_1+a_2\·\frac{1-\mathrm{\ϵ}}{\mathrm{\ϵ}}+a_2\·\frac{\mathrm{\Δ}\mathrm{\ϵ}}{{\mathrm{\ϵ}}^2})\frac{T_{IR1}+T_{IR2}}{2}+(a_4+a_5\·\frac{1-\mathrm{\ϵ}}{\mathrm{\ϵ}}+a_6\·\frac{\mathrm{\Δ}\mathrm{\ϵ}}{{\mathrm{\ϵ}}^2})\frac{T_{IR1}-T_{IR2}}{2}---\left(1\right)$

Wherein, LST is surface temperature；T_IR1And T_IR2It it is the brightness temperature of two Detection Using Thermal Infrared Channels of FY-2 data Degree；ε is the meansigma methods of the two Detection Using Thermal Infrared Channel emissivity；Δ ε is that two Detection Using Thermal Infrared Channels compare spoke Penetrate the difference of rate；a₀～a₆It is division window coefficient, can be by atmospheric radiation transmission known to those skilled in the art (such as MODTRAN) simulation obtains.

1.2) earth's surface net short-wave radiation estimation

Earth's surface net short-wave radiation estimation equation:

NSSR=(1-A) S₀·cos(SZA)·d_r·τ (2)

Wherein, NSSR is earth's surface net short-wave radiation；A is surface albedo, by wind and cloud visible light wave range Data acquisition；S₀It is solar constant (1367W/m²)；SZA is solar zenith angle, can be by satellite data Obtain；d_rIt it is the solar distance represented with astronomical unit；τ is atmospheric transmittance, under the conditions of fine day, τ can be expressed as:

$\mathrm{\τ}=e^{-{\mathrm{\τ}}_{H_{2}O}}\·e^{-{\mathrm{\τ}}_{O_3}}\·e^{-{\mathrm{\τ}}_{Aer+{\mathrm{CO}}_2+O_2}}---\left(3\right)$

Wherein,WithIt is respectively as follows:

${\mathrm{\τ}}_{H_{2}O}=0.102{\[W/cos\left(SZA\right)\]}^{0.29}---\left(4\right)$

${\mathrm{\τ}}_{O_3}=0.041{\[U_{O_3}/cos\left(SZA\right)\]}^{0.57}---\left(5\right)$

${\mathrm{\τ}}_{Aer+{\mathrm{CO}}_2+O_2}=0.1012/cos\left(SZA\right)---\left(6\right)$

Wherein,Being ozone content (atm.cm), every day, near real-time ozone content can be from TEMIS (Tropospheric Emission Monitoring Internet Service) obtains (http://www. temis.nl/protocols/O3total.html)；

W is Water Vapor Content (g/cm²), available two Detection Using Thermal Infrared Channel T of FY-2 data_IR1And T_IR2 The ratio inverting of Land surface emissivity obtain, its computing formula is:

$W=c_1+c_2\×\frac{{\mathrm{\ϵ}}_i}{{\mathrm{\ϵ}}_j}\×\frac{\underset{k=1}{\overset{N}{\mathrm{\Σ}}}(T_{i,k}-{\stackrel{\‾}{T}}_i)(T_{j,k}-{\stackrel{\‾}{T}}_j)}{\underset{k=1}{\overset{N}{\mathrm{\Σ}}}{(T_{i,k}-{\stackrel{\‾}{T}}_i)}^2}---\left(7\right)$

Wherein, ε_iAnd ε_jIt is to be the emissivity of two Detection Using Thermal Infrared Channels respectively,WithIt is two thermal infrareds Passage IR1 and IR2 meansigma methods of brightness temperature on the star of (N=7 × 7=49) in 7 neighborhoods,^kRepresent 7 Pixel sequence number in neighborhood.c₁And c₂For coefficient, the function of view zenith angle can be expressed as:

c₁=28.104-14.996/cos (VZA)+3.211/cos²(VZA) (8)

c₂=-28.056+14.954/cos (VZA)-3.206/cos²(VZA) (9)

Wherein, view zenith angle VZA can directly read from remotely-sensed data.

1.3) surface temperature LST and earth's surface net short-wave radiation NSSR is utilized to calculate Remote sensing parameters:

Using surface temperature diurnal variation model to describe the change procedure of surface temperature on daytime, this model can To be expressed as:

LST_day(t)=T₀+T_acos[β(t-t_m)], t ＜ t_s (10)

Wherein, LST_dayT () is the surface temperature (K) of t on daytime, T₀And T_aIt is fitting parameter (K), β is angular frequency, t_mIt is that surface temperature arrives maximum moment (h), t_sBe temperature start decay moment (h).Similarly, earth's surface on daytime net short-wave radiation can also describe with a cosine function:

NSSR_day(t)=S₀+S_acos[α(t-t_r)] (11)

Wherein, NSSR_dayT () is the earth's surface net short-wave radiation (W/m of t on daytime²), S₀And S_aIt is to intend Close parameter (W/m²), α is angular frequency, t_rIt is that earth's surface net short-wave radiation arrives the maximum moment (h).

For ease of calculating, the diurnal variation to surface temperature Yu earth's surface net short-wave radiation is carried out at nondimensionalization Reason:

$x=\frac{{\mathrm{LST}}_{day}\left(t\right)-275}{50}=p_{1}cos\[\mathrm{\β}(t-t_m)\]+q_1---\left(12\right)$

$y=\frac{{\mathrm{NSSR}}_{day}\left(t\right)}{1200}=p_{2}cos\[\mathrm{\α}(t-t_r)\]+q_2---\left(13\right)$

Wherein, x and y is the surface temperature after nondimensionalization processes and earth's surface net short-wave radiation respectively.

Owing to earth's surface net short-wave radiation reaches the moment of maximum the most at noon, and surface temperature reaches With Δ t, the big moment, typically in the afternoon, represents that surface temperature arrives maximum moment and the clean spoke of earth's surface shortwave It is mapped to reach the difference in the moment of maximum:

Δ t=t_m-t_r (14)

Assume angular frequency and the angular frequency phase of earth's surface net short-wave radiation diurnal variation of surface temperature diurnal variation Deng, obtain:

p₂ ²(x-q₁)²-2p₁p₂[cos(β·Δt)](x-q₁)(y-q₂)+p₁ ²(y-q₂)²=[p₁p₂sin(β·Δt)]² (15)

Formula (15) is under the conditions of a given fine day being the Equation of ellipse of a standard, wherein p₁、q₁、p₂、q₂, β and Δ t for given underlying surface (include the soil texture, soil moisture and Vegetative coverage) condition is definite value, then five elliptic parameters are expressed as:

$\left\{\begin{array}{c}{x}_0=q_1\\ y_0=q_2\\ \mathrm{\θ}=\frac{1}{2}{\cot }^{-1}\[\frac{{p_1}^2-{p_2}^2}{2p_1{p}_{2}cos(\mathrm{\β}\·\mathrm{\Δ}t)}\]\\ a=p_{1}sin(\mathrm{\β}\·\mathrm{\Δ}t)\\ b=p_{2}sin(\mathrm{\β}\·\mathrm{\Δ}t)\end{array}\right.---\left(16\right)$

Elliptic parameter shown in formula (16) is i.e. the remote sensing input of this method estimation day yardstick evapotranspiration Parameter, wherein: x₀It is elliptical center abscissa, y₀Being elliptical center vertical coordinate, a is oval semi-major axis, B is oval semi-minor axis, and θ is ELLIPTIC REVOLUTION angle.

Step 2, process meteorological data calculate with evapotranspiration model coefficient

This method estimation average evapotranspiration on daytime needs to calculate 6 evapotranspiration model coefficient n₀～n₅, this A little coefficients are to utilize ALEX (Atmosphere-Land Exchange) modeling data to calculate, specifically For:

First, meteorological data is carried out pretreatment, is organized into the form of ALEX modeling requirement, Meteorological data needed for ALEX is: wind speed, temperature, vapour pressure, shortwave radiation and atmospheric pressure；

Secondly, ALEX model is initialized, arrange the different soil texture, soil moisture with And vegetative coverage condition.Wherein, 12 kinds of soil textures that the soil texture is recommended with reference to international food and agricultural organization Classification；For every kind of soil texture, soil moisture scope is uniform to saturation moisture content from its wilting coefficient Take 10 initial soil moisture values；Vegetation coverage is from 0 to 1 change, and step-length is 0.1.To arrange Good meteorological data forces condition as air, carries out digital simulation, obtain mould under initialization condition Intend data；

Finally, according to simulation different initialize underlying surfaces (include different soils quality, soil moisture and Vegetative coverage condition) the surface temperature on daytime of simulation, net short-wave radiation calculate elliptic parameter, Jin Erji In elliptic parameter and the per day evapotranspiration data of simulation of simulation, utilize method of least square to calculate and evapotranspire Send out model coefficient.

Step 3, average evapotranspiration estimation on daytime:

Obtain remote sensing elliptic parameter utilizing remotely-sensed data and utilize simulation of climatic data to obtain evapotranspiration After model coefficient, the estimation equation of average evapotranspiration on daytime is:

ET_d=n₀+n₁×x₀+n₂×y₀+n₃×a+n₄×b+n₅×θ (17)

Wherein, ET_dIt is average evapotranspiration on daytime (mm)；n₀～n₅It is evapotranspiration model coefficient, can profit Obtain with ALEX modeling；x₀、y₀, a, b and θ be remote sensing elliptic parameter, directly by time many Phase GMS data calculate.

With Maqu County, Yellow River source, Maqu, Luqu County and Ruoergai country for demonstration study area, with FY-2E Data are data source, utilize Yellow River source, Maqu weather and the ring being positioned at this center, demonstration study area The meteorological data of border INTEGRATED SIGHT research station obtains coefficient, and then this demonstration study area of quantitative estimation daytime Average evapotranspiration.

Fig. 2 to Fig. 9 shows and utilizes GMS data Detection Using Thermal Infrared Channel data, according to formula (1) area, Yellow River Source Maqu Beijing time on the daytime 10:00-17:00 of Split-window algorithm inverting by hour Surface temperature.Wherein, the brightness temperature of two Detection Using Thermal Infrared Channels of FY-2E it is extracted as formula (2) Input parameter；Division window coefficient a₀～a₆Obtain according to the simulation of atmospheric radiation transmission MODTRAN. The result of Fig. 2 is one in the method for the present invention and directly inputs parameter, will be with earth's surface net short-wave radiation Come together to calculate remote sensing elliptic parameter；

Figure 10 to Figure 17 shows and utilizes GMS data visible channel data, according to public affairs Area, Yellow River Source Maqu Beijing time on the daytime 10:00-17:00 that formula (2) is estimated by hour earth's surface shortwave clean Radiation.Wherein, the apparent reflectance hourly of FY-2E observation is carried out by simplifying Dark-Object Methods After atmospheric correction obtains Reflectivity for Growing Season, directly as the surface albedo A of short-wave band；Fig. 3 Result be the present invention method in one directly input parameter, will with surface temperature come together calculate Remote sensing elliptic parameter；

Figure 18 shows and utilizes multidate GMS data and meteorological data, according to the present invention's The result of the average evapotranspiration on area, Yellow River Source Maqu daytime of method estimation.The result of Fig. 4 is the present invention Final region average evapotranspiration on the daytime data that obtain of method, it can show that this method can obtain Take the spatial distribution of region average evapotranspiration on daytime.

Described above illustrate and describes some preferred embodiments of invention, but as previously mentioned, it should reason Solve invention and be not limited to form disclosed herein, be not to be taken as the eliminating to other embodiments, And can be used for various other combination, amendment and environment, and can in invention contemplated scope described herein, It is modified by above-mentioned teaching or the technology of association area or knowledge.And what those skilled in the art were carried out Change and change is without departing from the spirit and scope invented, the most all should be in the protection of invention claims In the range of.

Claims (5)

1. combine multi-temporal remote sensing data and a method for meteorological data estimation average evapotranspiration on daytime, It is characterized in that, comprise the following steps:

1) remote sensing elliptic parameter is calculated by surface temperature LST and earth's surface net short-wave radiation NSSR；

2) ALEX modeling data is utilized to calculate evapotranspiration model coefficient n₀～n₅；

3) utilize step 1) in remote sensing elliptic parameter and step 2) in evapotranspiration coefficient n₀～n₅Estimate Calculate average evapotranspiration on daytime ET_d。

Associating multi-temporal remote sensing data the most according to claim 1 and meteorological data estimation are put down daytime All methods of evapotranspiration, it is characterised in that described step 1) in by surface temperature LST and ground Table net short-wave radiation NSSR calculate Remote sensing parameters particularly as follows:

1.1) inverting surface temperature:

Computer is utilized to extract wind and cloud two Detection Using Thermal Infrared Channel data IR1 of series GMS FY-2 With the brightness temperature of IR2, utilize Split-window algorithm inverting surface temperature:

$LST=a_0+\left(a_1+a_2\·\frac{1-\mathrm{\ϵ}}{\mathrm{\ϵ}}+a_2\·\frac{\mathrm{\Δ}\mathrm{\ϵ}}{{\mathrm{\ϵ}}^2}\right)\frac{T_{IR1}+T_{IR2}}{2}+\left(a_4+a_5\·\frac{1-\mathrm{\ϵ}}{\mathrm{\ϵ}}+a_6\·\frac{\mathrm{\Δ}\mathrm{\ϵ}}{{\mathrm{\ϵ}}^2}\right)\frac{T_{IR1}-T_{IR2}}{2}---\left(1\right)$

Wherein, LST is surface temperature；T_IR1And T_IR2It it is the brightness temperature of two Detection Using Thermal Infrared Channels of FY-2 data Degree；ε is the meansigma methods of the two Detection Using Thermal Infrared Channel emissivity；Δ ε is that two Detection Using Thermal Infrared Channels compare spoke Penetrate the difference of rate；a₀～a₆It is unknowm coefficient, atmospheric radiation transmission simulation obtains；

1.2) estimation earth's surface net short-wave radiation:

Earth's surface net short-wave radiation estimation equation:

NSSR=(1-A) S₀·cos(SZA)·d_r·τ (2)

Wherein, NSSR is earth's surface net short-wave radiation；A is surface albedo, by wind and cloud visible light wave range Data acquisition；S₀It is solar constant, 1367W/m²；SZA is solar zenith angle, satellite data obtain Take；d_rIt it is the solar distance represented with astronomical unit；τ is atmospheric transmittance, under the conditions of fine day, and τ table It is shown as:

$\mathrm{\τ}=e^{-{\mathrm{\τ}}_{H_{2}O}}\·e^{-{\mathrm{\τ}}_{O_3}}\·e^{-{\mathrm{\τ}}_{Aer+{\mathrm{CO}}_2+O_2}}---\left(3\right)$

Wherein,WithIt is respectively as follows:

${\mathrm{\τ}}_{H_{2}O}=0.102{\[W/cos\left(SZA\right)\]}^{0.29}---\left(4\right)$

${\mathrm{\τ}}_{O_3}=0.041{\[U_{O_3}/cos\left(SZA\right)\]}^{0.57}---\left(5\right)$

${\mathrm{\τ}}_{Aer+{\mathrm{CO}}_2+O_2}=0.1012/cos\left(SZA\right)---\left(6\right)$

Wherein,Ozone content (atm.cm), every day near real-time ozone content from TEMIS (Tropospheric Emission Monitoring Internet Service) obtains；W is atmospheric water Vapour content (g/cm²), utilize two Detection Using Thermal Infrared Channel T of FY-2 data_IR1And T_IR2Land surface emissivity it Obtain than inverting；

1.3) surface temperature LST and earth's surface net short-wave radiation NSSR is utilized to calculate remote sensing elliptic parameter:

Surface temperature diurnal variation model is used to describe the change procedure of surface temperature on daytime, this model table It is shown as:

LST_day(t)=T₀+T_acos[β(t-t_m)], t ＜ t_s (10)

Wherein, LST_dayT () is the surface temperature (K) of t on daytime, T₀And T_aIt is fitting parameter (K), β is angular frequency, t_mIt is that surface temperature arrives maximum moment (h), t_sBe temperature start decay moment (h)；Similarly, earth's surface on daytime net short-wave radiation describes with a cosine function:

NSSR_day(t)=S₀+S_acos[α(t-t_r)] (11)

Wherein, NSSR_dayT () is the earth's surface net short-wave radiation (W/m of t on daytime²), S₀And S_aIt is to intend Close parameter (W/m²), α is angular frequency, t_rIt is that earth's surface net short-wave radiation arrives the maximum moment (h)；

For ease of calculating, the diurnal variation to surface temperature Yu earth's surface net short-wave radiation is carried out at nondimensionalization Reason:

$x=\frac{{\mathrm{LST}}_{day}\left(t\right)-275}{50}=p_{1}cos\[\mathrm{\β}(t-t_m)\]+q_1---\left(12\right)$

$y=\frac{{\mathrm{NSSR}}_{day}\left(t\right)}{1200}=p_{2}cos\[\mathrm{\α}(t-t_r)\]+q_2---\left(13\right)$

Wherein, x and y is the surface temperature after nondimensionalization processes and earth's surface net short-wave radiation respectively；

Owing to earth's surface net short-wave radiation reaches the moment of maximum the most at noon, and surface temperature reaches With Δ t, the big moment, typically in the afternoon, represents that surface temperature arrives maximum moment and the clean spoke of earth's surface shortwave It is mapped to reach the difference in the moment of maximum:

Δ t=t_m-t_r (14)

Assume angular frequency and the angular frequency phase of earth's surface net short-wave radiation diurnal variation of surface temperature diurnal variation Deng, obtain:

p₂ ²(x-q₁)²-2p₁p₂[cos(β·Δt)](x-q₁)(y-q₂)+p₁ ²(y-q₂)²=[p₁p₂sin(βΔt)]² (15)

Formula (15) is under the conditions of a given fine day being the Equation of ellipse of a standard, wherein p₁、q₁、p₂、q₂, β and Δ t be definite value for given land surface condition, then five elliptic parameters It is expressed as:

$\left\{\begin{array}{c}{x}_0=q_1\\ y_0=q_2\\ \mathrm{\θ}=\frac{1}{2}{\cot }^{-1}\[\frac{{p_1}^2-{p_2}^2}{2p_1{p}_{2}cos(\mathrm{\β}\·\mathrm{\Δ}t)}\]\\ a=p_{1}sin(\mathrm{\β}\·\mathrm{\Δ}t)\\ b=p_{2}sin(\mathrm{\β}\·\mathrm{\Δ}t)\end{array}\right.---\left(16\right)$

Elliptic parameter shown in formula (16) is i.e. the remote sensing input of this method estimation day yardstick evapotranspiration Parameter, wherein: x₀It is elliptical center abscissa, y₀Being elliptical center vertical coordinate, a is oval semi-major axis, B is oval semi-minor axis, and θ is ELLIPTIC REVOLUTION angle.

Associating multi-temporal remote sensing data the most according to claim 2 and meteorological data estimation are put down daytime All methods of evapotranspiration, it is characterised in that described W utilizes two Detection Using Thermal Infrared Channel T of FY-2 data_IR1With T_IR2The ratio inverting of Land surface emissivity obtain, its computing formula is:

$W=c_1+c_2\×\frac{{\mathrm{\ϵ}}_i}{{\mathrm{\ϵ}}_j}\×\frac{\underset{k=1}{\overset{N}{\Σ}}(T_{i,k}-{\stackrel{\‾}{T}}_i)(T_{j,k}-{\stackrel{\‾}{T}}_j)}{\underset{k=1}{\overset{N}{\Σ}}{(T_{i,k}-{\stackrel{\‾}{T}}_i)}^2}---\left(7\right)$

Wherein, ε_iAnd ε_jIt is to be the emissivity of two Detection Using Thermal Infrared Channels respectively,WithIt is two thermal infrareds The meansigma methods of brightness temperature on passage IR1 and IR2 star in 7 neighborhoods, wherein, 7 neighborhoods are concrete Being N=7 × 7=49, k represents the pixel sequence number in 7 neighborhoods；c₁And c₂For coefficient, it is expressed as observing sky The function of drift angle:

c₁=28.104-14.996/cos (VZA)+3.211/cos²(VZA) (8)

c₂=-28.056+14.954/cos (VZA)-3.206/cos²(VZA) (9)

Wherein, view zenith angle VZA directly reads from remotely-sensed data.

Associating multi-temporal remote sensing data the most according to claim 3 and meteorological data estimation are put down daytime The all method of evapotranspiration, described steps 2) in the ALEX modeling data that utilizes calculate evapotranspiration mould Type coefficient n₀～n₅Particularly as follows:

2.1) meteorological data is carried out pretreatment, be organized into the form of ALEX modeling requirement, ALEX Required meteorological data is: wind speed, temperature, vapour pressure, shortwave radiation and atmospheric pressure；

2.2) ALEX model is initialized, arrange the different soil texture, soil moisture and Vegetative coverage condition；Wherein, 12 kinds of soil textures that the soil texture is recommended with reference to international food and agricultural organization divide Class；For every kind of soil texture, soil moisture scope uniformly takes to saturation moisture content from its wilting coefficient 10 initial soil moisture values；Vegetation coverage is from 0 to 1 change, and step-length is 0.1；By put in order Meteorological data forces condition as air, carries out digital simulation under initialization condition, obtains simulating number According to；

2.3) surface temperature on daytime of simulation of underlying surface, the clean spoke of shortwave are initialized according to the different of simulation Penetrate calculating elliptic parameter, and then per day evapotranspiration data based on the elliptic parameter simulated with simulation, Method of least square is utilized to calculate evapotranspiration model coefficient.

Associating multi-temporal remote sensing data the most according to claim 4 and meteorological data estimation are put down daytime The all method of evapotranspiration, described steps 3) in utilize step 1) in remote sensing elliptic parameter and step 2) evapotranspiration coefficient estimate average evapotranspiration on the daytime ET in_dParticularly as follows:

The estimation equation of average evapotranspiration on daytime is:

ET_d=n₀+n₁×x₀+n₂×y₀+n₃×a+n₄×b+n₅×θ (17)

Wherein, ET_dIt is average evapotranspiration on daytime, mm；n₀～n₅It it is evapotranspiration model coefficient；x₀、y₀、 A, b and θ are remote sensing elliptic parameters.