CN103969268A - Method for inverting surface soil physical parameters through passive microwave remote sensing - Google Patents

Abstract

The invention discloses a method for inverting surface soil physical parameters through passive microwave remote sensing. The method comprises the steps as follows: step one, brightness temperature data and MPDIes (microwave polarization difference indexes) of an AMSR-E passive microwave remote sensing brightness temperature image are acquired by a launched satellite-borne passive microwave remote sensing meter AMSR-E; step two, a relationship of smooth surface reflectance, surface soil roughness parameters, MPDIes and vegetation optical thickness is acquired through deduction; step three, surface soil humidity mv and soil roughness h are calculated; and step four, the surface soil temperature TS is calculated. The method is based on a simplified passive microwave radiation transfer equation and combined with the MPDIes, a Q/H surface soil roughness model, a vegetation optical thickness model, a Fresnel equation and the like, so that a physical inversion model for the surface soil determinant attribute parameters including the soil temperature, the surface soil temperature and the soil roughness is constructed; and the method can be used for inverting the surface soil temperature, soil humidity and surface roughness distribution of China and even the global region and has important scientific significance.

Description

The passive microwave remote sensing inversion method of surface soil physical parameter

Technical field

The invention belongs to a kind of soil physics property parameters inversion method based on spaceborne passive microwave remote sensing, be specifically related to the passive microwave remote sensing inversion method of topsoil temperature, topsoil temperature and topsoil roughness etc.

Background technology

Spaceborne passive microwave remote sensing possesses large scale, accurate round-the-clock, accurate round-the-clock monitoring capability, and cloud, rain, atmosphere are had to certain penetration capacity, has unique superiority in surface soil parametric inversion.In the past in 10 years, in the world sequential transmissions multiple spaceborne passive microwave radiometer, the SMOS satellite that mainly comprises hyperchannel scanning microwave radiometer SMMR, microwave radiometer imager SSM/I, Advanced Microwave radiometer AMSR-E and European Space Agency's L-band, China is also just going into overdrive to develop the plan of satellite-borne microwave sensor, No. four (SZ-4) Spaceship Carryings of divine boat in 2002 the multi-modal microwave remote sensor (M of lotus ³rS), wind and cloud-No. 3 Seeds of First Post-flight of transmitting in 2008 microwave moisture meter etc.

The domestic and international research about AMSR or AMSR-E microwave data inverting topsoil temperature at present seldom, its temperature retrieval model mainly concentrates on the structure of empirical model/semiempirical model, these class methods can be with in the same period, with experiment sample district and atmosphere thereof, the similar region of underlying surface situation, but because it lacks physical basis on the determining of regression coefficient, when model is used for zones of different, different time, need correction coefficient again, but research method can be used, and simple, precision is higher.In addition, by the meaning that AMSR-E sensor carries out topsoil temperature retrieval, be not merely to obtain the global topsoil temperature of large scale itself, the more important thing is, the topsoil temperature results obtaining by AMSR-E can improve the precision of other Land Surface Parameters (as soil moisture) inverting.As in the inverting of soil moisture, the parameters such as topsoil roughness, vegetation water cut, topsoil temperature all can exert an influence to inversion result, if can obtain more exactly the wherein value of one or more variablees, so in theory can improve the precision of Soil Moisture Inversion result.

Soil moisture inversion method based on passive microwave remote sensing mainly comprises following four classes: empirical method, forward model method, neural network, data assimilation method.Wherein, the most frequently used is empirical model method and forward model method, and mainstream development direction is forward model method.But empirical model method has obvious shortcoming: 1. empirical model generally can be used the high frequency band of microwave, is subject to atmospheric effect larger; 2. vegetation index is more responsive to vegetation biomass, and along with the rising of audio range frequency, the penetration capacity of microwave region sharply weakens.Therefore, the empirical model method based on vegetation index does not have space-time universality, in actual promotion and application, has significant limitation.Forward model is the equation of describing topographical features parameter and the bright temperature correlationship of sensor reception, and its input variable is each parameter of earth's surface, and output variable is the radiation brightness that sensor receives.The so-called surface parameters inversion based on forward model is exactly by the anti-process that pushes away input variable (Land Surface Parameters) of output variable (the bright temperature of sensor).What wherein, be most widely used is exactly the physical model method of inversion based on passive microwave radiation transfer equation.Physical model is compared empirical model, has good space-time universality, is a kind of soil moisture inversion method comparatively reliably.But the method still has certain weak point: the area that 1. most earth's surfaces Soil Moisture Inversion model covers in many vegetation, its inversion accuracy can be a greater impact, and orographic factor (landform geometric properties, topsoil roughness etc.) even can cause 12% inversion error; 2. part Study need to accurately simulated on the basis of the Land Surface Parameters such as topsoil temperature, dielectric constant, vegetation extinction coefficient, topsoil roughness, just can obtain soil moisture data more accurately, yet the accurate simulation of the Land Surface Parameters such as topsoil temperature, dielectric constant, vegetation extinction coefficient, topsoil roughness is the problem in science needing to be studied too.

Utilize passive microwave remote sensing to carry out the research of topsoil roughness simulation both at home and abroad less, only have some scholars to launch the research of this respect.Majority is the scattering principle based on coarse earth's surface, utilizes main passive microwave remote sensing data, has carried out the simulation of topsoil roughness.First this model is the relational expression that builds the earth's surface backscattering coefficient of earth's surface emissivity and different angles, then, by building the method for look-up table, realizes the simulation of topsoil roughness.Yet, obtaining of high precision topsoil coarseness data is the prerequisite of the crucial Land Surface Parameters such as passive microwave remote sensing inverting topsoil temperature, topsoil temperature, therefore, how to build a set of simple, feasible topsoil roughness physics inverse model, there is important Research Significance.

In research about the inverting of current topsoil temperature passive microwave remote sensing, majority belongs to empirical model or take a large amount of empirical relationships as basic semiempirical model, physical model.Because the surface condition of the whole world or each department, region is intricate, the simulated data that this experimental relation relies on can not contain all earth's surfaces type in the whole world or regional extent, so empirical model cannot meet the Land Surface Parameters simulation of large scale.Physical model is a kind of microwave remote sensing method for large scale topsoil temperature simulation of relative efficiency.Yet, physical model is subject to the impact of atmospheric parameter (steam, gasoloid etc.) and underlying surface covering situation (water, snow, freezing, vegetation covering situation) etc. larger, and the precondition of its accurate inverting topsoil temperature is can carry out accurate atmospheric correction and obtain high-precision underlying surface supplemental characteristic.But the high-precision analog of accurate atmospheric correction and other ammeter parameters is also exactly one of current difficult point of microwave remote sensing.The shortcoming of these aspects, has limited the development of topsoil temperature physics inverse model to a certain extent.

In research about the inverting of topsoil temperature passive microwave remote sensing, because the precision of empirical method (vegetation index) is lower, space-time universality is poor, neural network and data assimilation method refutation process more complicated, and internal relations is undistinct.What therefore, relatively commonly use at present is forward model method (physical model method).But the method still has certain weak point: be 1. not suitable for high vegetation and cover or the higher area of humidity ratio; 2. model needs the Land Surface Parameters such as topsoil temperature, dielectric constant, vegetation extinction coefficient, topsoil roughness as input parameter.Along with launching of the microwave radiometers such as AMSR-E, SMOS, novel microwave sensor can provide dual polarization, multichannel microwave signal data, for building multichannel topsoil temperature retrieval method, provides possibility.And this multichannel inversion method exactly provides possibility for forward model solves a plurality of Land Surface Parameters.Therefore, based on dual polarization, multichannel novel microwave radiation counter certificate, build the topsoil temperature physics inversion method of base multi-channel multi-parameter, by a direction that is passive microwave remote sensing development.

Utilize passive microwave remote sensing to carry out the less of topsoil roughness simulation, and these researchs are substantially all empirical model, semiempirical model, study not deep enough.Yet, obtaining of high precision topsoil coarseness data is the prerequisite of the crucial Land Surface Parameters such as passive microwave remote sensing inverting topsoil temperature, topsoil temperature, therefore, how to build a set of simple, feasible topsoil roughness physics inverse model, there is important Research Significance.

Summary of the invention

It is the passive microwave remote sensing inversion method of surface soil physical parameter that patent of the present invention provides a kind of, its passive microwave radiation transfer equation based on simplifying, and in conjunction with MPDI index, Q/H topsoil roughness model, vegetation opacity model, Fresnel equation etc., built a set of surface soil determinant attribute parameter (soil moisture, topsoil temperature, soil roughness) physics inverse model based on AMSR-E spaceborne passive microwave remote sensing image 6.9GHz and two wave bands of 10.7GHz.

For realizing above object, the technical scheme that the present invention has taked is:

The passive microwave remote sensing inversion method of surface soil physical parameter, it comprises the following steps:

Step 1, the spaceborne passive microwave remote sensing meter AMSR-E by transmitting obtain brightness temperature data and the MPDI index of AMSR-E passive microwave remote sensing brightness temperature image, and described brightness temperature data are the brightness temperature data T of AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band horizontal polarization passage _b6.9h, described MPDI index comprises the MPDI index M PDI of AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band _6.9and the MPDI index M PDI of AMSR-E passive microwave remote sensing brightness temperature image 10.7GHz wave band _10.7;

The relation about level and smooth Reflectivity for Growing Season, topsoil roughness parameter, MPDI index and vegetation opacity is obtained in step 2, deduction, and it comprises the following steps:

Step 21, passive microwave radiation equation is simplified, can be obtained:

$T_{\mathrm{bp}}=T_S(1-r_{\mathrm{sp}}*e^{-2{\mathrm{\τ}}_c})---\left(1\right)$

Wherein, T _bpbe the brightness temperature that spaceborne passive microwave remote sensing meter AMSR-E receives, it comprises the brightness temperature T of vertical polarization wave band _bvbrightness temperature T with the brightness temperature numerical value of horizontal polarization wave band _bh; T _sit is topsoil temperature; r _spbe coarse Reflectivity for Growing Season, it comprises the reflectivity r of coarse earth's surface vertically polarized wave section _svand the reflectivity r of coarse earth's surface horizontal polarized wave section _sh; τ _cbe vegetation opacity, therefore, formula (1) can resolve into:

$T_{\mathrm{bv}}=T_S(1-r_{\mathrm{sv}}*e^{-2{\mathrm{\τ}}_c})---(1-1)$

$T_{\mathrm{bh}}=T_S(1-r_{\mathrm{sh}}*e^{-2{\mathrm{\τ}}_c})---(1-2)$

Step 22, according to Wang and Choudhury(1995) the Q/H topsoil roughness model set up, for AMSR-E dual polarization microwave region, obtain the relation of coarse Reflectivity for Growing Season with the level and smooth Reflectivity for Growing Season of vertical polarization and horizontal polarization wave band:

r _sv=[(1-Q)r _ov+Q*r _oh]e ^-h（2-1）

r _sh=[(1-Q)r _oh+Q*r _ov]e ^-h（2-2）

Wherein, r _ovit is the reflectivity of level and smooth earth's surface vertically polarized wave section; r _ohit is the reflectivity of level and smooth earth's surface horizontal polarized wave section; Q is the polarization difference ratio of soil roughness, 0≤Q<0.5, and h is surface soil roughness, described Q and h form topsoil roughness parameter;

Step 23, by the definition of formula (1-1), (1-2), (2-1), (2-2) substitution MPDI index:

$\mathrm{MPDI}=\frac{\mathrm{Tbv}-\mathrm{Tbh}}{\mathrm{Tbv}+\mathrm{Tbh}}---\left(3\right)$

Can obtain the relation of level and smooth Reflectivity for Growing Season, topsoil roughness parameter, MPDI index and vegetation opacity:

$\frac{1}{\mathrm{MPDI}}=\frac{r_{\mathrm{ov}}+r_{\mathrm{oh}}}{(1-2Q)(r_{\mathrm{ov}}-r_{\mathrm{oh}})}-\frac{2}{(1-2Q)(r_{\mathrm{ov}}-r_{\mathrm{oh}})}{e}^{2{\mathrm{\&tau;}}_c+h}---\left(4\right)$

Step 3, calculating top layer soil moisture m _vwith soil roughness h; It comprises the following steps:

Step 31, by De Jeu (2003) vegetation opacity τ _cexpression:

τ _c=C ₁ln(MPDI) ³+C ₂ln(MPDI) ²+C ₃ln(MPDI)+C ₄（5）

Substitution formula (4) can obtain:

(MPDI-1+2Q) * r _ov+ (MPDI+1-2Q) * r _oh=2 (MPDI) α * e ^β+h(6) wherein, C ₁, C ₂, C ₃and C ₄the coefficient that refers to empirical model, it is constant; α=6C ₁+ 4C ₂+ 2C ₃, β=2C ₄;

Step 32, formula (6) is applied to respectively to AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band and 10.7GHz wave band, can obtains:

(MPDI _6.9-1+2Q)×r _ov6.9+(MPDI _6.9+1-2Q)×r _oh6.9=2(MPDI _6.9) ^α×e ^β+h（7-1）

(MPDI _10.7-1+2Q)×r _ov10.7+(MPDI _10.7+1-2Q)×r _oh10.7=2(MPDI _10.7) ^α×e ^β+h（7-2）

Step 33, single 6.9GHz wave band or 10.7GHz wave band to AMSR-E passive microwave remote sensing brightness temperature image, according to the specific inductive capacity model of Dobson (1985), utilize AIEM to simulate the level and smooth Reflectivity for Growing Season of each frequency band and the relation of topsoil temperature:

r _ov6.9=0.7258*m _v+0.0314 （8-1）

r _oh6.9=0.7757*m _v ^0.4481（8-2）

r _ov10.7=0.7117*m _v+0.0284 （8-3）

$r_{{\mathrm{oh}}_{10.7}}=0.7619*{m_v}^{0.461}---(8-4)$

R wherein _ov6.9level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band vertical polarization passage; r _oh6.9level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band horizontal polarization passage; r _ov10.7level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 10.7GHz wave band vertical polarization passage; r _oh10.7level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 10.7GHz wave band horizontal polarization passage;

Step 34, by formula (8-1) and formula (8-2) substitution formula (7-1), by formula (8-3) and formula (8-4) substitution formula (7-2), obtain respectively formula (9-1) and formula (9-2):

(MPDI _6.9-1+2Q)×(0.7258*m _v+0.0314)+ （9-1）

(MPDI _6.9+1-2Q)×(0.7757m _v ^0.4481)=2(MPDI _6.9) ^α×e ^β+h

(MPDI _10.7-1+2Q)×(0.7117m _v+0.0284)+ （9-2）

(MPDI _10.7+1-2Q)×(0.7619m _v ^0.461)=2(MPDI _10.7) ^α×e ^β+h

Step 35, simultaneous formula (9-1) and formula (9-2), calculate top layer soil moisture m _vwith soil roughness h;

Step 4, calculating top layer soil moisture T _s, being specially simultaneous formula (1-2), (2-2), (5), (8-1), (8-2) can obtain:

$T_S=\frac{e^{(h+\mathrm{\&beta;})}-[(1-Q)(0.7757\&times;{m_v}^{0.4481})+Q(0.7258\&times;m_v+0.0341)]}{T_{b6.9h}\&times;{\left({\mathrm{MPDI}}_{6.9}\right)}^{\mathrm{\&alpha;}}\&times;e^{(h+\mathrm{\&beta;})}}---\left(10\right)$

Described Q is 0.09.

Described α is that-0.0261, β is-2.8073.

The present invention compared with prior art, tool has the following advantages: the present invention is based on AMSR-E passive microwave remote sensing brightness temperature image, from microwave transmission equation, exploration has built the crucial earth's surface parameter A MSR-E passive microwave remote sensing inversion methods such as a set of simple, reliable soil moisture and surface soil roughness, surface soil temperature.Based on the method can inverting China so that topsoil temperature, soil moisture and the roughness of ground surface of Global Regional distribute, there is important scientific meaning.

Accompanying drawing explanation

Fig. 1 is the schematic diagram of passive microwave remote sensing inversion method of the surface soil physical parameter of the embodiment of the present invention;

The national topsoil temperature profile that Fig. 2 obtains for the passive microwave remote sensing inversion method inverting based on surface soil physical parameter of the present invention;

The national topsoil moisture distribution chart that Fig. 3 obtains for the passive microwave remote sensing inversion method inverting based on surface soil physical parameter of the present invention;

The global topsoil roughness distribution plan that Fig. 4 obtains for the passive microwave remote sensing inversion method inverting based on surface soil physical parameter of the present invention.

Embodiment

Below in conjunction with the drawings and specific embodiments, content of the present invention is described in further details.

Embodiment

Take the whole nation or global monitoring as example, by passive microwave remote sensing inversion method, obtain the distribution of (comprising topsoil humidity, soil roughness, the soil moisture) of surface soil physical parameter, comprise the following steps:

(1) first the AMSR-E passive microwave remote sensing brightness temperature image that obtains of spaceborne passive microwave remote sensing meter AMSR-E of transmitting is carried out to large rough rectification, obtains AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz after large rough rectification and brightness temperature and the MPDI(MicrowavePolarization Difference Index microwave polarization difference index of 10.7GHz wave band) data set.

(2) the passive microwave radiation transfer equation based on classical then, utilizes vegetation opacity, smooth earth's surface Reflectivity Model, MPDI model and roughness of ground surface model etc., deduces and obtains topsoil humidity, soil roughness, soil moisture inversion method.

It specifically comprises following content:

(2.1) passive microwave radiation equation is simplified, can be obtained:

$T_{\mathrm{bp}}=T_S(1-r_{\mathrm{sp}}*e^{-2{\mathrm{\&tau;}}_c})---\left(11\right)$

The paper < < that the derivation of formula (11) can be delivered by people such as Wang Lei with reference to < < remote sensing journal > > the 10th volume the 5th phase 656-660 page utilizes a kind of new scaling method > > of AMSR-E microwave radiometer to earth's surface roughness parameter.Wherein, T _bpbe the brightness temperature that spaceborne passive microwave remote sensing meter AMSR-E receives, it comprises the brightness temperature T of vertical polarization wave band _bvbrightness temperature T with the brightness temperature numerical value of horizontal polarization wave band _bh; T _sit is topsoil temperature; r _spbe coarse Reflectivity for Growing Season, it comprises the reflectivity r of coarse earth's surface vertically polarized wave section _svand the reflectivity r of coarse earth's surface horizontal polarized wave section _sh; τ _cbe vegetation opacity, therefore, formula (11) can resolve into:

$T_{\mathrm{bv}}=T_S(1-r_{\mathrm{sv}}*e^{-2{\mathrm{\&tau;}}_c})---(11-1)$

$T_{\mathrm{bh}}=T_S(1-r_{\mathrm{sh}}*e^{-2{\mathrm{\&tau;}}_c})---(11-2)$

(2.2) according to Wang and Choudhury(1995) the Q/H topsoil roughness model set up, for AMSR-E dual polarization microwave region, obtain the relation of coarse Reflectivity for Growing Season with the level and smooth Reflectivity for Growing Season of vertical polarization and horizontal polarization wave band:

r _sv=[(1-Q)r _ov+Q*r _oh]e ^-h（12-1）

R _sh=[(1-Q) r _oh+ Q*r _ov] e ^-h(12-2) wherein, r _ovit is the reflectivity of level and smooth earth's surface vertically polarized wave section; r _ohit is the reflectivity of level and smooth earth's surface horizontal polarized wave section; Q is the polarization difference ratio of soil roughness, 0≤Q<0.5, and through the present invention simulation, obtaining Q, to be approximately equal to 0.09, h be surface soil roughness, Q and h form topsoil roughness parameter.

(2.3) by the definition of formula (11-1), (11-2), (12-1), (12-2) substitution MPDI index:

$\mathrm{MPDI}=\frac{\mathrm{Tbv}-\mathrm{Tbh}}{\mathrm{Tbv}+\mathrm{Tbh}}---\left(13\right)$

Can obtain the relation of level and smooth Reflectivity for Growing Season, topsoil roughness parameter, MPDI index and vegetation opacity:

$\frac{1}{\mathrm{MPDI}}=\frac{r_{\mathrm{ov}}+r_{\mathrm{oh}}}{(1-2Q)(r_{\mathrm{ov}}-r_{\mathrm{oh}})}-\frac{2}{(1-2Q)(r_{\mathrm{ov}}-r_{\mathrm{oh}})}{e}^{2{\mathrm{\&tau;}}_c+h}---\left(14\right)$

(2.4) formula (14) is out of shape, we can obtain:

$(\mathrm{MPDI}-1+2Q)\&times;r_{\mathrm{ov}}+(\mathrm{MPDI}+1-2Q)\&times;r_{\mathrm{oh}}=2\mathrm{MPDI}\&times;e^{2{\mathrm{\&tau;}}_c+h}---\left(15\right)$

By De Jeu (2003) vegetation opacity τ _cexpression:

τ _c=C ₁ln(MPDI) ³+C ₂ln(MPDI) ²+C ₃ln(MPDI)+C ₄（16）

Substitution formula (15) can obtain:

(MPDI-1+2Q) * r _ov+ (MPDI+1-2Q) * r _oh=2 (MPDI) α * e ^β+h(17) wherein, C ₁, C ₂, C ₃and C ₄the coefficient that refers to empirical model, it is constant; m _ijfor experience matrix, α=6C ₁+ 4C ₂+ 2C ₃, β=2C ₄, in preferred embodiment of the present invention, simulation show that α is approximately equal to-0.0261, β is approximately equal to-2.8073.

(2.5) formula (17) is applied to respectively to AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band and 10.7GHz wave band, can obtains:

(MPDI _6.9-1+2Q)×r _ov6.9+(MPDI _6.9+1-2Q)×r _oh6.9=2(MPDI _6.9) ^α×e ^β+h（18-1）

(MPDI _10.7-1+2Q)×r _ov10.7+(MPDI _10.7+1-2Q)×r _oh10.7=2(MPDI _10.7) ^α×e ^β+h（18-2）

(2.6) single 6.9GHz wave band or the 10.7GHz wave band to AMSR-E passive microwave remote sensing brightness temperature image, according to the specific inductive capacity model of Dobson (1985), utilize AIEM to simulate the level and smooth Reflectivity for Growing Season of each frequency band and the relation of topsoil temperature:

r _ov6.9=0.7258*m _v+0.0314 （19-1）

r _oh6.9=0.7757*m _v ^0.4481（19-2）

r _ov10.7=0.7117*m _v+0.0284 （19-3）

$r_{{\mathrm{oh}}_{10.7}}=0.7619*{m_v}^{0.461}---(19-4)$

Wherein: r _ov6.9level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band vertical polarization passage; r _oh6.9level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band horizontal polarization passage; r _ov10.7level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 10.7GHz wave band vertical polarization passage; r _oh10.7level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 10.7GHz wave band horizontal polarization passage;

(2.7), by formula (19-1) and formula (19-2) substitution formula (18-1), by formula (19-3) and formula (19-4) substitution formula (18-2), obtain respectively formula (20-1) and formula (20-2):

(MPDI _6.9-1+2Q)×(0.7258*m _v+0.0314)+ （20-1）

(MPDI _6.9+1-2Q)×(0.7757m _v ^0.4481)=2(MPDI _6.9) ^α×e ^β+h

(MPDI _10.7-1+2Q)×(0.7117m _v+0.0284)+ （20-2）

(MPDI _10.7+1-2Q)×(0.7619m _v ^0.461)=2(MPDI _10.7) ^α×e ^β+h

At formula (20-1) with (20-2), MPDI _6.9and MPDI _10.7can obtain from AMSR-E remote sensing image; Q is approximately equal to 0.09; α is approximately equal to-0.0261, and β is approximately equal to-2.8073, that is to say formula (20-1) and (20-2) only have two common unknown quantity (soil moisture m _vwith soil roughness h).Therefore, can utilize Matlab programming combinatorial formula (20-1) and (20-2) system of equations of composition, known variables soil moisture m is tried to achieve in inverting _vsolution with soil roughness h.

(2.8) simultaneous formula (11-2), (12-2), (16), (19-1), (19-2), calculate top layer soil moisture T _s:

$T_S=\frac{e^{(h+\mathrm{\&beta;})}-[(1-Q)(0.7757\&times;{m_v}^{0.4481})+Q(0.7258\&times;m_v+0.0341)]}{T_{b6.9h}\&times;{\left({\mathrm{MPDI}}_{6.9}\right)}^{\mathrm{\&alpha;}}\&times;e^{(h+\mathrm{\&beta;})}}---\left(21\right)$

Certainly, also can simultaneous formula (11-1), (12-1), (16), (19-1), (19-2) calculate top layer soil moisture T _s.

(3) on the basis of step (2), take respectively the whole nation or global AMSR-E passive microwave remote sensing brightness temperature image and data be survey region, inverting topsoil temperature, soil moisture and soil roughness, result is exported as in Figure 2-4.

Although the present invention describes by specific embodiment, it will be appreciated by those skilled in the art that, without departing from the present invention, can also carry out various conversion and be equal to alternative the present invention.In addition, for particular condition or application, can make various modifications to the present invention, and not depart from the scope of the present invention.Therefore, the present invention is not limited to disclosed specific embodiment, and should comprise the whole embodiments that fall within the scope of the claims in the present invention.

Claims (3)

1. the passive microwave remote sensing inversion method of surface soil physical parameter, is characterized in that, it comprises the following steps:

Step 1, the spaceborne passive microwave remote sensing meter AMSR-E by transmitting obtain brightness temperature data and the MPDI index of AMSR-E passive microwave remote sensing brightness temperature image, and described brightness temperature data are the brightness temperature data T of AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band horizontal polarization passage _b6.9h, described MPDI index comprises the MPDI index M PDI of AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band _6.9and the MPDI index M PDI of AMSR-E passive microwave remote sensing brightness temperature image 10.7GHz wave band _10.7;

The relation about level and smooth Reflectivity for Growing Season, topsoil roughness parameter, MPDI index and vegetation opacity is obtained in step 2, deduction, and it comprises the following steps:

Step 21, passive microwave radiation equation is simplified, can be obtained:

$T_{\mathrm{bp}}=T_S(1-r_{\mathrm{sp}}*e^{-2{\mathrm{\&tau;}}_c})---\left(1\right)$ Wherein, T _bpbe the brightness temperature that spaceborne passive microwave remote sensing meter AMSR-E receives, it comprises the brightness temperature T of vertical polarization wave band _bvbrightness temperature T with the brightness temperature numerical value of horizontal polarization wave band _bh; T _sit is topsoil temperature; r _spbe coarse Reflectivity for Growing Season, it comprises the reflectivity r of coarse earth's surface vertically polarized wave section _svand the reflectivity r of coarse earth's surface horizontal polarized wave section _sh; τ _cbe vegetation opacity, therefore, formula (1) can resolve into:

$T_{\mathrm{bv}}=T_S(1-r_{\mathrm{sv}}*e^{-2{\mathrm{\&tau;}}_c})---(1-1)$

$T_{\mathrm{bh}}=T_S(1-r_{\mathrm{sh}}*e^{-2{\mathrm{\&tau;}}_c})---(1-2)$

Step 22, according to Wang and Choudhury(1995) the Q/H topsoil roughness model set up, for AMSR-E dual polarization microwave region, obtain the relation of coarse Reflectivity for Growing Season with the level and smooth Reflectivity for Growing Season of vertical polarization and horizontal polarization wave band:

r _sv=[(1-Q)r _ov+Q*r _oh]e ^-h（2-1）

r _sh=[(1-Q)r _oh+Q*r _ov]e ^-h（2-2）

Wherein, r _ovit is the reflectivity of level and smooth earth's surface vertically polarized wave section; r _ohit is the reflectivity of level and smooth earth's surface horizontal polarized wave section; Q is the polarization difference ratio of soil roughness, 0≤Q<0.5, and h is surface soil roughness, described Q and h form topsoil roughness parameter;

Step 23, by the definition of formula (1-1), (1-2), (2-1), (2-2) substitution MPDI index:

$\mathrm{MPDI}=\frac{\mathrm{Tbv}-\mathrm{Tbh}}{\mathrm{Tbv}+\mathrm{Tbh}}---\left(3\right)$

Can obtain the relation of level and smooth Reflectivity for Growing Season, topsoil roughness parameter, MPDI index and vegetation opacity:

$\frac{1}{\mathrm{MPDI}}=\frac{r_{\mathrm{ov}}+r_{\mathrm{oh}}}{(1-2Q)(r_{\mathrm{ov}}-r_{\mathrm{oh}})}-\frac{2}{(1-2Q)(r_{\mathrm{ov}}-r_{\mathrm{oh}})}{e}^{2{\mathrm{\&tau;}}_c+h}---\left(4\right)$

Step 3, calculating top layer soil moisture m _vwith soil roughness h; It comprises the following steps:

Step 31, by De Jeu (2003) vegetation opacity τ _cexpression:

τ _c=C ₁ln(MPDI) ³+C ₂ln(MPDI) ²+C ₃ln(MPDI)+C ₄（5）

Substitution formula (4) can obtain:

(MPDI-1+2Q) * r _ov+ (MPDI+1-2Q) * r _oh=2 (MPDI) α * e ^β+h(6) wherein, C ₁, C ₂, C ₃and C ₄the coefficient that refers to empirical model, it is constant; α=6C ₁+ 4C ₂+ 2C ₃, β=2C ₄;

Step 32, formula (6) is applied to respectively to AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band and 10.7GHz wave band, can obtains:

(MPDI _6.9-1+2Q)×r _ov6.9+(MPDI _6.9+1-2Q)×r _oh6.9=2(MPDI _6.9) ^α×e ^β+h（7-1）

(MPDI _10.7-1+2Q)×r _ov10.7+(MPDI _10.7+1-2Q)×r _oh10.7=2(MPDI _10.7) ^α×e ^β+h（7-2）

Step 33, single 6.9GHz wave band or 10.7GHz wave band to AMSR-E passive microwave remote sensing brightness temperature image, according to the specific inductive capacity model of Dobson (1985), utilize AIEM to simulate the level and smooth Reflectivity for Growing Season of each frequency band and the relation of topsoil temperature:

r _ov6.9=0.7258*m _v+0.0314 （8-1）

r _oh6.9=0.7757*m _v ^0.4481（8-2）

r _ov10.7=0.7117*m _v+0.0284 （8-3）

$r_{{\mathrm{oh}}_{10.7}}=0.7619*{m_v}^{0.461}---(8-4)$

Wherein: r _ov6.9level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band vertical polarization passage; r _oh6.9level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 6.9GHz wave band horizontal polarization passage; r _ov10.7level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 10.7GHz wave band vertical polarization passage; r _oh10.7level and smooth Reflectivity for Growing Season for AMSR-E passive microwave remote sensing brightness temperature image 10.7GHz wave band horizontal polarization passage;

Step 34, by formula (8-1) and formula (8-2) substitution formula (7-1), by formula (8-3) and formula (8-4) substitution formula (7-2), obtain respectively formula (9-1) and formula (9-2):

(MPDI _6.9-1+2Q)×(0.7258*m _v+0.0314)+ （9-1）

(MPDI _6.9+1-2Q)×(0.7757m _v ^0.4481)=2(MPDI _6.9) ^α×e ^β+h

(MPDI _10.7-1+2Q)×(0.7117m _v+0.0284)+ （9-2）

(MPDI _10.7+1-2Q)×(0.7619m _v ^0.461)=2(MPDI _10.7) ^α×e ^β+h

Step 35, simultaneous formula (9-1) and formula (9-2), calculate top layer soil moisture m _vwith soil roughness h;

Step 4, calculating top layer soil moisture T _s, being specially simultaneous formula (1-2), (2-2), (5), (8-1), (8-2) can obtain:

$T_S=\frac{e^{(h+\mathrm{\&beta;})}-[(1-Q)(0.7757\&times;{m_v}^{0.4481})+Q(0.7258\&times;m_v+0.0341)]}{T_{b6.9h}\&times;{\left({\mathrm{MPDI}}_{6.9}\right)}^{\mathrm{\&alpha;}}\&times;e^{(h+\mathrm{\&beta;})}}---\left(10\right)$

2. the passive microwave remote sensing inversion method of surface soil physical parameter according to claim 1, is characterized in that, described Q is 0.09.

3. the passive microwave remote sensing inversion method of surface soil physical parameter according to claim 1 and 2, is characterized in that, described α is that-0.0261, β is-2.8073.