CN101187630A - A method for monitoring drought in farmland - Google Patents

Abstract

The invention discloses a method for monitoring farmland drought. The farmland drought monitoring method is to obtain the soil or leaf water content of the observation point on the surface to be monitored, establish a functional relationship between the soil or leaf water content of the observation point and the corresponding index representing the farmland drought, and bring the remote sensing data to Enter the functional relational expression to obtain the water content of soil or plant leaves on the surface to be monitored; wherein, the corresponding index representing the surface drought is determined according to the following method: 1) on the farmland surface with vegetation coverage≤15%, use vertical Drought index; 2) On the surface of the farmland with vegetation coverage > 15% and ≤ 65%, use the vegetation condition albedo drought index; 3) On the surface of the farmland with vegetation coverage > 65% to 100%, use the following three indexes At least one of: short-wave infrared vertical water loss index, vegetation water content index and vegetation water deficit index. The surface drought monitoring method of the invention is suitable for monitoring the surface drought of farmland with different vegetation coverage.

Description

A method for monitoring drought in farmland

technical field

The invention relates to a farmland drought monitoring method.

Background technique

The occurrence process of drought is potential and not easy to discover; the occurrence of drought in farmland is characterized by a large area of influence, which brings serious catastrophic consequences and economic losses; drought involves many factors, such as meteorological parameters, surface water conditions, human socio-economic conditions, etc. Activities and levels of agricultural production and production structure, etc. Remote sensing technology can provide multi-source, multi-dimensional and multi-temporal information of farmland, which opens up a new way for farmland drought monitoring.

The traditional farmland drought monitoring is to monitor the degree and scope of the drought with the data on the observation point, and the meteorological observation data is the most widely used. Due to the periodic and regional characteristics of drought, remote sensing technology is one of the feasible ways to realize drought monitoring in a large area. Since the 1970s, satellite remote sensing drought monitoring has produced many models and methods, such as the NOAA Drought Index (the NOAA Drought Index , NDI, Strommen et al, 1980), the vegetation condition index (Vegetation Condition Index, VCI) based on the normalized difference vegetation index (NDVI) and the temperature state index (Temperature Condition Index, TCI) etc. Among these methods, there is no large-scale farmland drought monitoring method suitable for different growth stages of farmland crops.

Contents of the invention

The purpose of the present invention is to provide a new method for surface drought monitoring using remote sensing technology.

The method for surface drought monitoring provided by the present invention is to obtain the soil or leaf water content of the observation point on the surface to be monitored, and establish a functional relationship between the soil or leaf water content of the observation point and the corresponding index representing the surface drought , bringing the remote sensing data into the functional relationship to obtain the moisture content of the soil or plant leaves on the surface to be monitored;

Wherein, the corresponding index representing the surface drought situation is determined according to the following method:

1) On the surface with vegetation coverage ≤ 15%, the vertical drought index is used;

The vertical drought index is calculated according to the following formula (I):

$\mathrm{<span\; class="notranslate">\; PDI</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; red</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; MR</span>}}_{\mathrm{<span\; class="notranslate">\; nir</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

Among them, PDI is the vertical drought index, M is the slope of the soil line, R _red is the reflectance of the red band after atmospheric correction, and R _nir is the reflectance of the near-infrared band after atmospheric correction;

2) On the surface with vegetation coverage > 15% and ≤ 65%, the vegetation condition albedo drought index is used;

The vegetation condition albedo drought index is calculated according to the following formula (II):

$\mathrm{<span\; class="notranslate">\; VCADI</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDV</span>}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{{\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{{\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; II</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

Among them, VCADI is the vegetation condition albedo drought index, $A_{\max ,\mathrm{NDV}{I}_i}=a+\mathrm{bNDV}{I}_i,$ $A_{\min ,\mathrm{NDV}{I}_i}=a^{\&prime;}+b^{\&prime;}{\mathrm{NDVI}}_i;$ A _{min, NDVIi} and A _{max, NDVIi} are the minimum and maximum albedo when the vegetation index (NDVI) of the area to be measured is equal to a certain value; a, b, a', b' are undetermined coefficients, a and a' are The intercept of the dry edge and the wet edge in the Albedo-NDVI feature space, b and b′ are the slopes of the dry edge and the wet edge in the Albedo-NDVI feature space, respectively, obtained from the scatter plot of the vegetation index and albedo of the area to be measured ( See accompanying drawing 3), A _{i, NDVIi} is the albedo of the point (A _i , NDVI _i ) in the Albedo-NDVI feature space, and NDVI _i is the vegetation index of D (A _i , NDVI _i ) in the Albedo-NDVI feature space;

3) In farmland with vegetation coverage > 65%, use at least one of the following three indexes: short-wave infrared vertical water loss index, vegetation water content index and vegetation water deficit index for monitoring;

The short-wave infrared vertical dehydration index is calculated according to the following formula (III):

$\mathrm{<span\; class="notranslate">\; SPSI</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; SWIR</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; MR</span>}}_{\mathrm{<span\; class="notranslate">\; NIR</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; III</span>}<span\; class="notranslate">\; )</span>\text{<span class="notranslate"> ,</span>}$

Among them, SPSI is the short-wave infrared vertical water loss index, R _SWIR and R _NIR are the reflectance of short-wave red light (1550-1750nm) and near-infrared band (780-900nm) after atmospheric correction, respectively, and M is the NIR-SWIR baseline slope;

Described vegetation moisture content index is calculated according to following formula (IV):

$\mathrm{<span\; class="notranslate">\; VWCI</span>}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; IV</span>}<span\; class="notranslate">\; )</span>$

Among them, VWCI is the vegetation water content index, referring to Figure 4, M is the slope of the soil baseline in the NIR-SWIR space; M ₁ and M ₂ are the slopes of the AB and CD lines, and I ₁ and I ₂ are the slopes of the AB and CD lines, respectively. Intercept, NIR _G and SWIR _G are the albedo values of the near-infrared and short-wave infrared corresponding to the position to be measured;

Described vegetation water deficit index is calculated according to following formula (V):

VWSI＝1-VWCI (V)

Among them, VWSI is the vegetation water deficit index; VWCI is the vegetation water content index, calculated according to formula (IV).

Among them, vegetation coverage refers to the percentage of the vertical projected area of vegetation (including leaves, stems, and branches) on the ground to the total area of the statistical area.

In the above-mentioned method, in the above-mentioned 1), on the farmland surface with vegetation coverage ≤ 15%, the vertical drought index PDI is used, and the M value involved in the calculation formula is the red-near-infrared band reflectance in the two-dimensional space, which is characterized by The slope of the soil line formed by the points of the soil. In the above 2), on the farmland surface with vegetation coverage > 15% and ≤ 65%, the vegetation condition albedo drought index VCADI is used; wherein, the slope and intercept of the dry edge and wet edge are extracted in the Albedo-NDVI feature space ( a, b, a', b'), and calculate VCADI on this basis. In the above 3), on the surface of the farmland where the vegetation coverage is >65% to 100%, the short-wave infrared vertical water loss index or the vegetation moisture content index is used; the parameters involved in each formula are based on the short-wave infrared-near infrared spectral feature space .

In the above method, the water content of the leaves may specifically be the equivalent water content of the leaves, and the vegetation coverage in the above 3) is greater than 65%, which belongs to the case of high vegetation coverage.

In the above method, the R _red can be atmospherically corrected 630-690nm red band reflectance, the R _nir or R _NIR can be atmospherically corrected 780-900nm near-infrared band reflectance, and the R _SWIR can be The R NIR is atmospherically corrected 1550-1750nm shortwave red light band reflectance, and the R _NIR may be atmospherically corrected 780-900nm near-infrared band reflectance.

The present invention utilizes multi-source, multi-dimensional, multi-temporal remote sensing images, starts from the construction of reflectance spectra sensitive to farmland moisture bands or n-dimensional spectral feature space composed of ecological and physical parameters representing farmland drought, and establishes a remote sensing model that can quantitatively reflect farmland water deficit information Grasp the two key indicators of farmland drought, soil moisture and leaf (canopy) water content, in the form of various combinations of spectral reflectance or surface ecological physical parameters, by calculating the distance of the farmland ecological drought index in the multidimensional spectral feature space , quantitatively extract the dynamic information of farmland drought, on this basis, monitor and evaluate the farmland drought, and verify it through the radiative transfer model and field observation data. According to the different conditions of surface vegetation cover, select the corresponding index to represent the drought situation of farmland. Select the appropriate observation point to obtain the soil or leaf water content, and establish a functional relationship with the corresponding index that characterizes the farmland drought, and then invert the soil or crop leaf water content in the entire monitoring area. The corresponding index of the selected characterizing farmland drought situation of the present invention is as follows:

1) In the case of no vegetation coverage and very low vegetation coverage (coverage ≤ 15%), the vertical drought index (PDI) is constructed according to the spatial spectral characteristics of NIR and RED for the remote sensing monitoring of soil moisture indicators, taking into account the soil Moisture content, vegetation coverage, soil organic matter content and differences among different soil types, the model is simple, practical and highly operable.

2) Under the condition of partial cover of farmland vegetation, according to NDVI and albedo feature space, the vegetation conditional albedo drought index (VCADI) is proposed. The introduction of albedo can obtain richer reflection information of ground objects in the hemispheric space, and the reflection characteristics of the entire sunlight range from visible light to infrared can be better reflected, which can more accurately retrieve farmland drought information. The model is easy to operate and has high monitoring accuracy for drought assessment.

3) In the case of full vegetation coverage, three drought monitoring methods based on leaf equivalent water content (EWT) or leaf relative water content (FMC) were proposed for the leaf water content index of drought monitoring in farmland. The present invention finds that the wet and dry conditions in the NIR-SWIR feature space are typical trapezoidal or triangular distributions, and utilizes the leaf radiation transfer model PROSPECT, the canopy radiation transfer model (Lillesaeter/SailH) and the surface-atmosphere radiation transfer model 6S to propose a method for High vegetation coverage canopy moisture content and water deficit monitoring model - short-wave infrared vertical water loss index (SPSI), suitable for high coverage vegetation water content index (VWCI) and vegetation water deficit index (VWSI). The model is suitable for inversion and evaluation of drought monitoring parameters in high vegetation coverage areas.

The present invention utilizes remote sensing technology to monitor farmland drought dynamics, selects different wave bands of remote sensing to establish drought monitoring indices for different growth stages of farmland crops, and combines actual water content data measured at ground observation points to obtain large-area drought conditions in the monitoring area. The practical application test shows that the method is simple, efficient, easy to operate and accurate in results, and can be widely used in farmland drought monitoring in northern and northwestern my country. The surface drought monitoring method of the invention is suitable for monitoring the surface drought of farmland with different vegetation coverage.

Description of drawings

Figure 1 shows the triangular distribution of dry and wet conditions in the red-near-infrared spectral feature space

Figure 2 is a schematic diagram of the Vertical Drought Index (PDI)

Figure 3 is a schematic diagram of the Vegetation Conditional Albedo Drought Index (VCADI)

Figure 4 is a schematic diagram of short-wave infrared vertical water loss index (SPSI) and vegetation water content index (VWCI)

Figure 5 is a schematic diagram of the relationship between vertical drought index and soil moisture

Figure 6 is a schematic diagram of the relationship between the vegetation condition albedo drought index and soil moisture at different depths

Figure 7 shows the comparison between the short-wave infrared vertical water loss index (SPSI) estimated EWT _canopy and field observation data

Figure 8 shows the comparison between the vegetation water content index (VWCI) estimated EWT _canopy and field observation data

Figure 9 shows the comparison between the vegetation water content index (VWCI) estimated FMC and field observation data

Detailed ways

Embodiment 1, the construction of drought monitoring model

For different vegetation coverage, select the appropriate remote sensing band, and calculate the index suitable for characterizing the moisture status of the farmland under the vegetation coverage. The technical solution is as follows:

1. For the surface with no vegetation coverage or low coverage (vegetation coverage ≤ 15%), the NIR-Red spectral feature space scatter diagram constructed using remote sensing related band data presents a typical triangular distribution (as shown in Figure 1), B-C is the soil baseline, and the soil gradually dries from B to C, and the mathematical expression of B-C as the soil baseline can be obtained through the spatial statistical characteristics:

R _nir,s =MR _red,s +I (1)

Among them, R _{red, s} , R _{nir, s} are the reflectance of red light band and near-infrared band after atmospheric correction, M is the slope of the soil line, and I is the intercept of the soil line on the ordinate.

In Figure 1, the vertical line L (Figure 2) passing through the origin of the coordinates and perpendicular to the soil baseline can be obtained to obtain the normal equation (2) of equation (1)

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; nir</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; red</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

In the NIR-Red feature space, the distance from any point E(R _red , R _nir ) to the straight line L can explain the aridity of the surface, that is, the farther away from the line L, the drier the surface is, and vice versa. For the water body, its aridity index is the smallest, which just falls on the origin of the coordinates, and any other object with a certain reflective ability gets closer to the origin as it gets wetter. Generally speaking, the space closest to the L line is distributed in water bodies or wet areas. Spaces away from the L line are drier areas. Therefore, the distance from any point E(R _red , R _nir ) on the NIR-Red feature space to the straight line L can be used to describe the drought situation, and a drought monitoring model based on the NIR-Red spectral space characteristics can be established, that is, the vertical drought Index (Perpendicular Drought Index, PDI).

$\mathrm{<span\; class="notranslate">\; PDI</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; red</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; MR</span>}}_{\mathrm{<span\; class="notranslate">\; nir</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Among them, M is the slope of the soil line, which is obtained by selecting points that fall near the soil line and performing linear regression. R _red is the reflectance of the red light band (630-690nm) after atmospheric correction, and R _nir is the reflectance of the near-infrared band (775-900nm) after atmospheric correction, obtained by radiation calibration for the corresponding red light and near-infrared bands.

2. For the farmland surface with medium surface vegetation coverage (vegetation coverage greater than 15% to less than or equal to 65%), the impact of vegetation coverage needs to be considered. Albedo (Albedo)--Normalized Difference Vegetation Index (NDVI) scatter diagram also has "dry edge" and "wet edge", "dry edge" is the high albedo part, "wet edge" is the low albedo rate part. Similarly, the characteristics of land cover can be reflected by ABC. If the influence of other parameters is not considered, since the vegetation changes the roughness of the surface, in arid and semi-arid areas, the surface albedo decreases with the increase of vegetation, and the increase of soil moisture also causes the albedo to decrease. Soil moisture and vegetation growth are positive. correlation, while there is a negative correlation between vegetation, soil moisture and albedo.

As shown in Figure 3, let albedo=A _i , there is a point D(A _i , NDVI _i ) in the Albedo-NDVI feature space, and the difference between the maximum and minimum albedo on the straight line passing through this point can reflect the point’s The degree of drought, the index is the vegetation condition albedo drought index (Vegetation Condition Albedo Drought Index, VCADI).

$\mathrm{<span\; class="notranslate">\; VCADI</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDV</span>}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{{\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{{\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

in, $A_{\max ,\mathrm{NDV}{I}_i}=a+\mathrm{bNDV}{I}_i,$ $A_{\min ,\mathrm{NDV}{I}_i}=a^{\&prime;}+b^{\&prime;}{\mathrm{NDVI}}_i.$ A _{min, NDVIi} and A _{max, NDVIi} are the minimum and maximum albedo when NDVI is equal to a certain value in the study area, respectively. a, b, a', b' are undetermined coefficients, a and a' are the intercepts of dry and wet edges in the Albedo-NDVI feature space, respectively, b and b' are the dry and wet edges in the Albedo-NDVI feature space The slope of the edge can be obtained through the scatter diagram of NDVI and albedo in the study area, and the points near the AC and BC lines can be selected to obtain the corresponding value by linear regression. A _{i, NDVIi} is the albedo of a certain point D(A _i , NDVI _i ), and NDVI _i is the vegetation index of a certain point D in the Albedo-NDVI feature space.

3. For the surface with high vegetation coverage (vegetation coverage is greater than 65% to 100% coverage), on the spectral feature space scatter diagram (Fig. 4) of near infrared (NIR)-short wave infrared (SWIR), after spatial statistics Analysis can get NIR-SWIR baseline BC, mathematical expression:

R _NIR = MR _SWIR + I (5)

Among them, R _SWIR and R _NIR are the atmospherically corrected short-wave red light and near-infrared band reflectance, M is the slope of the NIR-SWIR baseline, and I represents the intercept of the NIR-SWIR baseline on the ordinate.

Obtain the vertical line L (Fig. 4) that passes through the origin of the coordinates and is perpendicular to the soil baseline, then the normal equation (6) of equation (5) can be obtained

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; NIR</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; SWIR</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

In the NIR-SWIR feature space, the distance from any point G(R _SWIR , R _NIR ) to the straight line L can describe the drought situation on the surface. Generally speaking, the space closest to the L line is distributed in water bodies or wet areas . Spaces away from the L line are drier areas. For the water body, its aridity index is the smallest, which just falls on the origin of the coordinates, and any other object with a certain reflective ability gets closer to the origin as it gets wetter. Therefore, the distance from any point G(R _SWIR , R _NIR ) on the NIR-SWIR feature space to the straight line L can be used to describe the drought situation, and a drought monitoring model based on the NIR-SWIR spectral space characteristics can be established. Infrared Vertical Water Loss Index (Shortwave Infrared Perpendicular Water Stress Index, SPSI).

$\mathrm{<span\; class="notranslate">\; SPSI</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; SWIR</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; MR</span>}}_{\mathrm{<span\; class="notranslate">\; NIR</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

The SPSI is the short-wave infrared vertical water loss index, R _SWIR , R _NIR are the atmospherically corrected short-wave red light and near-infrared band reflectance respectively, and M is the NIR-SWIR baseline slope, which can be obtained by linear regression.

向量越短，说明冠层植被水分含量越高，因此，

可以代表像元在平行于NIR-SWIR基线方向的水分状况。

向量的长度表明像元远离NIR-SWIR基线的距离，即像元远离NIR-SWIR基线的距离越大，

向量越短，像元植被覆盖越好，反之亦然。

和

向量的比值不仅能反映植被水分含量，还能减少大气作用和植被冠层对红光的散射效应。把

和

和

的比值分别命名为植被水分含量指数(Vegetation Water Content Index，VWCI)和植被水分亏缺指数(Vegetation WaterStress Index，VWSI)。As mentioned above, the direction away from the NIR-SWIR baseline and the direction parallel to the NIR-SWIR baseline jointly determine the vegetation moisture content. The vegetation coverage conditions of E, G, and F points are the same, but their vegetation moisture content is quite different. vector, gradually decreasing, The longer the vector, the less vegetation moisture,

The shorter the vector, the higher the water content of the canopy vegetation, therefore,

It can represent the moisture status of the pixel in the direction parallel to the NIR-SWIR baseline.

The length of the vector indicates the distance of the pixel away from the NIR-SWIR baseline, that is, the greater the distance of the pixel away from the NIR-SWIR baseline,

The shorter the vector, the better the vegetation coverage of the cell, and vice versa.

and

The ratio of vectors can not only reflect the moisture content of vegetation, but also reduce the scattering effect of atmospheric action and vegetation canopy on red light. Bundle

and

and

The ratios of these values are named Vegetation Water Content Index (VWCI) and Vegetation Water Stress Index (VWSI) respectively.

$\mathrm{<span\; class="notranslate">\; VWSI</span>}<span\; class="notranslate">\; =</span>\frac{\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; EG</span>}}}{\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; EF</span>}}}<span\; class="notranslate">\; ,</span>$ $\mathrm{<span\; class="notranslate">\; VWCI</span>}<span\; class="notranslate">\; =</span>\frac{\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; GF</span>}}}{\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; EF</span>}}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\frac{\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; EG</span>}}}{\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; EF</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 8</span>}<span\; class="notranslate">\; )</span>$

Set the slope and intercept of lines AB and CD as M ₁ , M ₂ , I ₁ , I ₂ , and their linear equations are respectively NIR＝M ₁ ×SWIR+I ₁ , NIR＝M ₂ ×SWIR+I ₂ , input Point G(SWIR _G , NIR _G ) is known, and EF and BC are parallel, then the equation of EF line can be written as

NIR＝M×(SWIR-SWIR _G )+NIR _G

The coordinates of point E(SWIR _E , NIR _E ) can be obtained by solving the equation system:

${\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; E.</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; ,</span>$ ${\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; E.</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="m"\; filenames="S2007101788074E00031.png"\; state="\{\{state\}\}">m</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}}$

The coordinates of the point F(SWIR _F , NIR _F ) can be obtained by solving the equations:

${\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; ,</span>$ ${\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; f</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}}$

Finally, the distance between E(SWIR _E , NIR _E ), G(SWIR _G , NIR _G ), F(SWIR _F , NIR _F ), the general expression of VWCI can be obtained.

$\mathrm{<span\; class="notranslate">\; VWCI</span>}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 9</span>}<span\; class="notranslate">\; )</span>$

Among them, VWCI is the vegetation water content index, referring to Figure 4, M is the slope of the soil baseline in the NIR-SWIR space; M ₁ and M ₂ are the slopes of the AB and CD lines, respectively, and I ₁ and I ₂ are the AB and CD lines, respectively The intercept of , NIR _G , SWIR _G are the albedo values of the near-infrared and short-wave infrared corresponding to the position to be measured, respectively;

VWCI and VWSI can simultaneously retrieve vegetation water content and water stress.

4. According to the vegetation coverage of the monitoring area, choose one of the above three types of indexes for calculation. At the same time, the actual soil water content is measured at the observation station in the monitoring area (for high vegetation coverage, the actual leaf water content should be measured). Linear regression is performed between the measured data and the monitoring index value of the corresponding location, and the functional relationship between the index and water content is established, so as to obtain the distribution of soil water content (or leaf water content) in the entire monitoring area. Combining specific indicators to delineate the drought level.

Embodiment 2, utilize the drought monitoring model of embodiment 1 to monitor farmland drought

1. Take the area with a single soil type as a monitoring area.

2. Select several points in the monitoring area to measure soil water content, field water capacity and wilting coefficient. It can be considered that the field water capacity and wilting coefficient in a single soil type area are consistent, so the two measured values at each measuring point can be averaged to obtain the field water capacity and wilting coefficient of the entire monitoring area.

3. According to the vegetation coverage of the monitoring area, select the appropriate band to calculate the drought monitoring index:

Perform PDI calculations using red and near-infrared bands.

Using albedo and normalized difference vegetation index data, the calculation of VCADI is carried out.

Use NIR and SWIR bands for SPSI, VCWI, VCSI calculations.

4. Carry out correlation analysis between the soil water content (or leaf) at the measuring point and the drought monitoring index, and obtain the numerical relationship (generally linear relationship) between the soil (or leaf) water content and the monitoring index in the monitoring area.

5. From the result obtained in 4, the soil (or leaf) moisture content in the entire monitoring area is obtained.

6. Drought assessment based on soil moisture content:

Soil moisture content > 75% of field water holding capacity Moist;

The soil moisture content is between 75% and 45% of the field water capacity under normal moisture conditions;

Soil water content is less than 45% of the field water holding capacity, greater than the wilting coefficient of slight drought;

Soil water content is less than wilting adjustment coefficient Severe drought.

The application demonstration areas are Ningxia Hui Autonomous Region and Beijing Shunyi Remote Sensing Experimental Field.

Ningxia Hui Autonomous Region is located in the hinterland of the northwest of the motherland, in the upper middle reaches of the Yellow River, between 104°10′E-107°30′E and 35°25′N-39°25′N, with a land area of 51,800 km ² . Gansu, Inner Mongolia, Shaanxi and other provinces (regions) are adjacent. Drought in this area severely restricts agricultural production, and severe drought occurs almost every year. Field observation data come from 25 standard meteorological observation stations and ecological monitoring stations in the area. All the data are determined according to the "Agricultural Meteorological Observation Specifications" compiled by the National Meteorological Administration in 1993, mainly including:

1) Farmland moisture parameters

The soil moisture at different depths of 0-20cm in the surface layer on the 8th, 18th and 28th of each month from 1990 to 2005; the groundwater level in the same period; soil physiological parameters including wilting coefficient and field water capacity.

2) Crop characteristic parameters, including water demand, air temperature demand, and soil moisture requirements of wheat in different phenological stages;

3) Synchronize remote sensing data MODIS.

Beijing Shunyi Test Site is located between E 116°26’-117°E and N 40°-40°21’. From the end of March to mid-June 2001, the Shunyi test site conducted a comprehensive synchronous test of satellite-machine-ground remote sensing. Taking Shunyi, Beijing as the airborne data acquisition test site and key ground test site, the airborne multi-angle and multi-spectral data were obtained. and hyperspectral data. The observed content includes multi-angle, multi-temporal, multi-platform, multi-band, multi-scale satellite remote sensing images and aerial remote sensing images during the growth of winter wheat, various meteorological data and atmospheric parameters, various biochemical and physical and chemical parameters during the growth of winter wheat, biological Quantities and various structural parameters, spectral data and farmland microclimate parameters, etc. The data of the Shunyi test site used in this application include soil moisture, albedo, water content and chlorophyll content of wheat leaves, leaf area index, surface temperature, vegetation canopy temperature, near-surface air temperature and other data. All the data are determined according to the "Agricultural Meteorological Observation Specifications" compiled by the National Meteorological Administration in 1993.

1. Using PDI for drought monitoring:

This verification uses MODIS remote sensing data and corresponding soil moisture data in Ningxia Autonomous Region.

The key to drought monitoring lies in the acquisition and quantification of drought information, the classification and grading of drought types, and the finding of comprehensive indicators for verifying drought indices. The growth of various crops has a certain range of suitable soil moisture. When the soil moisture content is lower than the lower limit of the suitable soil moisture content of the crop (generally speaking, the lower limit of the suitable soil moisture content of the crop is about 20% of the field water holding capacity, and the upper limit When the crop does not absorb enough water to compensate for the expenditure of transpiration, drought occurs. However, crop absorption rate is closely related to soil field water capacity, crop wilting coefficient and soil effective moisture content. In the remote sensing monitoring of farmland drought conditions, the comprehensive drought index combined with soil moisture and farmland water balance can reflect the surface drought conditions more comprehensively. Therefore, the comprehensive drought index K (the following formula) is used to verify the PDI.

$\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; W</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}}{{\mathrm{<span\; class="notranslate">\; W</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; W</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; )</span>$

In the formula, K is the drought index; W is the soil water content (%); W _h is the upper limit of the soil can hold water, that is, the field water capacity (%); W _p is the soil wilting coefficient (%). It can be seen that the change of K can truly reflect the satisfaction degree of soil moisture to crop water demand, and reflect the drought condition. Crops can get sufficient water supply and no drought occurs, then K=0; when W=W _p , soil available water is completely lost, and crops die due to lack of water supply, then K=1.

Through the overlay analysis of the drought index and the cloud detection results of MODIS remote sensing images, it was found that in the MODIS remote sensing image data of each time phase corresponding to the field observation data set to be compared, some pixels covered by clouds were first processed. Find the cloud-free pixel around the pixel, and replace the drought index value of the cloud-covered pixel with the average value of the four surrounding pixels. If the surrounding pixels are all covered by clouds, the pixel is eliminated.

Taking the Ningxia political area between 104°10′E-107°30′E and 35°25′N-39°25′N as the monitoring area, the vegetation coverage of this monitoring area on March 26, 2005 was less than < 15%. Using the MODIS Ningxia range data on March 26, 2005, the PDI of the monitoring area was calculated and compared with the 0-20cm average soil moisture content of 22 meteorological and ecological observation stations in Ningxia. The soil moisture content used is the field observation data on March 28.

The monitoring area was divided into 22 observation points, and the average soil moisture content (W _0-20 ) of the surface layer 0-20cm of each sample point was measured by CNC-503DR intelligent neutron moisture meter.

Calculate the PDI of the 22 observation points respectively.

Vertical Drought Index (Perpendicular Drought Index, PDI).

$\mathrm{<span\; class="notranslate">\; PDI</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; red</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; MR</span>}}_{\mathrm{<span\; class="notranslate">\; nir</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Among them, M is the slope of the soil line, R _red is the reflectance of the red band (630-690nm) after atmospheric correction, and R _nir is the reflectance of the near-infrared band (775-900nm) after atmospheric correction.

Table 1. Soil moisture content and vertical drought index PDI at each observation point in Ningxia on March 28, 2005

longitude latitude PDI 0-20cm soil water content (%) Red Nir Yinchuan 106°18′ 38°25′ 0.267111 24.5 0.169 0.207 Qingtong Gorge 105°54′ 38°01′ 0.313107 3 0.2 0.241 Dawukou 106°24′ 39°02′ 0.217588 25.4 0.143 0.164 Guyuan 106°16′ 36°00′ 0.302125 13.8 0.189 0.236 106°18′ 35°58′ 0.258194 16 0.175 0.19 Huinong 106°46′ 39°13′ 0.213155 23.8 0.142 0.159 Kaihara 105°39′ 36°34′ 0.266456 8.7 0.168 0.207 105°32′ 36°42′ 0.299505 5.2 0.185 0.36 Sigi 105°43′ 35°58′ 0.289883 16.4 0.183 0.225 Jingyuan 106°19′ 35°30′ 0.226153 19.2 0.133 0.184 106°21′ 35°30′ 0.230939 15.7 0.138 0.186 Lund 106°07′ 35°37′ 0.251091 22.9 0.148 0.204 Tao Le 106°42′ 38°48′ 0.258647 27.9 0.163 0.201 Pin Luo 106°34′ 38°54′ 0.212802 23.7 0.138 0.162 Wu Zhong 106°08′ 37°47′ 0.290437 4 0.185 0.224 Xingren 105°15′ 36°56′ 0.343133 5.4 0.217 0.266 Yongning 106°15′ 38°15′ 0.270789 24.1 0.17 0.211 Zhongning 105°39′ 37°26′ 0.227009 twenty one 0.147 0.173 salt pond 107°23′ 37°48′ 0.311647 4.6 0.192 0.246 Lingwu 106°29′ 38°05′ 0.308976 2.6 0.196 0.239 central defender 105°13′ 37°20′ 0.265046 28.4 0.167 0.206 concentric 105°54′ 36°58′ 0.333914 6.7 0.211 0.259

According to the PDI of 22 observation points in the monitoring area and the W _0-20 of the 22 observation points in the monitoring area, the functional relationship is established with the data processing software Excell, and the functional relationship is as follows: W _0-20 =-0.0033PDI+0.3215, R ² =0.5596( Figure 5).

Vertical drought index and soil moisture data (W _0-20 ) showed a high correlation, which may be related to vegetation growth. The maximum value of the vegetation index (NDVI) was 0.34 on March 28. The crops had just started to rise, the vegetation coverage was low, and the vegetation had little interference with the surface reflectance. The reflectance spectrum was mainly determined by the physical and chemical properties of the soil.

2. Use the albedo drought index VCADI for drought monitoring:

Taking the Beijing Shunyi Test Site (the scope as above) as the monitoring area, the vegetation coverage in this monitoring area is greater than 15% and less than 65%.

The monitoring area is divided into 24 observation points, and the CNC-503DR intelligent neutron moisture meter is used to measure the soil moisture content of 5cm, 10cm, 20cm, and 0-20cm in the surface layer of each sample point. The data in table 2 is the surface layer 5cm, 10cm, 20cm, 0-20cm average soil water content (W ₅ , W ₁₀ , W ₂₀ , W _0-20 ) of 24 observation points on April 17, 2001, and soil field Water holding capacity w _h , soil wilting coefficient w _p .

Table 2. Soil moisture content at each observation point in Beijing on April 17, 2001

Observation point number Longitude latitude W ₅ (%) W ₁₀ (%) W ₂₀ (%) W _0-20 (%) W _h (%) W _p (%) 25 116°34′33.4″ 40°11′41.2″ 6.510 12.699 17.067 12.09178 22.5 10.4 26 116°34′34.3″ 40°11′43.6″ 3.406 10.200 16.678 10.09486 22.5 10.4 27 116°34′34.0″ 40°11′46.0″ 4.982 13.298 21.927 13.40233 22.5 10.4 28 116°34′33.7″ 40°11′48.5″ 7.445 15.554 22.681 15.22647 22.5 10.4 29 116°34′37.2″ 40°11′49.0″ 4.909 13.578 22.177 13.55472 22.5 10.4 30 116°34′37.5″ 40°11′46.6″ 9.881 16.839 22.507 16.40919 22.5 10.4 31 116°34′37.9″ 40°11′44.1″ 4.842 10.992 16.343 10.72571 22.5 10.4 32 116°34′38.1″ 40°11′41.8″ 6.045 12.283 17.249 11.85891 22.5 10.4 33 116°34′42.4″ 40°11′41.6″ 6.890 11.570 16.211 11.55713 22.5 10.4 34 116°34′42.1″ 40°11′44.0″ 3.579 8.061 13.665 8.434948 22.5 10.4 35 116°34′41.8″ 40°11′46.5″ 7.308 12.786 20.920 13.6712 22.5 10.4 36 116°34′41.6″ 40°11′49.0″ 8.520 14.292 20.990 14.60063 22.5 10.4 46 116°34′24.8″ 40°11′34.1″ 4.411 10.309 16.410 10.37664 22.1 10 47 116°34′25.2″ 40°11′30.1″ 7.624 14.026 17.082 12.91068 22.1 10 48 116°34′31.1″ 40°11′30.4″ 6.828 13.522 17.083 12.4777 22.1 10 49 116°34′30.8″ 40°11′34.4″ 6.380 13.256 17.005 12.21337 22.1 10 52 116°34′34.4″ 40°11′34.7″ 6.228 11.834 16.429 11.49704 22.1 10 53 116°34′34.8″ 40°11′30.6″ 8.114 14.146 17.094 13.11814 22.1 10 54 116°34′31.7″ 40°11′53.6″ 16.274 22.079 24.440 20.931 22.7 10.4 55 116°34′31.2″ 40°11′56.9″ 13.599 20.024 23.542 19.05479 22.7 10.4 58 116°34′37.1″ 40°11′57.6″ 5.278 12.650 20.097 12.67541 22.7 10.4 59 116°34′37.5″ 40°11′54.6″ 3.974 11.429 19.492 11.6315 22.7 10.4 60 116°34′42.8″ 40°11′54.6″ 13.101 19.949 23.173 18.74107 22.7 10.4 61 116°34′42.4″ 40°11′58.0″ 17.926 22.195 23.777 21.2994 22.7 10.4

The albedo drought index is calculated according to the following formula (II):

$\mathrm{<span\; class="notranslate">\; VCADI</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; A</span>}}_{{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{{\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}}{{\mathrm{<span\; class="notranslate">\; A</span>}}_{{\mathrm{<span\; class="notranslate">\; max</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDVI</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; NDV</span>}{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; II</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

Among them, VCADI is the albedo drought index, $A_{{\max ,\mathrm{NDVI}}_i}=a+\mathrm{bNDV}{I}_i,$ $A_{{\min ,\mathrm{NDVI}}_i}=a^{\&prime;}+b^{\&prime;}{\mathrm{NDVI}}_i;$ A _{min, NDVIi} and A _{max, NDVIi} are the minimum and maximum albedo when the vegetation index of the area to be measured is equal to a certain value; The scatter diagram (see attached drawing 3) is obtained, a and a' are the intercepts of the dry edge and the wet edge in the Albedo-NDVI feature space respectively, b and b' are the dry edge and the wet edge in the Albedo-NDVI feature space The slope of ; A _{i , NDVIi} , is the albedo of the point (A _i , NDVI _i ) in the Albedo-NDVI feature space; NDVI _i is the vegetation index of D(A _i , NDVI _i ) in the Albedo-NDVI feature space;

Using the albedo-NDVI feature space constructed from the ETM+ remote sensing data of Beijing Shunyi research area on April 17, 2001, on the basis of extracting dry and wet edges, the relevant parameters of VCADI are further determined through statistical analysis. VCADI was calculated using near-infrared albedo and corresponding NDVI data.

According to the VCADI of 24 observation points in the monitoring area and the 24 observation points W ₅ , W ₁₀ , W ₂₀ , W _0-20 in the monitoring area, use the data processing software Excel to establish the functional relationship, among which, use the near-infrared band (0.7-4.0μm) The functional relationship between VCADI calculated by albedo and soil water content is as follows:

The functional relationship between VCADI and W ₅ is W ₅ =-9.3465Ln(VCADI)+1.4711, R ² =0.8057;

The functional relationship between VCADI and W ₁₀ is W ₁₀ =-22.019*VCADI+26.032, R ² =0.7962;

The functional relationship between VCADI and W ₂₀ is W ₂₀ =-16.968*VCADI+28.556, R ² =0.6585;

The functional relationship between VCADI and W _0-20 is W _0-20 =-20.578*VCADI+24.873, R ² =0.8363.

The results showed that the VCADI calculated from the near-infrared albedo showed a good negative correlation with soil moisture, especially the relationship between the 0-20cm average soil moisture and the surface 10cm soil moisture was the best, and there was a linear relationship (Figure 6). The squares of the correlation coefficients with the 0-20cm average soil moisture (W _0-20 ) are R ² =0.84, 0.83, respectively. Figure 6 is a graph of VCADI and soil water content calculated using the albedo in the near-infrared band (0.7-4.0 _μm ). good. In Figure 6, W ₅ , W ₁₀ , W ₂₀ , and W _0-20 represent the average soil moisture in 5cm, 10cm, 20cm, and 0-20cm, respectively, and R2 is the square of the correlation coefficient.

3. Using SPSI and VWCI (VWSI) for drought monitoring:

Taking Beijing Shunyi Test Site (the scope as above) as the monitoring area, the vegetation coverage in this monitoring area on May 19, 2001 was greater than 65%.

In order to verify SPSI and VWCI, using the TM/ETM+ image of Beijing Shunyi test area on May 19, 2001, the leaf equivalent water content (EWT) and leaf relative water content (FMC) calculated by SPSI and VWCI were synchronized with satellite compared with field observation data.

The leaf water content per unit area (pixel scale) is related to the leaf area index. With the increase of the leaf area index, the EWT per unit area will increase. The pixel reflects the average moisture content at the pixel scale. Therefore, after multiplying the EWT observed in the field by the leaf area index (such as EWT _canopy = LAI × EWT _leaf ), it can be compared with the vegetation moisture content obtained from remote sensing data. FMC is the average value of the sampling area, and the observation point data can be directly compared with the FMC value of the remote sensing pixel scale.

The monitoring area on May 19, 2001 was divided into 13 observation points. The specific values of longitude, latitude, LAI, calculated VWCI, SPSI and EWT _canopy of each point are shown in Table 3:

Table 3. Parameter values of vegetation moisture status at various observation points in Beijing on May 19, 2001

sampling number longitude latitude VWCI SPSI LAI EWT _canopy FWC NW1-1 40°11′42.1″ 116°34′01.1″ 0.862698 0.473955 1.607 242.1298 372.32 NW1-2 40°11′42.6″ 116°34′04.6″ 0.928463 0.465037 2.483 250.3996 249.52 NW2-1 40°11′42.1″ 116°34′16.2″ 0.805081 0.482557 1.487 187.6734 311.62 NW2-2 40°11′47.7″ 116°34′19.1″ 0.70719 0.527388 1.489 173.2655 287.33 NW3-1 40°11′34.1″ 116°34′24.8″ 0.835301 0.485952 1.862 153.6023 203.70 NW3-2 40°11′34.4″ 116°34′30.8″ 0.704406 0.479479 1.357 124.1287 225.81 NW3-3 40°11′34.7″ 116°34′34.4″ 0.915049 0.467324 2.203 212.0869 238.05 NW4-26 40°11′43.6″ 116°34′34.3″ 0.859784 0.467799 1.780 186.3046 258.42 NW4-31 40°11′44.1″ 116°34′37.9″ 0.870286 0.455644 2.020 191.3939 233.67 NW4-34 40°11′44.0″ 116°34′42.1″ 0.907015 0.461642 2.067 239.0386 284.94 NW5-55 40°11′56.9″ 116°34′31.2″ 0.645697 0.500237 0.735 72.36716 243.53 NW5-58 40°11′57.6″ 116°34′37.1″ 0.75568 0.458563 1.321 135.1022 251.92 NW5-61 40°11′58.0″ 116°34′42.4″ 0.99157 0.461325 2.624 330.1371 310.83

The short-wave infrared vertical water loss index is calculated according to the following formula (III):

$\mathrm{<span\; class="notranslate">\; SPSI</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; SWIR</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; MR</span>}}_{\mathrm{<span\; class="notranslate">\; NIR</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; III</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

Among them, SPSI is the vertical short-wave infrared dehydration index; R _NIR is the reflectivity of the 780-900nm near-infrared band after atmospheric correction, and the R _SWIR is the reflectance of the short-wave infrared band of 1550-1750nm after atmospheric correction; M is NIR-SWIR Baseline slope.

The vegetation moisture content index is calculated according to the following formula (IV):

$\mathrm{<span\; class="notranslate">\; VWCI</span>}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; NIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; SWIR</span>}}_{\mathrm{<span\; class="notranslate">\; G</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; IV</span>}<span\; class="notranslate">\; )</span>$

Among them, VWCI is the vegetation water content index; M is the slope of soil baseline in NIR-SWIR space; as shown in Figure 4, M ₁ and M ₂ are the slopes of AB and CD lines, and I ₁ and I ₂ are the slopes of AB and CD lines, respectively. Intercept; NIR _G and SWIR _G are the albedo values of the near-infrared and short-wave infrared corresponding to the position to be measured.

According to the SPSI and VWCI of each observation point in the monitoring area and the EWT _canopy and FMC of each observation point in the monitoring area, the functional relationship is established with the data processing software Excell, where

The functional relationship between SPSI and EWT _canopy is as follows:

SPSI = 1.0986*EWT _canopy + 16.213, ^R2 = 0.6864;

The functional relationship between VWCI and EWT _canopy is as follows:

VWCI=0.9871*EWT _canopy +9.1872, ^R2 =0.7365;

The functional relationship between VWCI and FMC is as follows:

VWCI=0.7525*FMC+64.449, ^R2 =0.7053;

Using R ² and root mean square error (RMSE), the precision analysis of remote sensing observation data was carried out. The results show that there is a high correlation between model estimates and field observation data (see Figures 7, 8, and 9).

It can be seen from the analysis that VWCI (VWSI) and SPSI are effective in monitoring vegetation moisture, and the monitoring accuracy of VWCI (VWSI) is higher than that of SPSI.

Claims (6)

1. the method for a face of land draught monitor, be soil or the leaf water content that obtains set observation station on the face of land to be monitored, the soil of described observation station or the corresponding index of leaf water content and sign face of land damage caused by a drought are set up functional relation, bring remotely-sensed data into described functional relation, draw the soil or the plant leaf blade water cut on the face of land to be monitored;

Wherein, the corresponding index of described sign face of land damage caused by a drought is determined as follows:

1) on the face of land, farmland of vegetation coverage≤15%, adopts vertical drought index;

Described vertical drought index calculates according to following formula (I):

$\mathrm{PDI}=\frac{1}{\sqrt{M^2+1}}(R_{\mathrm{red}}+{\mathrm{MR}}_{\mathrm{nir}})---\left(I\right),$

Wherein, PDI is vertical drought index, and M is the soil line slope, R _RedBe the red spectral band reflectivity through atmospheric correction, R _NirBe near-infrared band reflectivity through atmospheric correction;

2) vegetation coverage＞15% and≤65% the face of land, farmland, adopt vegetation condition albedo drought index;

Described vegetation condition albedo drought index calculates according to following formula (II):

$\mathrm{VCADI}=\frac{A_{i,\mathrm{NDV}{I}_i}-A_{{\min ,\mathrm{NDVI}}_i}}{A_{\max ,{\mathrm{NDVI}}_i}-A_{{\min ,\mathrm{NDVI}}_i}}---\left(\mathrm{II}\right),$

Wherein, VCADI is a vegetation condition albedo drought index, $A_{\max ,\mathrm{NDV}{I}_i}=a+\mathrm{bNDV}{I}_i,$ $A_{\min ,\mathrm{NDV}{I}_i}=a^{\&prime;}+b^{\&prime;}{\mathrm{NDVI}}_i;$ A _{Min, NDVIi}And A _{Max, NDVIi}Be respectively minimum and the maximum albedo of district to be measured vegetation index when equaling certain particular value; A, b, a ', b ' is a undetermined coefficient, and a and a ' are respectively the intercepts of doing limit and wet limit in the Albedo-NDVI feature space, and b and b ' are respectively the slopes of doing limit and wet limit in the Albedo-NDVI feature space, scatter diagram by district to be measured vegetation index and albedo obtains A _{I, NDVIi}Be Albedo-NDVI feature space mid point (A _i, NDVI _i) albedo, NDVL is D (A in the Albedo-NDVI feature space _i, NDVI _i) vegetation index;

3), adopt at least a in following three kinds of indexes: the vertical dehydration index of short-wave infrared, vegetation moisture exponential sum vegetation water deficit index on the face of land, farmland of vegetation coverage＞65% to 100%;

The vertical dehydration index of described short-wave infrared calculates according to following formula (III):

$\mathrm{SPSI}=\frac{1}{\sqrt{M^2+1}}(R_{\mathrm{SWIR}}+{\mathrm{MR}}_{\mathrm{NIR}})---\left(\mathrm{III}\right)\text{,}$

Wherein, SPS worker is the vertical dehydration index of short-wave infrared, R _SWIR, R _NIRBe respectively short-wave infrared and near-infrared band reflectivity through atmospheric correction, M is a soil baseline slope in the empty palace gate of NIR-SWIR;

Described vegetation moisture index calculates according to formula (IV):

$\mathrm{VWCI}=\frac{(M_1-M)\&times;({\mathrm{NIR}}_G-M_2\&times;{\mathrm{SWIR}}_G-I_2)}{(M_1-M_2)\&times;({\mathrm{NIR}}_G-M\&times;{\mathrm{SWIR}}_G)-(M_1-M)\&times;I_2+(M_2-M)\&times;I_1}---\left(\mathrm{IV}\right)$

Wherein, VWCI is a vegetation moisture index; M is a soil baseline slope in the NIR-SWIR space; As Fig. 4, M ₁, M ₂Be respectively the slope of AB and CD line, I ₁, I ₂Be respectively the intercept of AB and CD line; NIR _G, SWIR _GAlbedo value for the near infrared and the short-wave infrared of position to be measured correspondence;

Described vegetation water deficit index calculates according to following formula (V):

VWSI＝1-VWCI (V)

Wherein, VWSI is a vegetation water deficit index; VWCI is a vegetation moisture index, calculates according to formula (IV).

2. method according to claim 1, it is characterized in that: described 1), the face of land, farmland at vegetation coverage≤15%, adopt vertical arid indices P DI, the M value that its computing formula relates to, in ruddiness-near-infrared band reflectivity two-dimensional space, the slope of the soil line that the each point of sign soil constitutes.

3. method according to claim 1 is characterized in that: described 2), vegetation coverage＞15% and≤65% the face of land, farmland, adopt vegetation condition albedo drought index VCADI; Wherein, extract slope and the intercept of doing limit, wet limit, calculate VCADI on this basis at the Albedo-NDVI feature space.

4. method according to claim 1 is characterized in that: described 3), on the farmland face of land of vegetation coverage＞65% to 100%, adopt vertical dehydration index of short-wave infrared or vegetation moisture index; Wherein the related parameter of each formula is all based on short-wave infrared-near infrared spectrum feature space.

5. method according to claim 1 is characterized in that: described R _RedBe the 630-690nm red spectral band reflectivity through atmospheric correction, described R _NirBe the 780-900nm near-infrared band reflectivity through atmospheric correction, described R _SWIRBe 1550-1750nm short-wave infrared wave band reflectivity through atmospheric correction.

6. according to arbitrary described method in the claim 1 to 5, it is characterized in that: the described face of land is the face of land, farmland.

Images