correlative factor

Vegetation index (VI) is a method to monitor the state of plants on the surface by means of remote sensing. Through a lot of research on plant reflectance spectrum, it is found that plant spectral information has obvious absorption band in red light band and strong reflection band in near infrared band. Therefore, we can use this characteristic of plants to design vegetation index through joint calculation of reflectance in red light band and near infrared band and carry out basic research on vegetation ecology. After years of development, the vegetation index can be divided into many types according to different monitoring and calculation methods, such as normalized vegetation index (NDVI), specific vegetation index (RVI), soil-adjusted vegetation index (SAVI) and vertical vegetation index (PVI). At present, the normalized vegetation index (NDVI) is one of the most widely used vegetation indices. The significance of each indicator is as follows:

NDVI (reflectance difference between near infrared region and red region/reflectance sum between near infrared region and red region) is expressed by the following formula: NDVI = (NIR-R)/(NIR+R).

NDVI can be used to detect vegetation growth, vegetation coverage and eliminate some radiation errors. The range of values is-1 ~ 1, and a negative value indicates that the ground is covered by clouds, water and snow. Which has high reflectivity for visible light; 0 means there are rocks or bare soil, etc. , and NIR and r are approximately equal; A positive value indicates that there is vegetation coverage, which increases with the increase of coverage. The limitation of NDVI is that nonlinear stretching enhances the contrast of reflectivity of NIR and R. For the same image, when calculating RVI and NDVI respectively, it will be found that the increase speed of RVI value is higher than that of NDVI value, that is, NDVI is less sensitive to high vegetation areas. NDVI can reflect the background influence of plant canopy, such as soil, wetland, snow, dead leaves and roughness, and it is related to vegetation coverage.

Although NDVI has the following advantages: wide space coverage; The sensitivity of plant detection is high; The data is comparable. However, these advantages can only be obtained through the following processing: eliminating the influence of the atmosphere, many components in the atmosphere, such as water vapor and ozone, will affect the reflection of red light and near infrared band, and at the same time, when the sensor receives the signal of the target, it will also receive external noise; In order to eliminate the influence of plant background soil, the sensor receives information from plant background soil at the same time, which will lead to different spectral information of the same plant cover under the influence of different backgrounds.

There are many methods to extract vegetation index. At present, the commonly used method is to extract various vegetation indexes by processing remote sensing images. Before NDVI is extracted by software, the image needs to be processed, including image preprocessing, image correction, projection conversion, format conversion and so on. Band analysis and combination, find the most suitable band collocation for research, so as to improve the image effect; Image clipping or stitching to meet the needs of research field. Then, according to NDVI formula, the difference sum between near infrared band and red band is calculated, and finally it is divided.

Second, the damaged vegetation area

When the area is large, remote sensing images, aerial photographs, geographic information system (GIS) and global positioning system (GPS) are used to measure the damaged area of vegetation. When the area is small, the damaged area of vegetation is determined by manual field measurement.

Third, precipitation.

Among all the external triggering factors that lead to landslide deformation or sharp sliding, concentrated continuous rainstorm is the most common. Rainfall-induced landslide is caused by the decrease of shear strength of landslide surface due to seepage. The influence of rainfall on small and medium-sized landslides is very obvious, which often has the characteristics of simultaneity and regionality. The research shows that when the precipitation reaches a certain critical precipitation or critical precipitation intensity, a large number of landslides will occur in an area. Therefore, in recent years, precipitation monitoring has been regarded as an important part of landslide monitoring and an effective factor of forecasting and early warning abroad.

Rain gauges should be used to monitor rainfall. There are many kinds of rain gauges, such as tipping bucket rain gauge, siphon rain gauge and so on. Taking the most widely used tipping bucket rain gauge as an example, its working principle is introduced.

The tipping bucket rain gauge is a telemetering rainfall instrument, which consists of a sensor and a signal recorder. The sensor consists of a water container, an upper tipping bucket, a metering tipping bucket, a counting tipping bucket and a spring switch. The recorder consists of a counter, a recording pen, a self-recording clock and a control circuit board. Its working principle is: rainwater enters the water container from the top nozzle, falls into the water receiving funnel, and flows into the tipping bucket through the funnel dripper. When the accumulated water reaches a certain height (e.g. 0. 1mm), the tipping bucket loses its balance and turns over. Every time the dumper topples, the switch is turned on, and a pulse signal is sent to the recorder and recorded, so that the rainfall process can be measured, as shown in Figure 5-4. In addition, all the rain gauges used now also have data acquisition modules and communication modules, which can directly transmit the monitoring values to the designated database.

Siphon rain gauge can continuously record liquid precipitation and precipitation hours, and precipitation intensity can also be known from precipitation records. Siphon rain gauge consists of water container, float chamber, self-recording clock and shell. Rainwater enters the water container from the top water inlet, is collected through the lower funnel, and leads to the float chamber. The float chamber consists of a cylinder equipped with a float, which rises with the increase of the injected rain and drives the self-recording pen to rise. The self-recording clock is fixed on the seat plate, and the drum is driven by the clock machine to rotate, so that the siphon rainfall recording pen can draw a curve on the recording paper around the drum. The ordinate on the recording paper records the rainfall, and the abscissa is driven by the self-recording clock, indicating the time. When the rainfall reaches a certain height (such as 10mm), the water level in the float chamber rises to the siphon connected with the float chamber, which leads to the start of siphon, and the rainwater in the float chamber is quickly discharged into the water storage bottle. At the same time, the self-recording pen drops vertically to the zero line position on the recording paper, and starts to rise again with the inflow of rainwater, thus continuously recording the rainfall process back and forth.

Figure 5-4 Structure diagram of tipping bucket rain gauge

Four. pore water pressure

Since Karl Terzaghi (1883 ~ 1963) put forward the theory of pore water pressure in saturated clay, this monitoring content has been widely used in landslide stability evaluation abroad. In China, it is mainly used for monitoring geological disasters, and can also be used for monitoring in foundation treatment projects. Monitor the parameters such as pore water pressure in the landslide, especially in the sliding zone, and predict the possibility and danger degree of landslide caused by rainstorm according to the changes of these parameters, so as to make advance prediction, so as to reduce or avoid huge economic losses and casualties caused by rainstorm landslide, and provide scientific basis for drainage and prevention of rainstorm landslide.

The general principle of pore water pressure monitoring is as follows: when the external temperature is constant and the osmometer is only subjected to the osmotic (pore water) pressure, the pressure value p has the following linear relationship with the output frequency modulus δ f:

p = kδF

δF = F0-F

Where: k is the minimum pressure reading measured by the osmometer (kPa/f); δf—— the change of reference value of osmometer relative to real-time measurement value (f); F- real-time measurement value of osmometer (f); F0 refers to the reference value (f) of osmometer.

When the osmotic (pore) water pressure acting on the osmometer is constant (that is, Δ p ′ = 0) and the temperature increases Δ t, the osmometer has an output Δ f ′, which is only caused by the temperature change, so it should be deducted in the calculation.

Experiments show that δ f' and δ t have the following linear relationship:

δP′= kδF′+bδT = 0

kδF′=-bδT

δT = T-T0

Where: b-temperature correction coefficient of osmometer (kPa/℃); Δ t —— the change of real-time temperature measurement value relative to reference value (℃); T refers to the real-time measured value of temperature (℃); T0 —— Reference value of temperature (℃).

When the osmometer is subjected to the dual action of osmotic (pore) water pressure and temperature, the general calculation formula of osmometer is as follows:

pm = kδF+bδT = k(F0-F)+b(T-T0)+Q

Where: PM is the measured seepage (pore) hydraulic pressure (kPa); Q—— Correction parameter. If the atmospheric pressure changes greatly, it should be corrected.