Monitoring and Forecast of Sandstorm

In recent decades, the study of sandstorms in China has been carried out in the field of geology, mainly from the perspective of Quaternary geology to study loess and dust deposition.

In 1950s, dust weather began to be included in the weather phenomenon that must be monitored in China's meteorological routine observation. At that time, the network of fixed-point observation stations on the ground was used.

Ground fixed-point monitoring is divided into centralized observation and long-term observation. The centralized observation of dust disaster mainly includes: optical thickness measurement of dust, meteorological elements recorded by automatic weather stations, soil condition analysis of dust source areas, vertical stratification of dust observed by lidar, dust flux observed along airflow direction, vertical observation of dust flux, visibility (naked eye, visibility meter), Doppler acoustic detector, sampling and analysis of physical and chemical characteristics of dust particles, etc. Long-term monitoring includes: automatic weather station meteorological elements observation, atmospheric aerosol optical thickness monitoring, sky radiometer radiation observation records, dust flux estimation, visibility observation, historical climate data collation and analysis, soil observation research, sandstorm historical database establishment, etc. Centralized observation can get the measured data in the first time, and long-term observation can accumulate a lot of environmental background data. Due to the large scale, dust disasters may occur from local, regional to continental. However, in the northwest of China, there are few monitoring stations, especially the formation of sandstorms is mainly in desert areas with extremely sparse population, and there are even fewer monitoring stations. This makes the conventional ground observation data have great limitations on the study of sandstorm monitoring. At the same time, the time distribution of conventional data is also difficult to capture and track the source, dynamic evolution and intensity change of strong sandstorm caused by mesoscale system.

Since 1970s, remote sensing technology has developed rapidly, and a large number of new remote sensing data with high spatial resolution, high temporal resolution and high spectral resolution have been produced, which makes it possible to monitor sand-dust disasters by using remote sensing satellites. Remote sensing data is multi-source, dynamic, real-time and accurate, which not only makes up for the lack of temporal and spatial resolution of ground observation data, but also verifies the accuracy and complements each other with ground observation data, deepening the study of sand and dust disasters.

7.4.6. 1 principle of satellite remote sensing monitoring

The physical basis of monitoring sandstorm by satellite remote sensing is the difference of spectral characteristics of observation objects.

The spectral characteristics of sandstorm, cloud, snow, sand, vegetation, water and bare land are different, and the spectral channels of polar-orbiting meteorological satellites can be roughly divided into two categories: one is located in the visible light band, which can receive the reflectivity from the target and measure the reflectivity of the underlying surface; The other is located in the infrared window band, which can receive the thermal radiation of the target. Because there are differences in albedo and surface temperature between the top of sandstorm and observation objects such as the surface and clouds, meteorological satellites can be used to monitor sandstorms.

In the processing of observation data, the characteristics of dust particles, such as particle size, shape and texture, are also important factors that determine the light radiation characteristics of dust particles, such as emission and scattering characteristics.

Selection of 7.4.6.2 Satellite

Monitoring accuracy and monitoring frequency are the important basis for us to monitor sandstorms and choose satellites. For this specific sandstorm-prone area in northwest China, the temporal and spatial scale of sandstorms is mainly restricted by the weather system. Its time scale ranges from tens of minutes to several days, sometimes even as high as ten days; The spatial scale is from n× 10 m to n× 100km, so the monitoring tools are required to have high temporal and spatial resolution, and the monitoring ability of the monitoring tools is required to reach the level of identifying sandstorm weather. At present, China's public reception and operational use of civil satellites include: American Landsat, American geostationary meteorological satellite (STOP satellite), Japanese geostationary meteorological satellite (GMS), American polar-orbiting meteorological satellite (NO AA) and FY 1C polar-orbiting meteorological satellite developed by China. The horizontal resolution of the first two satellites is very high, and the point below the satellites is tens of meters. However, the observation period of a certain area is divided into 16d and 26d, which is expensive and inconvenient for real-time monitoring of sandstorms. GMS is located at east longitude 127 above the equator, which can observe the earth once every hour and can be used for real-time extraction and monitoring of sandstorm information. But the resolution of the point under the satellite on the infrared channel is 5 km. For the northwest region, the distortion of the northwest corner data of the image is very serious. However, polar-orbiting meteorological satellites NOAA and FY 1C have moderate horizontal resolution and time resolution, and the 5 ~ 10 band set by spaceborne scanning radiometer has certain detection ability for sandstorms. The cost is low, the meteorological department has established corresponding receiving stations, and it has great prospects to monitor sandstorm weather with polar-orbiting meteorological satellites. However, for the sandstorm process with short duration, the time resolution of 6 h may miss the effective time of monitoring.

Spectral characteristics of satellite observations in 7.4.6.3

NOAA satellite is a polar orbiting satellite composed of two stars. The average orbital height is 850 km, the resolution of satellite points is 1. 1 km, and the scanning width is about 2800km；. . The space-borne detector is an improved very high resolution scanning radiometer (AVHRR) with five observation channels: the wavelength of channel 1(ch 1) is 0.58 ~ 0.68 micron, which belongs to the visible light band; The wavelength of channel 2(ch2) is 0.73 ~ 1. 1 micron, which belongs to the near infrared band. The wavelength of channel 3(ch3) is 3.55 ~ 3.93 microns, which belongs to the mid-infrared band. The wavelength of channel 4(ch4) is 10.3 ~ 1 1.3 micron, which belongs to thermal infrared band. The wavelength of channel 5(ch5) is11.5 ~12.5 μ m, which belongs to the thermal infrared band.

The polar-orbiting satellite FY- 1C launched by China mainly carries two scanning radiometers with 10 channels, and the wavelength coverage of the channels is equivalent to that of NOAA satellites.

These bands contain abundant atmospheric and surface information, which makes it possible to identify clouds, snow, sandstorms and various surfaces. Different detection surfaces have different detection values for each channel. According to their respective spectral characteristics, the sandstorm information is extracted and the monitoring model is established.

7.4.6.4 information extraction

To identify sandstorms, information such as clouds, snow, sand, vegetation, water and bare land must be identified, and based on the spectral response curve, the following parameters are used to screen and distinguish, so as to extract sandstorm information.

(1) vegetation index

Vegetation index is an important basis for monitoring vegetation and non-vegetation areas by remote sensing. NDVI is used to distinguish vegetation. The vegetation of NDVI is greater than 0, while the bare land, dusty area, cloud water and snow cover of NDVI are less than 0.

Introduction to environmental geophysics

Where ρch 1 and ρch2 are the reflectivity values of channel 1 and 2, respectively.

(2) the extraction value t of water body

The reflectivity (ρch2) value of channel 2 reflects the obvious land-water boundary, so ρ CH2 < t is adopted as the method to distinguish water areas, generally T=5.

(3) Extracting the water vapor index (WI) of cloud snow, sand dust and sand dust.

There is a pure rotation zone of water vapor in the far infrared band > 1 1μ m, but in AVHRR data, the bands of channel 4(ch4) and channel 5(ch5) are just between the rotation zones, that is, the difference between ch4 and ch5 can reflect the water content. The water content of clouds in the atmosphere and snow on the ground is much higher than that in dust or sandstorm areas, and the water vapor index of dust or sandstorm is less than zero. Therefore, the water vapor index (WI) is used to distinguish clear deserts or sandstorms from clouds and snow.

Introduction to environmental geophysics

Where ρch4 and ρch5w are the radiation values of channel 4 and channel 5, respectively.

(4) the difference between land and sandstorm

The method of combining the reflectivity value of 1 2 channel with the WI index value is used to distinguish dust storms from sandstorms. In clear and cloudless remote sensing data, the reflectivity of sandstorm is greater than that of sand dust, and the WI index is WI (sandstorm) < Wi (sand dust).

7.4.6.5 data interpretation

The main tasks of remote sensing monitoring data interpretation are: to identify and locate the spatial distribution range and influence area of sand and dust disasters; Dynamically monitor the changing process of dust migration path and migration law; Quantitative extraction of dust information by remote sensing: monitoring of background conditions such as atmosphere and underlying surface caused by dust disaster; Dynamic simulation of dust disaster.

In order to analyze satellite remote sensing data, it is necessary to obtain ground measurement data. The ground investigation information mainly includes: ① using spectrometer to measure the spectral reflectivity of ground objects with stable reflection value, such as desertified land and desert; (2) Spectral reflectance values of lake and sea surface; (3) measuring solar spectral radiation with a solar photometer; (4) pocket thermal infrared radiometer for measuring surface temperature; ⑤ Portable infrared radiometer measures the sky temperature; ⑥ Near-surface temperature and humidity; ⑦ Panoramic photos of scenery, etc. Therefore, in the study of dust disaster, we should pay attention to the research of ground experimental remote sensing. Through the measurement of the spectral characteristics of desertification land on the ground, the research of remote sensing monitoring of dust disaster began to develop from qualitative and semi-quantitative research to quantitative extraction of dust attribute characteristic parameters. Ground-based remote sensing not only deepens the satellite remote sensing theory and data inversion method, but also perfects the remote sensing image processing technology, greatly expands the application field of satellite data, and plays an important role in sand and dust prediction, sand and dust monitoring and dust cause research.

Sandstorm forecast in 7.4.6.6

The operational system of national sandstorm monitoring and early warning service organized by China Meteorological Bureau was put into trial operation on March 1 2006. On the same day, the Central Meteorological Observatory incorporated sandstorm forecast and strong sandstorm warning into the daily weather forecast business scope. Like weather forecast, people all over the country can learn about sandstorms from TV, radio, internet and other media, prevent them as early as possible and minimize disaster losses.

In February 2002, China Meteorological Bureau started the first phase of the national sandstorm monitoring and early warning service system, deployed instruments and equipment in meteorological stations in Xinjiang, Gansu, Inner Mongolia, Ningxia, Shaanxi, Beijing and other places, established a comprehensive monitoring network of sandstorm weather with a high degree of automation, increased the special observation items and observation density of sandstorm weather, and obtained the data of sandstorm formation, movement, distribution and related environmental changes at the first time, which improved the accuracy and timeliness of sandstorm weather forecast.