阅读说明：本技术 一种基于被动微波遥感的多年冻土监测与分类方法 (Permafrost monitoring and classifying method based on passive microwave remote sensing ) 是由 张万昌 高会然 于 2021-06-24 设计创作，主要内容包括：本发明提出一种基于被动微波遥感的多年冻土监测与分类方法,包括如下步骤：步骤1、利用判别算法获取近地表土壤冻结状态；步骤2、基于步骤1获取的近地表土壤冻结状态,建立基于近地表土壤冻结状态的冻结指数方法；步骤3、通过气象站点数据,建立基于地表土壤冻融状态的冻结指数与年平均气温之间的关系模型,获得多年冻土分类的冻结指数阈值,结合多年冻土热学稳定性分区理论进行监测与分类。(The invention provides a permafrost monitoring and classifying method based on passive microwave remote sensing, which comprises the following steps: step 1, obtaining a near-surface soil frozen state by using a discrimination algorithm; step 2, establishing a freezing index method based on the near-surface soil freezing state obtained in the step 1; and 3, establishing a relation model between the freezing index based on the freeze-thaw state of the earth surface soil and the annual average air temperature through meteorological station data, obtaining a freezing index threshold value of the classification of the perennial frozen soil, and monitoring and classifying by combining a thermal stability zoning theory of the perennial frozen soil.)

1. A permafrost monitoring and classifying method based on passive microwave remote sensing is characterized by comprising the following steps:

step 1, obtaining a near-surface soil frozen state by using a discrimination algorithm;

step 2, establishing a freezing index method based on the near-surface soil freezing state obtained in the step 1;

and 3, establishing a relation model between the freezing index based on the freeze-thaw state of the earth surface soil and the annual average air temperature through meteorological station data, obtaining a freezing index threshold value of the classification of the perennial frozen soil, and monitoring and classifying by combining a thermal stability zoning theory of the perennial frozen soil.

2. The permafrost monitoring and classifying method based on passive microwave remote sensing according to claim 1, wherein in step 1, a discrimination algorithm is used to obtain the near-surface soil frozen state, a dual-index DIA algorithm based on improved passive microwave remote sensing is used as the most widely applied discrimination algorithm for surface soil freeze-thaw remote sensing, and the dual-index DIA algorithm of improved passive microwave remote sensing requires three parameters: 37GHz vertical polarization bright temperature Tb_37v19-37GHz negative brightness temperature spectrum gradient SG and soil moisture local variance LVSM to judge the freeze-thaw state of the earth surface, and the core algorithm is formula (1) and formula (2)

Tb_37v≤P₃₇ (1)

Wherein, Tb_37vIs 36.5GHz vertical polarization bright temperature with the unit of K,shows the negative bright temperature spectrum gradient between 18.7GHz and 36.5GHz with the unit of K/GHz, P₃₇And P_SGRespectively representing two indexes Tb_37vAnda threshold value of (d); p_SGThe value is 0; and when the surface soil state meets the conditions of the formula (1) and the formula (2), judging the surface soil to be frozen soil, otherwise, judging the surface soil to be melted soil.

3. The permafrost monitoring and classifying method based on passive microwave remote sensing according to claim 1, wherein the step 2 is to establish a freezing index F based on the near-surface soil freezing state_iA process, represented by formula (4):

in the formula (4), D_f ^1/2And D_t ^1/2The days for freezing and thawing the soil in one year respectively; the annual sequence freezing index spatial distribution is obtained, the information is an index for representing the spatial distribution and the climatic sensitivity of the permafrost, and the multi-type permafrost distribution with different continuity is obtained by combining with a permafrost classification method.

4. The permafrost monitoring and classifying method based on passive microwave remote sensing according to claim 1, wherein the permafrost monitoring and classifying in step 3 is carried out by combining a permafrost thermal stability partition theory, specifically as follows,

dividing the permafrost into four sub-zones or types by taking the average annual temperature as a main basis for sub-zones:

1) the place with the temperature contour of-5.0 ℃ in the mean temperature of the year and north is a continuous permafrost zone;

2) -5.0 to-3.0 ℃ isotherms are in the discontinuous permafrost zone;

3) an island-shaped permafrost zone is arranged between the isothermal line of-3.0 ℃ and the isothermal line of 0 ℃ in the south bound of permafrost;

4) the south boundary takes the south area as a seasonal frozen soil zone;

then, establishing a relational model between the freezing index based on the freeze-thaw state of the earth surface soil and the annual average air temperature through meteorological station data to obtain a freezing index threshold value of the classification of the perennial frozen soil; the relationship model between the freezing index and the annual average air temperature is expressed as equation (5).

F_i＝a·ln(k-T_am)+b (5)

Wherein a, b and k are constants, T_amThe annual average temperature, k value according to the annual maximum annual average temperature T_am,maxDetermination of k-T_am,maxNot less than 1; therefore, corresponding freezing index thresholds under different annual average air temperature thresholds are obtained.

Technical Field

The invention relates to the field of remote sensing monitoring, in particular to a method for remotely sensing, monitoring and classifying permafrost.

Background

The permafrost is an important component factor in a freezing circle layer, and the formation and development of the permafrost are influenced and determined by various factors such as geography, geology, weather and the like. Due to the characteristics of wide permafrost area, bad development environment and the like, the spatial distribution of the permafrost is difficult to obtain through ground sampling investigation on the characteristics of the permafrost. With the accumulation of observation data and the advancement of technical means, some methods or models for estimating frozen soil distribution by using the existing observation data sets (such as surface temperature, vegetation coverage and the like) are proposed. Cold climatic conditions are essential elements of permafrost formation and development. The permafrost south bound at high latitude and the frozen soil lower bound of the permafrost at high altitude are basically consistent with the annual average temperature contour line of-1 to-2 ℃. According to the development relationship between permafrost and climate, the permafrost thermal stability zoning theories in different areas are successively proposed, and an important theoretical basis is provided for permafrost distribution monitoring and research.

The Earth observation System science plan (EOS) ranks frozen soil remote sensing as one of important research targets, and the application potential of remote sensing in frozen soil research is increasingly emphasized. In the early research of monitoring permafrost by using a remote sensing technology, a visible light multispectral satellite sensor is mostly adopted, and the visible light multispectral satellite sensor commonly comprises Landsat, MODIS (model Resolution Imaging spectrometer), ASTER (Advanced Space-borne Thermal Emission and Reflection Radiometer) and SPOT and the like.

Generally speaking, there are areas where permafrost is distributed, the annual average air temperature is usually less than 0 ℃, while the earth's surface also has seasonal permafrost distribution, i.e. a moving layer of permafrost. The matter exchange and energy transfer can be directly carried out between the permafrost active layer and the permafrost layer, and the thermal stability of the permafrost layer also influences the spatial distribution, the freezing depth and the like of the active layer. Therefore, in a certain degree, the freeze-thaw spatiotemporal characteristics of the permafrost active layer reflect the spatiotemporal distribution state of the permafrost layer. Since the 70 s of the last century, scholars at home and abroad use passive microwave remote sensing technology to conduct seasonal frozen soil space-time monitoring research in the global range, develop a series of earth surface freeze-thaw state discrimination algorithms based on passive microwave remote sensing data, and obtain remarkable results. In addition to these studies, Park et al (2016) first conducted studies to predict perennial frozen soil distribution using surface freeze-thaw status obtained by passive microwave remote sensing. The study considers that the ground surface freezing days are more than the thawing days for two consecutive years, and the ground surface freezing days are classified as permafrost.

The permafrost usually develops below the earth's surface, and in the permafrost monitoring method based on optical remote sensing, monitoring by utilizing visible light and infrared sensors has great limitation:

1) the return visit period of the visible light remote sensing satellite is long, and the visible light remote sensing satellite is easily influenced by cloud layers;

2) research on permafrost in local areas is difficult to popularize and apply, and the like;

3) although the development and the time-space evolution of the permafrost are influenced by the altitude, the dimensionality, the terrain, the climate conditions and the like, the permafrost is subjected to drawing and the like through the correlation between the distribution characteristics of the permafrost and various environmental variables, the complex physical structure, the soil environment and the space heterogeneity of the permafrost bring great uncertainty to the research of permafrost monitoring by using an optical sensor.

In a permafrost monitoring method based on passive microwave remote sensing, the conventional method utilizes the freeze-thaw state of the earth surface to predict the distribution of the permafrost. The classification result shows that the classification of the permafrost is approximately consistent with the permafrost classification chart, but the frozen soil subareas have larger difference in partial areas due to the existence of multi-type permafrost such as sporadic distribution, discontinuity and the like. The method is used as an experimental study for permafrost monitoring based on passive microwave remote sensing, has no perfect theoretical basis, and is difficult to popularize and apply.

Disclosure of Invention

In order to solve the technical problems, the invention provides a permafrost monitoring and classifying method based on passive microwave remote sensing. Then, carrying out correlation analysis on the freezing index and the annual average air temperature, and carrying out frozen soil type classification according to a thermal stability zoning theory of the permafrost soil to finally obtain the spatial distribution of the multi-type permafrost soil.

The technical scheme of the invention is as follows: a permafrost monitoring and classifying method based on passive microwave remote sensing comprises the following steps:

step 1, obtaining a near-surface soil frozen state by using a discrimination algorithm;

step 2, establishing a freezing index method based on the near-surface soil freezing state obtained in the step 1;

and 3, establishing a relation model between the freezing index based on the freeze-thaw state of the earth surface soil and the annual average air temperature through meteorological station data, obtaining a freezing index threshold value of the classification of the perennial frozen soil, and monitoring and classifying by combining a thermal stability zoning theory of the perennial frozen soil.

Further, in the step 1, a discrimination algorithm is used for obtaining the near-surface soil freezing state, an improved passive microwave remote sensing-based dual-index DIA algorithm is used as the most widely applied surface soil freezing and thawing remote sensing discrimination algorithm, and the improved passive microwave remote sensing-based dual-index DIA algorithm needs three parameters: 37GHz vertical polarization bright temperature Tb_37v19-37GHz negative brightness temperature spectrum gradient SG and soil moisture local variance LVSM to judge the freeze-thaw state of the earth surface, and the core algorithm is formula (1) and formula (2)

Tb_37v≤P₃₇ (1)

Wherein, Tb_37vIs 36.5GHz vertical polarization bright temperature with the unit of K,shows the negative bright temperature spectrum gradient between 18.7GHz and 36.5GHz with the unit of K/GHz, P₃₇And P_SGRespectively representing two indexes Tb_37vAnda threshold value of (d); p_SGThe value is 0; and when the surface soil state meets the conditions of the formula (1) and the formula (2), judging the surface soil to be frozen soil, otherwise, judging the surface soil to be melted soil.

Further, the step 2 establishes a freezing index F based on the near-surface soil freezing state_iA process, represented by formula (4):

in the formula (4), D_f ^1/2And D_t ^1/2The days for freezing and thawing the soil in one year respectively; the annual sequence freezing index spatial distribution is obtained, the information is an index for representing the spatial distribution and the climatic sensitivity of the permafrost, and the multi-type permafrost distribution with different continuity is obtained by combining with a permafrost classification method.

Further, the step 3 of monitoring and classifying by combining the permafrost thermal stability zoning theory specifically includes that the permafrost is divided into four zones or types by taking the annual average air temperature as a main basis for the zone division:

1) the place with the temperature contour of-5.0 ℃ in the mean temperature of the year and north is a continuous permafrost zone;

2) -5.0 to-3.0 ℃ isotherms are in the discontinuous permafrost zone;

3) an island-shaped permafrost zone is arranged between the isothermal line of-3.0 ℃ and the isothermal line of 0 ℃ in the south bound of permafrost;

4) the south boundary takes the south area as a seasonal frozen soil zone;

then, establishing a relational model between the freezing index based on the freeze-thaw state of the earth surface soil and the annual average air temperature through meteorological station data to obtain a freezing index threshold value of the classification of the perennial frozen soil; the relationship model between the freezing index and the annual average air temperature is expressed as equation (5).

F_i＝a·ln(k-T_am)+b (5)

Wherein a, b and k are constants, T_amThe annual average temperature, k value according to the annual maximum annual average temperature T_am,maxDetermination of k-T_am,maxNot less than 1; therefore, corresponding freezing index thresholds under different annual average air temperature thresholds are obtained.

Drawings

FIG. 1 is a flow chart of a permafrost monitoring and classification method based on passive microwave remote sensing;

FIG. 2 is a comparison of permafrost distribution and classification results with existing permafrost zoning maps;

FIG. 3(a) Classification of annual Arctic permafrost in 1997;

FIG. 3(b) permafrost profile of northern hemisphere 2003;

FIG. 3(c) permafrost profile of northern hemisphere 2010;

FIG. 3(d) 2017. northern hemisphere permafrost profile;

fig. 4 is an illustration of fig. 3.

Detailed Description

The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments, and all other embodiments obtained by a person skilled in the art based on the embodiments of the present invention belong to the protection scope of the present invention without creative efforts.

According to an embodiment of the invention, as shown in fig. 1, a flowchart of a permafrost monitoring and classifying method based on passive microwave remote sensing of the invention includes the following steps:

step 1, obtaining a near-surface soil frozen state by using a discrimination algorithm;

step 2, establishing a freezing index method based on the near-surface soil freezing state obtained in the step 1;

and 3, establishing a relation model between the freezing index based on the freeze-thaw state of the earth surface soil and the annual average air temperature through meteorological station data, obtaining a freezing index threshold value of the classification of the perennial frozen soil, and monitoring and classifying by combining a thermal stability zoning theory of the perennial frozen soil.

Firstly, an improved Double Index (DIA) algorithm is used for obtaining a near-surface soil freezing state, after precision verification is carried out on ground observation data, improvement of a freezing index method is carried out according to the characteristics of the near-surface soil freezing state data, and the freezing index method based on the near-surface soil freezing state is established. Then, carrying out correlation analysis on the freezing index and the annual average air temperature, and carrying out frozen soil type classification according to a thermal stability zoning theory of the permafrost soil to finally obtain the spatial distribution of the multi-type permafrost soil.

According to one embodiment of the invention, in the step 1, a discrimination algorithm is used for obtaining the near-surface soil frozen state; specifically, in step 1, a passive microwave remote sensing-based Dual Index (DIA) algorithm is used as the most widely applied earth surface soil freeze-thaw remote sensing discrimination algorithm, and the improved DIA algorithm needs three parameters: 37GHz vertical polarization bright temperature (Tb)_37v) And the freeze-thaw state of the earth surface is judged by a negative bright temperature Spectrum Gradient (SG) of 19-37GHz and a Local Variance of Soil Moisture (LVSM), and the core algorithm can be summarized as an expression (1) and an expression (2).

Tb_37v≤P₃₇ (1)

Wherein, Tb_37vIs 36.5GHz vertical polarization bright temperature (K),shows a negative bright temperature spectral gradient (K/GHz), P, between 18.7GHz and 36.5GHz₃₇And P_SGRespectively representing two indexes Tb_37vAndthe threshold value of (2). In the normal case, P_SGThe value is 0. When the surface soil state satisfies the conditions of the formula (1) and the formula (2) at the same time, the surface soil can be judged as frozen soil, otherwise, the surface soil is judged as melted soil. According to an embodiment of the invention, the step 2, a freezing index method based on the near-surface soil freezing state is established;

to study the susceptibility of permafrost to climate change and to estimate permafrost distribution change in the context of global warming, Nelson and Outcale propose a simple "freezing index" (F) based on meteorological observation statistics_i) The method can be used for describing the continuous distribution of the permafrost on a small spatial scale, and is widely applied to multiple permafrostIn the research of frozen soil. The original freeze index is defined as:

in the formula (3), DDF₊ ^1/2And DDT^1/2The subscript "+" indicates that when the index calculation relates to winter snow depth, snow density and snow cover heat transfer characteristics, the snow cover has a certain effect on the air temperature and is therefore adjusted by negative index expression to reduce this effect. The freezing index method based on meteorological statistics is based on ground observation data, is suitable for research on the continuity distribution of small-scale permafrost and is difficult to be applied to the research on large-scale discrete space grids. Therefore, on the basis of the freezing index method, the freezing index method based on the freezing and thawing state of the surface soil is innovatively provided by fusing the surface soil freezing and thawing state information obtained by judging and using the passive microwave remote sensing data, and the method is expressed as the following formula (4):

in the formula (4), D_f ^1/2And D_t ^1/2The days for freezing and thawing the soil in one year. Thereby obtaining year-by-year sequence freezing index spatial distribution, and the information is an index for representing the space distribution and the climate sensitivity of the perennial frozen soil. By combining with the permafrost classification method, various types of permafrost distributions with different continuity can be obtained.

According to an embodiment of the invention, the step 3 is monitored and classified by combining a permafrost thermal stability zoning theory. Specifically, the permafrost latitudes are disturbed to a certain extent due to the influence of the landform and the mountain trend in the northeast of China. Therefore, the formation, distribution and development of the frozen soil in the area have both latitudinal and regional regularity, which is the basis and theoretical basis for the concept of the frozen soil partition in the area.

According to the method for partitioning the permafrost thermal stability in the northeast China area, which is proposed by scholars of Guo Dongxi and the like, firstly, the latitudes of the permafrost are fully expressed, and meanwhile, the regional regularity can be obviously reflected. Since the zonal characteristics of the permafrost are based on the distribution of climatic factors, the climatic factors largely determine and control the latitudinal zonal regularity of the permafrost. The actual distribution condition and the main characteristics of the permafrost in the northeast region have close relation with the annual average air temperature and are approximately consistent with certain air temperature isolines. Based on this point, taking the average annual temperature as the main basis of zoning, the permafrost is divided into four zones (or types):

1) the place with the temperature contour of-5.0 ℃ in the mean temperature of the year and north is a continuous permafrost zone;

2) -5.0 to-3.0 ℃ isotherms are in the discontinuous permafrost zone;

3) an island-shaped permafrost zone is arranged between the isothermal line of-3.0 ℃ and the permafrost south bound (isothermal line of 0 ℃);

4) the south boundary takes the south area as a seasonal frozen soil zone.

And then, establishing a relational model between the freezing index based on the freeze-thaw state of the earth surface soil and the annual average air temperature through meteorological station data, and obtaining the freezing index threshold value of the classification of the perennial frozen soil. According to the characteristics of the freezing index calculation formula, the natural logarithm of the average annual temperature and the freezing index have good linear relation, and in order to simplify the correlation relation between the freezing index and the average annual temperature, a relational model between the freezing index and the average annual temperature is expressed as an expression (5).

F_i＝a·ln(k-T_am)+b (5)

Wherein a, b and k are constants, T_amThe average annual temperature is k value according to the maximum annual average annual temperature (T)_am,max) Determination of k-T_am,maxNot less than 1. Therefore, the corresponding freezing index threshold values under different annual average air temperature threshold values can be obtained.

The permafrost has strong thermal stability, so the spatial distribution of the permafrost has one in a time sequenceStable and gradual change. However, the freezing index based on the freeze-thaw state of the surface soil has a certain mutation property because it is relatively largely affected by weather. Therefore, the regional regularity of the space-time distribution of the permafrost is described by considering the space distribution condition of the permafrost in the previous year when predicting the space distribution of the permafrost in the current year. In view of the above permafrost distribution and change characteristics, the present invention employs a modified freezing index (F)_i+) The method represents the spatial continuity of the permafrost and is calculated according to the formula (6).

In the formula, F_i+ ^tIndicating the modified freeze index for the current year. Alpha is a weight coefficient (0)<α<1)。

According to a specific application embodiment of the invention, the method is applied and verified in northeast China by using the method, firstly, compared with a frozen soil classification map in more than 2000 years, the distribution result of the frozen soil in more than 2003 has better classification precision. Due to the degradation of the continuous permafrost, in the classification system of the 2006-year permafrost region chart, the continuous permafrost and the discontinuous permafrost are classified into one class, the range of spatial continuity is changed, and the method is different from the permafrost classification system used by the method. Therefore, the classification accuracy of the perennial frozen soil in 2006 is verified by the total area statistics of the perennial frozen soil classification map in 2006. The statistical results of the validation in 2003 and 2006 are shown in table 1. The results show that the total classification error of frozen soil in 2003 and 2006 is 2.97% and 1.37%. However, as permafrost in research areas deteriorates and the area of the permafrost is continuously reduced, the actual classification error of the permafrost in 2003 should be less than 2.97%.

TABLE 1 area statistics for different types of permafrost

Fig. 2 shows that the distribution and classification results of the permafrost in 2003 and 2006 obtained by the method are compared with the existing permafrost region maps in 2000 and 2006 respectively, and the distribution difference can be found visually. Due to the comprehensive characteristics of the existing permafrost region maps, the difference between the permafrost distribution and classification result obtained by the method and the permafrost region map is mainly expressed on the boundary of classification. But the overall distribution state is relatively consistent, especially in the south edge of perennial frozen soil in the research area. It is noteworthy that in the perennial frozen soil compartmentalization chart in 2006, the perennial frozen soil of the western corners of the study area had disappeared, which is also reflected in the present results.

The permafrost monitoring method and parameters constructed by the invention are applied to the permafrost monitoring method and parameters in the 30-90-degree N area of the northern hemisphere to monitor and classify the permafrost. The effectiveness of the method can be discussed by combining the classification chart of the annual permafrost in the ring north of 1997. The results are shown in fig. 3, and fig. 2a-d are the classification chart of the annual north pole permafrost in 1997, and the permafrost distribution chart of the northern hemisphere in 2003, 2010 and 2017 in turn.

Fig. 3(a), (b), (c), (d) are the classification chart of the annual north pole permafrost in 1997 and the permafrost distribution chart of the northern hemisphere in 2003, 2010 and 2017.

Fig. 4 is illustrated in bitmap 3.

Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, but various changes may be apparent to those skilled in the art, and it is intended that all inventive concepts utilizing the inventive concepts set forth herein be protected without departing from the spirit and scope of the present invention as defined and limited by the appended claims.