CN113008843A - Winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing - Google Patents

Winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing Download PDF

Info

Publication number
CN113008843A
CN113008843A CN201911332612.XA CN201911332612A CN113008843A CN 113008843 A CN113008843 A CN 113008843A CN 201911332612 A CN201911332612 A CN 201911332612A CN 113008843 A CN113008843 A CN 113008843A
Authority
CN
China
Prior art keywords
drought
index
nibs
chlorophyll fluorescence
winter wheat
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201911332612.XA
Other languages
Chinese (zh)
Other versions
CN113008843B (en
Inventor
李强子
王思远
王红岩
杜鑫
张源
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Aerospace Information Research Institute of CAS
Original Assignee
Institute of Remote Sensing and Digital Earth of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Institute of Remote Sensing and Digital Earth of CAS filed Critical Institute of Remote Sensing and Digital Earth of CAS
Priority to CN201911332612.XA priority Critical patent/CN113008843B/en
Publication of CN113008843A publication Critical patent/CN113008843A/en
Application granted granted Critical
Publication of CN113008843B publication Critical patent/CN113008843B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/188—Vegetation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N2021/635—Photosynthetic material analysis, e.g. chrorophyll

Landscapes

  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

本发明公开基于TROPOMI叶绿素荧光遥感的冬小麦旱情监测方法，收集TROPOMI叶绿素荧光数据、冬小麦空间分布数据和土壤水分实地调查数据；进行等经纬栅格化、线性插补和SG滤波重建处理，得到研究区逐日空间连续SIF数据集；综合冬小麦空间分布数据和重建后的研究区逐日空间连续SIF数据集，构建归一化叶绿素荧光旱情指数NIBS；利用同期温度植被干旱指数TVDI和土壤水分实地调查数据，对NIBS指数进行精度评价；基于NIBS指数对研究区春旱旱情进行动态监测。NIBS指数在时空尺度上指示研究区农作物旱情，与同期其他旱情监测方法有较高相关性，同时其高时空分辨率的优势，能够为旱情监测提供丰富的时空变化信息。

Figure 201911332612

The invention discloses a winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing, collects TROPOMI chlorophyll fluorescence data, winter wheat spatial distribution data and soil moisture field survey data; performs equal latitude and longitude rasterization, linear interpolation and SG filter reconstruction processing to obtain a study area Daily spatial continuous SIF data set; integrated the winter wheat spatial distribution data and the reconstructed daily spatial continuous SIF data set of the study area to construct the normalized chlorophyll fluorescence drought index NIBS; using the temperature vegetation drought index TVDI and soil moisture field survey data in the same period, the The NIBS index was used to evaluate the accuracy; based on the NIBS index, the dynamic monitoring of the spring drought in the study area was carried out. The NIBS index indicates the drought conditions of crops in the study area on the temporal and spatial scales, and has a high correlation with other drought monitoring methods in the same period. At the same time, its high temporal and spatial resolution advantages can provide abundant temporal and spatial variation information for drought monitoring.

Figure 201911332612

Description

Winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing
Technical Field
The invention relates to the technical field of remote sensing monitoring. In particular to a winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing.
Background
The frequent occurrence of natural disasters and large-scale climate abnormalities are great challenges for human development, wherein drought disasters are wide in coverage range and long in duration, and are disasters which have the largest global natural influence range and influence human activities and loss most seriously. With the more severe climate change situation, the frequency degree and duration of drought also tend to increase, and crop yield reduction caused by the trend also threatens the grain safety of the country or region, so that the drought monitoring of the main grain producing area is particularly important. The traditional drought monitoring method mostly adopts a fixed ground monitoring and random investigation method, and has the characteristics of labor waste, low efficiency, poor precision and the like in the implementation process, the remote sensing technology has the development advantages of high time, high space, high spectrum and multiple platforms, the satellite remote sensing data is utilized to carry out continuous deepening of large-scale regional agricultural drought monitoring research, and the macroscopicity, economy, dynamics and effectiveness which are not possessed by the traditional monitoring method can be supplemented.
The drought monitoring based on the crop growth vigor can utilize satellite data inversion calculation to obtain various vegetation indexes, reflect the crop growth change state and further reflect the drought condition. The drought control index is mainly normalized vegetation index NDVI, vegetation state index VCI, range-flat vegetation index AVI, normalized difference water index NDWI and the like, although the indexes can indicate drought conditions in a large range, vegetation greenness information cannot reflect the instantaneous photosynthetic rate of vegetation, and a certain hysteresis disadvantage exists in the drought degree characterization based on the indexes. Compared with the method which only depends on the crop growth information to supplement the consideration of the canopy temperature, the drought remote sensing monitoring which integrates the canopy temperature and the crop growth, such as the temperature drought vegetation index TVDI, the vegetation water supply index VSWI and the like, is stronger in principle and is widely applied, but the method still highly depends on the vegetation greenness information in essence, and the hysteresis quality cannot be effectively improved.
Chlorophyll fluorescence is a long-wave signal released after plant chlorophyll absorbs energy, and can effectively indicate the physiological growth state and the stressed degree of vegetation. Compared with the active chlorophyll fluorescence induction which depends on an artificial radiation light source, the solar chlorophyll fluorescence induction (SIF) is obtained by taking solar radiation as an excitation source and can be obtained in multiple scales such as leaves, canopies, sky and the like. The SIF based on satellite remote sensing can be used as a probe for vegetation photosynthesis, can indicate the growth conditions and environmental stress of vegetation more directly than the traditional vegetation index, and although no satellite specially used for measuring SIF exists at present, the application potential of the SIF based on satellite remote sensing has prompted the academia to carry out SIF inversion and application according to the existing meteorological satellite platform.
Photosynthesis weakening and fluorescence quantum yield reduction caused by water stress provide theoretical support for SIF-based drought monitoring of different spatial scales. In the scale of the leaf and the canopy, the fluorescence yield of the vegetation can be directly reduced by water stress, even if the corresponding green degree information has no obvious change; the SIF data are obtained based on the inversion of the satellite-borne sensor, and the researchers also develop the research on the influence of drought events on vegetation in different regions of the world, such as Russia, Amazon forest, American middle plains, Australia and the like. The TROPOMI sensor carried by a sentinel-5p satellite and transmitted in 2017 in 10 months is an atmospheric monitoring spectrometer with the most advanced technical performance and the highest spatial resolution, has obvious space-time resolution advantage compared with the prior atmospheric monitoring satellite sensor, and brings new potential for drought monitoring research based on SIF data obtained by inversion of the TROPOMI sensor.
Disclosure of Invention
Therefore, the technical problem to be solved by the invention is to provide a winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing, which overcomes the defects of inconsistent spatial resolution and data loss of the original chlorophyll fluorescence remote sensing primary product.
In order to solve the technical problems, the invention provides the following technical scheme:
the winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing comprises the following steps:
(1) collecting TROPOMI chlorophyll fluorescence data, winter wheat spatial distribution data and soil moisture field investigation data;
(2) on the basis of TROPOMI chlorophyll fluorescence data, carrying out equal longitude and latitude rasterization, linear interpolation and SG filtering reconstruction processing to obtain a solar induced chlorophyll fluorescence SIF data set with continuous space day by day in a research area;
(3) integrating the spatial distribution data of the winter wheat in the step (1) and the solar-induced chlorophyll fluorescence SIF data sets of the reconstructed study area in the step (2) in a spatial continuous day-by-day manner to construct a normalized chlorophyll fluorescence drought index NIBS;
(4) performing precision evaluation on the normalized chlorophyll fluorescent drought index NIBS by using the contemporaneous temperature vegetation drought index TVDI and the soil moisture field investigation data in the step (1);
(5) and dynamically monitoring spring drought in the research area based on the normalized chlorophyll fluorescence drought index NIBS index.
The winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing comprises the following steps of (1):
TROPOMI chlorophyll fluorescence data: a solar induced chlorophyll fluorescence SIF product data set inverted based on a Tropomi sensor carried by a sentinel-5p satellite;
spatial distribution data of winter wheat: by collecting domestic high-grade first data during the growing season of the winter wheat in 2017-2018, combining ground sample data of field investigation in 2018, obtaining spatial distribution data of the winter wheat by using a maximum likelihood classification method, and resampling a classification result to be 1 km;
soil moisture field survey data: in the period from 23 days in 3 months to 28 days in 3 months, field investigation on the soil moisture is carried out in two areas of the Hebei stone village, the Chenge platform, the Shandong Jining and the Neze, and a Time Domain Reflectometer (TDR) is adopted to observe the soil moisture content of different soil layers.
In the above method for monitoring drought conditions of winter wheat based on TROPOMI chlorophyll fluorescence remote sensing, in the step (2), the method for rasterizing TROPOMI chlorophyll fluorescence data in equal longitude and latitude is as follows: respectively establishing equal longitude and latitude grids with step lengths of 0.1 degree, 0.05 degree and 0.2 degree in the spatial ranges of 30-43 degrees of north latitude and 110-113 degrees of east longitude in a research area; through traversing longitude and latitude coordinates of the centers of sampling points of non-rasterized sun-induced chlorophyll fluorescence SIF signals, the mean value of SIF signal sequences in each grid is counted, and three solar-induced chlorophyll fluorescence SIF signals with spatial resolution continuously rasterized are obtained.
In the above winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing, in the step (2), the linear interpolation method is as follows: by analyzing SIF sequence data on the same spatial position, assuming that the change of the sun-induced chlorophyll fluorescence SIF signal is in a linear trend in the signal loss period, and indexing the nearest effective value from the head to the tail of the loss interval to linearly interpolate the sun-induced chlorophyll fluorescence SIF sequence;
SG filtering reconstruction processing: smoothing the SIF sequence signal after linear interpolation by using a Savitzky-Golay algorithm, wherein the calculation method comprises the following steps:
Figure BDA0002330065640000041
in the formula (I), the compound is shown in the specification,
Figure BDA0002330065640000042
representing the reconstructed SIF sequence; y isj+iRepresenting the original SIF data after linear interpolation; ciRepresents a filter coefficient; m is the window width; n is the length of the filter and can be expressed as (2m + 1);
in the filtering process, two parameters need to be set: the window width m, the polynomial fitting times; the larger the window width m is, the smoother the filtered sequence data is; the degree of polynomial fit is set to 2-4, the smaller the degree, the smoother.
The winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing comprises the following steps in step (3):
(3-1) a winter wheat crop distribution density calculation method;
and (2) combining the spatial distribution data of the winter wheat in the step (1) to construct a spatial distribution density map of the winter wheat: on the basis of a preprocessed 0.1-degree sun-induced chlorophyll fluorescence SIF data set, the total distribution area of the winter wheat in each equal longitude and latitude grid is statistically analyzed, a winter wheat distribution density graph is generated by taking the crop area as a grid numerical value, wherein the spatial range and the resolution of the equal longitude and latitude grid are consistent with those of a rasterized sun-induced chlorophyll fluorescence SIF signal, namely the north latitude is 30-43 degrees, the east longitude is 110-113 degrees, and the step length is 0.1 degree;
(3-2) construction of normalized chlorophyll fluorescent drought index NIBS
Selecting grids with the distribution density of all the winter wheat being more than 5 on the basis of the winter wheat crop distribution density diagram in the step (3-1), counting sun-induced chlorophyll fluorescence SIF fluorescence signals which are subjected to filtering reconstruction in the same spatial position in the step (2), calculating the maximum value and the minimum value of the grids meeting the statistical conditions, constructing and calculating drought indexes, and providing normalized chlorophyll fluorescence drought indexes NIBS according to the fluorescence signal statistical characteristic values of the winter wheat distribution areas in the research area at the same moment, wherein the specific calculation methods are respectively as follows:
Figure BDA0002330065640000051
in the formula, SIFminThe minimum value of SIF signal meeting the statistical condition of winter wheat distribution at a certain time phase in a research region, SIFmaxThe maximum value of the SIF signal meeting the statistical condition of the distribution of the winter wheat at a certain time phase in a research area.
According to the winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing, in the step (4), the temperature vegetation drought index TVDI integrates vegetation coverage information and land surface temperature information, and can reflect the surface drought condition; the temperature vegetation drought index is obtained by a dry-wet edge fitting equation calculated by normalizing the vegetation index NDVI and the surface temperature LST, and the specific calculation formula is as follows:
Figure BDA0002330065640000052
LSTNDVIi max＝a1+b1NDVIi
LSTNDVIi max＝a2+b2NDVIi
in the formula: LSTNDVIiIs the land surface temperature; LSTNDVIi maxRepresents the highest surface temperature represented by a certain NDVI value, namely a dry edge value; LSTNDVIi minRepresents the lowest surface temperature represented by a certain NDVI value, i.e., the wet edge value; a1, b1 and a2, b2 are coefficients of dry-edge and wet-edge fitting equations, respectively; NDVIi represents a certain normalized vegetation index.
According to the winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing, the precision evaluation parameter of drought monitoring comprises a decision coefficient R2Root-Mean-Square Error (RMSE); the calculation methods of the decision coefficient and the root mean square error are respectively as follows:
Figure BDA0002330065640000053
Figure BDA0002330065640000054
in the formula:
Figure BDA0002330065640000061
respectively representing TVDI or soil moisture SM values and their corresponding mean values, yiIs an NIBS index value, and m is the number of sampling points;
wherein R is2For describing the correlation between the constructed NIBS index and TVDI and soil moisture, R2The larger, the higher its correlation; and the RMSE is used for describing the stability of the correlation between the NIBS index and the real soil moisture, and the smaller the RMSE value, the higher the stability.
The winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing comprises the following steps: the normalized chlorophyll fluorescence index NIBS is related to the contemporaneous temperature vegetation drought index TVDI and the contemporaneous soil moisture field investigation data, and winter wheat spring drought monitoring can be monitored through the normalized chlorophyll fluorescence drought index NIBS.
The winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing,
correlation of normalized chlorophyll fluorescence index NIBS with contemporaneous temperature vegetation drought index TVDI: the NIBS indexes of different subareas of the research area are in negative correlation with the contemporaneous remote sensing drought monitoring index TVDI, which shows that the NIBS indexes can have the same indication function of better separation degree as the TVDI in terms of the drought degree at the same moment, thereby proving the effectiveness of the NIBS index drought indication on the spatial distribution;
correlation analysis of the NIBS index with soil moisture: the NIBS index has a clear positive correlation with soil moisture SM.
The technical scheme of the invention achieves the following beneficial technical effects:
aiming at the characteristic that sun-Induced chlorophyllin Fluorescence (SIF) can effectively indicate water stress of land vegetation, the Normalized chlorophyll Fluorescence drought Index (NIBS) is used for drought monitoring of winter wheat in Huang-Huai-Hai region. The method comprises the steps of firstly rasterizing an SIF original product set obtained by inversion of a Sentinel-5p satellite (Sentinel-5p) troposphere observation Instrument (TROPOMI) sensor into space continuous data through longitude and latitude step lengths of 0.1 degree and the like, then carrying out linear interpolation on missing values based on time sequence analysis, reconstructing through S-G filtering to obtain a high-space-time resolution fluorescence data set, and constructing the NIBS index by combining winter wheat distribution data in a research area on the basis of the data set. Through selecting typical drought events for comparative analysis, the NIBS Index and the Temperature Vegetation Drought Index (TVDI) at the same period have good correlation in different ripening areas, wherein the R in the Yanshan area is2Maximum, 0.66, Huang-Huai plain region R2Minimum, 0.44; the NIBS index is also highly correlated with field soil moisture survey results, where R is the northern river sample area20.53, Shandong-like region R2Is 0.54, bulk R2Is 0.51. Experimental results show that the NIBS index can effectively indicate the drought of winter wheat in Huang-Huai-Hai region on a space-time scale.
The application provides an equal longitude and latitude rasterization and filtering reconstruction processing method based on a TROPOMI sensor original SIF data set, then synthesizes processed SIF signals and winter wheat distribution data in a research area, constructs a normalized chlorophyll fluorescence drought index NIBS, and evaluates the drought monitoring capability of the NIBS. Research results show that the NIBS index can indicate the drought conditions of crops in a research area on a time-space scale, has higher correlation with other drought monitoring methods in the same period, has the advantage of high time-space resolution, and can provide richer time-space change information for drought monitoring.
The study of the application is based on the assumption that the fluorescent signal of winter wheat stressed by drought is weaker than that of healthy crops, and the drought monitoring experiment by utilizing TROPOMI chlorophyll fluorescence data is developed: by utilizing the equal longitude and latitude rasterization and filtering reconstruction method, the defects of inconsistent spatial resolution and data loss of the original chlorophyll fluorescence remote sensing primary product are overcome, and the normalized chlorophyll fluorescence drought index NIBS is constructed by combining the reconstructed fluorescence data and the winter wheat planting distribution of the research area; by comparing the NIBS index and the TVDI in the same period, the two drought indexes in different ripening areas have good correlation relationship, and meanwhile, soil moisture data and the NIBS index at the corresponding time are highly correlated based on field investigation, so that the NIBS index can effectively indicate the drought stress degree of crops on the spatial distribution; and then, drought monitoring is carried out on the spring of the research area based on the NIBS index, and the result shows that the NIBS index can reflect the evolution process of the drought and prove the effectiveness of the NIBS index from the time change.
The method has the advantages that (1) the longitude and latitude rasterization and reconstruction method for TROPOMI sensor chlorophyll fluorescence inversion products and the like is provided. The method expands the superiority of chlorophyll fluorescence data obtained by TROPOMI sensor inversion on a space-time scale, and the space-time continuous chlorophyll fluorescence data set can provide powerful data source guarantee for drought analysis. (2) The normalized chlorophyll fluorescence index NIBS capable of effectively monitoring drought is constructed. The NIBS index integrates crop distribution data and a reconstructed chlorophyll fluorescence signal, can indicate drought stress degrees of different areas at the same time phase, and can update monitoring results day by day to dynamically track the drought evolution process; compared with the traditional remote sensing drought monitoring index TVDI which relies on eight-day synthetic data products, the NIBS index can monitor the change rule of the drought on a finer time scale.
Drawings
FIG. 1 is a plot of the geographic location and winter wheat distribution for a study area of the present invention;
FIG. 2 is a diagram of a soil moisture sample point distribution for field investigation; a is 145 sampling points arranged in Shijiazhuang City and the Chenchen Tai City in Hebei province; b is 52 sampling points arranged in Jining and Yuze of Shandong province; c is 6 sample points of several surveys in Hebei province; d is a specific measurement method;
FIG. 3 is a technical flow chart of the method for monitoring drought of winter wheat based on TROPOMI chlorophyll fluorescence remote sensing;
FIG. 4 shows results of equal longitude and latitude rasterization for different step lengths; the upper three graphs are the result graphs of rasterization in steps of 0.05 degrees, 0.1 degrees and 0.2 degrees respectively for 3, 25 and 2019; the lower three graphs are the result graphs of rasterization in steps of 0.05 degrees, 0.1 degrees and 0.2 degrees respectively for 5, month and 2 days in 2019;
FIG. 5 is a schematic diagram of linear interpolation and S-G filter reconstruction of SIF sequence data; a is SIF sequence data of a grid with 37 degrees of Beijing at the north latitude and 120 degrees of Tokyo (the original point is the statistical daily corrected SIF mean value in the area of the grid, and the cross point is SIF signal loss in the grid at the moment); b is a linear interpolated SIF sequence of a grid of Beijing 120 degrees with 37 degrees of Beijing latitude; c is filtered sequence data of a grid of 37 degrees tokyo 120 degrees north latitude (a smooth solid line is S-G filtered reconstruction data, and a dotted line is an interpolated SIF sequence); a is SIF sequence data of a 33-degree Beijing 116-degree grid in northern latitude (the original point is the daily corrected SIF mean value obtained by statistics in the area of the grid, and the cross point is SIF signal loss in the grid at the moment); b is a SIF sequence after linear interpolation of a 33-degree Tokyo 116-degree grid of the north latitude; c is filtered sequence data of a grid of 33 ° tokyo 116 ° north latitude (smooth solid line is reconstructed data after S-G filtering, dotted line is SIF sequence after interpolation);
FIG. 6a is the distribution data of winter wheat with 1km resolution;
FIG. 6b is a graph of winter wheat distribution density at the corresponding locations;
FIG. 6c is a graph of winter wheat distribution density throughout the study area;
fig. 7 is a graph comparing SIF data before and after filtering reconstruction: a-h are respectively SIF data after equal longitude and latitude rasterization of 16 days in 5-month in 2019, 17 days in 5-month in 2019, 18 days in 5-month in 2019, 19 days in 5-month in 2019, 20 days in 5-month in 2019, 21 days in 5-month in 2019, 22 days in 5-month in 2019 and 23 days in 5-month in 2019; A-H are SIF data after filtering reconstruction of 16 days in 5-month in 2019, 17 days in 5-month in 2019, 18 days in 5-month in 2019, 19 days in 5-month in 2019, 20 days in 5-month in 2019, 21 days in 5-month in 2019, 22 days in 5-month in 2019 and 23 days in 5-month in 2019 respectively;
FIG. 83 is a graph showing a calculation result of NIBS index at 30 days;
fig. 93 is a graph of calculation results of TVDI index at 30 days of month;
FIG. 10a shows the correlation analysis of NIBS index and TVDI in Yanshan region;
FIG. 10b is the correlation analysis result of NIBS index and contemporaneous TVDI in Heilongjiang area;
FIG. 10c is the correlation analysis of NIBS index and contemporaneous TVDI in the North Roxburgh region;
FIG. 10d shows the correlation analysis of NIBS index and TVDI in mountainous hilly areas;
FIG. 10e is the correlation analysis result of NIBS index and contemporaneous TVDI in Yuxi region;
FIG. 10f shows the correlation analysis of NIBS index and TVDI in Huang-Huai-Ping-Yuan region;
FIG. 10g shows the correlation analysis result of the NIBS index and the contemporaneous TVDI in Yu Wan district of Hubei province;
FIG. 10h shows the correlation analysis result of the NIBS index and the TVDI at the same period in Jianghuai plain region;
FIG. 10i shows the correlation analysis of NIBS index and TVDI in Huang-Huai-Hai region;
FIG. 11 correlation analysis results of NIBS with soil moisture: a is a scatter diagram of the NIBS index relation between soil moisture measured by Hebei stone house and a chenchen platform area and corresponding time; b is the correlation between soil moisture of field survey of Shandong Jining and Nethereto and NIBS indexes in the same period, c is the correlation between all field survey data of the Hebei and Shandong fields and the NIBS indexes;
FIG. 12 shows the drought monitoring results of spring wheat, in which the normalized chlorophyll fluorescence index NIBS was calculated day by day during the whole spring (2.15 to 6.1 days) in Huang-Huai-Hai region (2.15 to 3.31 days in FIG. 12-1, 4 to 5.15 days in FIG. 12-2, and 6 to 1 day in FIG. 12-3).
Detailed Description
First, research method
In the embodiment, equal longitude and latitude rasterization and SG filtering reconstruction processing are carried out on the basis of a TROPOMI chlorophyll fluorescence remote sensing original product to obtain a continuous SIF data set of a research area day by day space; then, synthesizing the distribution of the winter wheat and the reconstructed fluorescence signal to construct a normalized chlorophyll fluorescence drought index NIBS; the precision evaluation is carried out on the NIBS index by using the TVDI and the field investigation data of the soil moisture in the same period; and finally, dynamically monitoring the spring drought in the research region based on the NIBS index. The general technical flow is shown in fig. 3.
Second, research area and data source thereof
1. Overview of the region of investigation
In the embodiment, a Huang-Huai-Hai winter wheat growing area is selected as a research area, the Huang-Huai-Hai area covers five provinces of Hebei, Shandong, Jiangsu, Anhui and Henan and two direct prefectures of Beijing and Tianjin, the cultivated land area of the Huang-Huai-Hai winter wheat growing area accounts for about one fifth of the whole country, and the Huang-Huai-Hai winter wheat growing area is an important grain, cotton and oil production base in China.
In continental monsoon type climates in a temperate zone of a research area, photo-thermal resources are sufficient, four seasons change obviously, wherein high temperature and raininess are achieved in summer, cold and dry are achieved in winter, annual rainfall is 500 mm-600 mm, and summer is mainly focused. The Huang-Huai-Hai area is one of the areas with the largest drought-stricken area in China, the frequent pumping and adjustment of a large amount of underground water are needed for regular irrigation to relieve drought conditions, the precipitation in winter in a research area is less, the arrival time of a frontal rain zone in spring is late, the rising speed of the air temperature is increased, the increase of evaporation capacity and the insufficient water storage capacity are superposed, so that the spring drought conditions are easily caused, and the growth vigor and the yield of crops are severely limited.
2. Data Source introduction
2.1TROPOMI chlorophyll fluorescence data
The sentinel-5p satellite is the first satellite of the European Global environmental and monitoring System (Copenny) plan, and a carried troposphere observation instrument (TROPOMI) sensor can provide global service for dynamic monitoring of atmospheric chemistry, environmental pollution, ozone and aerosol. The TROPOMI sensor can effectively monitor trace gas components in the atmosphere of each global place, the resolution of the sub-satellite point is as high as 7km multiplied by 3.5km, the swath width is 2600km, and the imaging capability can cover the global every day.
The chlorophyll fluorescence remote sensing data adopted by the embodiment is derived from SIF product data sets inverted by a tropimai sensor carried on a Sentiniel-5 p, such as Philipp and Christian, a chlorophyll fluorescence signal is obtained by inversion of a data driving algorithm in a range from 743 to 758nm, the window contains a strong Fraunhofer dark line, and the absorption influence of moisture, oxygen and the like on the signal can be effectively avoided.
2.2MODIS data
A middle-resolution Imaging spectrometer (MODIS) is one of the main sensors mounted on Terra and Aqua satellites, and the high spatial-temporal resolution and rich spectral bands of the MODIS can be widely applied to extraction and analysis of information such as ground surface temperature, vegetation index, atmospheric water vapor and aerosol. In the embodiment, the MODIS Land Surface Temperature (LST) product MYD11A2 and the reflectivity product MYD09A1 are mainly used for calculating the drought index TVDI by remote sensing monitoring, wherein the Land Surface reflectivity product calculates the normalized vegetation index NDVI by using a red light wave band (620-670nm) and a near infrared wave band (841-876 nm).
2.3 soil moisture field survey data
Soil moisture is the most direct characterizing parameter of agricultural drought and is a function of crop growth conditions and environmental factors. In this example, crop drought characterization in the study area was quantified by field investigation of soil moisture. According to the meteorological drought monitoring result issued by the national climate center in 2019, 3 and 14 days, more than moderate drought occurs in the Hebei stone village, the Chenge station, the Shandong Jining and the Neze, and the field investigation of the field soil moisture is respectively carried out in the two areas during the period from 23 days in 3 months to 28 days in 3 months.
Time Domain Reflectometry (TDR) is a sensor based on the frequency domain reflection principle, and can observe the soil moisture content of different soil layers. In the research, the TDR measuring instrument is used for measuring the soil moisture parameters within the range of 0-100%, the measurement error is +/-3%, and the comparison between soil moisture contents under different degrees of drought stress is met. As shown in fig. 2(a, b), 145 sampling points are distributed in the north-river province sample area, and 52 sampling points are distributed in the east-Shandong province sample area. The sampling points are distributed in the central zone of the large-area concentrated planting area of the winter wheat, so that the influence of other ground objects such as forest lands, construction lands and the like on the remote sensing monitoring result is avoided. Fig. 2(c) shows six spots of a centralized survey, in which nine measurements are performed at each spot, and the specific measurement method is shown in fig. 2 (d): 20X 20m centered on the spot2And randomly selecting three positions for sampling in the range, measuring three times during sampling each time, and taking the average value of nine records as the soil moisture representative value of the sampling point. The blue grid in fig. 2 represents a 0.1 ° × 0.1 ° iso-graticule corresponding to the sampling points for statistical analysis.
2.3 spatial distribution data of winter wheat
By collecting the domestic high-grade first-number data during the growing season of the winter wheat in 2017 and 2018 and combining with ground sample data of field investigation in 2018, the spatial distribution data of the winter wheat is obtained by using the maximum likelihood classification method, the classification result is resampled to 1km for subsequent statistical research, the overall precision of a distribution graph reaches 93.21%, the experimental requirement is met, and the spatial distribution of the winter wheat is shown in figure 1.
Three, chlorophyll fluorescence data and other longitude and latitude rasterization and filtering reconstruction
1. Equal longitude and latitude rasterization SIF data method
Due to the influence of swath width, earth curvature and observation angle change of the TROPOMI sensor, the transverse sampling width of the sensor is different from 3km of the sub-satellite point to 15km of the edge, the sub-satellite point can not be overlapped every day for a revisit period of 17 days, and further the consistency of the spatial resolution of the sampling footprints in the same region on a time sequence can not be ensured.
In the embodiment, equal graticules with step lengths of 0.1 °, 0.05 ° and 0.2 ° are respectively established in spatial ranges of 30 ° to 43 ° north latitude and 110 ° to 113 ° east longitude where the research area is located. And (3) counting the mean value of SIF signal sequences in each grid by traversing the longitude and latitude coordinates of the center of the sampling point of the non-rasterized SIF signal, and obtaining the SIF signals with three spatial resolutions and continuously rasterized.
As shown in fig. 4, taking 3 month, 25 days and 5 month, 2 days with higher quality of original data as an example, comparing the rasterization results with different step lengths, it can be found that a large number of fluorescence null value strips appear in the grid data with 0.05 ° step length, while the 0.2 ° step length is coarser than the grid data with 0.1 ° step length, and a large number of area difference details are lost, so that the original daily SIF data is rasterized and batched with 0.1 ° as the optimal statistical step length, and an equal latitude and longitude chlorophyll fluorescence remote sensing time sequence data set is obtained.
2. Linear interpolation and S-G filter reconstruction
Some time intervals SIF signals of part of research areas are lost due to cloud layer shielding, track change, data quality and the like, and effective signal representative values cannot be obtained on the target grid by the method; moreover, the data missing at the same time has the characteristics of wide distribution area and continuous space, and effective SIF signal reconstruction is difficult to realize by a space statistical interpolation method based on adjacent geographic space attributes.
Based on the data characteristics of the SIF signals, the sequence reconstruction is realized by analyzing the change trend of the SIF signals in a time sequence angle, and the reconstruction method is based on the principle that the change trend of the SIF signal sequence in the same space position in the similar time is stable, and the specific method comprises the following steps: by analyzing SIF sequence data at the same spatial position, the change of the SIF signal is assumed to be in a linear trend at the signal loss period, and the SIF sequence is linearly interpolated by the effective value with the nearest head and tail of the index loss interval. Taking two grid data of 0.1 ° × 0.1 ° in the research area as an example, the interpolation process is shown in fig. 5(a, B, a, B), wherein a red point value (a, a) represents that SIF signals are missing in the grid at the moment, and a blue point represents that daily corrected SIF mean values are obtained in statistics in the grid area; and (B, B) represents a SIF sequence after linear interpolation.
The interpolated SIF sequence has local oscillation and abnormal points, and the Savitzky-Golay algorithm is used to smooth the interpolated SIF sequence signal. The S-G filtering is a least square convolution algorithm (Savitzky and Golay,1964) proposed by Savitzky and Golay in 1964, and is widely applied to smoothing and denoising of time-domain data streams, and the core idea is that a reconstructed time series curve approaches to an upper envelope curve of an original sequence curve through repeated iteration processing; compared with other traditional denoising algorithms, the S-G filtering algorithm has the advantages of simple theory, easy implementation and no limitation of data time, space scale and sensors, and can smooth fine sawtooth noise of time sequence data without influencing key characteristics of the whole curve. The specific calculation method is as follows:
Figure BDA0002330065640000131
in the formula (I), the compound is shown in the specification,
Figure BDA0002330065640000132
representing the reconstructed SIF sequence; y isj+iRepresenting the original SIF data after linear interpolation; ciRepresents a filter coefficient; n is the length of the filter and can be expressed as (2m + 1). In the filtering process, two parameters need to be set: the first is the window width m, and the second is the polynomial fitting degree. Generally, the larger the window width m, the smoother the filtered sequence data, while the polynomial fitting degree is generally set between 2 and 4, and the smaller the degree, the smoother the sequence data.
By comparing the effects before and after filtering through repeated experiments, in order to remove obvious oscillation noise and ensure that key change information of the key SIF sequence is not changed, the result of finally determining that the width m of the filtering window is 7 and the smoothing times is 2 is shown in fig. 5(C, C), wherein a red curve is reconstructed data after S-G filtering, and blue is an interpolated SIF sequence.
Fourth, normalized chlorophyll fluorescence drought index calculation method
1. Winter wheat crop distribution density calculation method
In the embodiment, for the accuracy of monitoring the drought of the winter wheat, a spatial distribution density map of the winter wheat is constructed by combining with distribution data of the winter wheat, wherein the total distribution area of the winter wheat in each equal longitude and latitude grid is statistically analyzed based on a preprocessed 0.1-degree SIF data set, and the distribution density map of the winter wheat in the Huang-Huai-Hai region is generated by taking the crop area as a grid numerical value. Wherein the spatial range and resolution of the equal longitude and latitude grid are consistent with those of the rasterized SIF signal, namely 30-43 degrees of north latitude, 110-113 degrees of east longitude and latitude, the step length is 0.1 degree, and the calculation result is shown as follows, wherein (a) is winter wheat distribution data with the resolution of 1 km; (b) the distribution density of the winter wheat at the corresponding position is shown; (c) the distribution density of winter wheat throughout the study area is plotted.
2. Construction of normalized chlorophyll fluorescent drought index NIBS
And selecting all grids with the distribution density of the winter wheat more than 5 on the basis of the distribution density graph of the winter wheat, counting the fluorescence signals which are subjected to filtering reconstruction at the same spatial position, and calculating the maximum value and the minimum value of the grids which meet the statistical conditions for constructing and calculating the drought index. According to the statistical characteristic value of the Fluorescence signal of a winter wheat distribution area at the same moment in a research area, a Normalized chlorophyll Fluorescence drought Index NIBS (Normalized Index Based Solar-Induced chlorophyl Fluorescence) is provided, and the specific calculation methods are respectively as follows:
Figure BDA0002330065640000141
in the formula, SIFminThe minimum value of SIF signal meeting the statistical condition of winter wheat distribution at a certain time phase in a research region, SIFmaxThe maximum value of the SIF signal meeting the statistical condition of the distribution of the winter wheat at a certain time phase in a research area.
Fifthly, the precision evaluation is carried out on the NIBS index of the normalized chlorophyll fluorescence drought index
1. Temperature vegetation drought index calculation method
The temperature vegetation drought index TVDI integrates vegetation coverage information and land surface temperature information, and can reflect the surface drought condition. The temperature vegetation drought index is obtained by a dry-wet edge fitting equation calculated by normalizing the vegetation index (NDVI) and the surface temperature (LST), and the specific calculation formula is as follows:
Figure BDA0002330065640000142
LSTNDVIi max＝a1+b1NDVIi
LSTNDVIi max＝a2+b2NDVIi
in the formula: LSTNDVIiIs the land surface temperature; LSTNDVIi maxRepresents the highest surface temperature represented by a certain NDVI value, namely a dry edge value; LSTNDVIi minRepresents the lowest surface temperature represented by a certain NDVI value, i.e., the wet edge value; a1, b1 and a2, b2 are coefficients of dry-edge and wet-edge fitting equations, respectively; NDVIi represents a certain normalized vegetation index.
2. Precision evaluation method
The precision evaluation parameters selected for drought monitoring in the research comprise a determination coefficient R2Root-Mean-Square Error RMSE (Root-Mean-Square Error). Wherein R is2For a description of the correlation between the NIBS index and TVDI and soil moisture, R, constructed herein2The larger, the higher its correlation; and the RMSE is used for describing the stability of the correlation between the NIBS index and the real soil moisture, and the smaller the RMSE value, the higher the stability. The calculation methods of the decision coefficient and the root mean square error are respectively as follows:
Figure BDA0002330065640000151
Figure BDA0002330065640000152
in the formula:
Figure BDA0002330065640000153
respectively representing the TVDI or SM value and its corresponding mean, yiIs a NIBS fingerThe number value, m is the number of spots.
Sixthly, dynamic monitoring of spring drought in research area based on normalized chlorophyll fluorescence drought index NIBS index
1. Reconstructed high spatial and temporal resolution chlorophyll fluorescence data
And (3) performing equal longitude and latitude rasterization on day-by-day data of the TROPOMI chlorophyll fluorescence remote sensing inversion original product from 3 to 6 months in 2019, respectively analyzing a time sequence change curve of SIF signals in each grid, and performing linear interpolation of an invalid value and S-G filtering reconstruction, wherein the time sequence change curve from 16 to 23 months in 5 to 5 months in 2019 is taken as an example, FIGS. 7a-H are SIF data after equal longitude and latitude rasterization, and A-H show the SIF data after filtering reconstruction, and respectively take blue and green two different colors to stretch and contrast the effects before and after the reconstruction of filtering and the change condition of the fluorescence value in an eight-day interval.
The equal longitude and latitude rasterization method with the step size of 0.1 degree can better reflect the space difference of chlorophyll fluorescence on the basis of ensuring enough high space resolution, but the data loss of the original SIF product can cause large-area invalid grid numerical values such as 18 days in 5 months and 19 days in 5 months, and the fluorescence signal close to the loss period is taken as the necessary parameter and basis of linear interpolation through the time sequence data analysis of the space position of the loss signal, thereby ensuring the scientificity of fluorescence reconstruction.
The result shows that the change trend of the reconstructed filtered fluorescence data in the whole research area is basically consistent with the rasterized original data, and the phenomenon of integral enhancement and weakening is shown; the local fluorescence details of the reconstructed data can be well preserved after filtering, and by taking 5 months and 22 days as an example, the high-value regions in the northern Jiangsu and the southern Anhui can be effectively reflected after reconstruction and filtering, so that the regional key characteristics of the fluorescence signals can not be lost due to filtering based on time series analysis.
2. Correlation of NIBS with contemporaneous temperature vegetation drought index TVDI
In order to verify the effectiveness of the normalized chlorophyll fluorescence index NIBS, the present example selects a contemporaneous temperature vegetation drought index TVDI for comparison analysis. TVDI takes into account both terrestrial vegetation coverage information and temperature information, and has been proven by the academia to be a reasonable indicator for describing surface soil moisture changes and crop stress.
Since the plain regions of Huang-Huai-Hai are wide, and the difference of different hydrothermal conditions, climatic characteristics and production conditions leads to different agricultural planting systems in various regions, the research region is divided into eight sub-regions according to 'Chinese agricultural cooked region' in order to eliminate or weaken the error influence on the analysis result caused by the farming system, sowing time and climatic characteristics, which are respectively Yanshan region, Black dragon harbor region, North and Lu region, Shandong hilly region, Yuxi region, Huang-Huai plain region, Hu Yu region and Jiang-Huai plain region. 270 verification sample points are uniformly and randomly generated in eight ripening area winter wheat distribution areas, and the spatial distribution of the sample points is shown in figure 8; correlation analysis was performed by extracting the NIBS index of 3, 30 and 2019 and the contemporaneous TVDI calculation result (fig. 9) according to the geographic location of the sampling point, and the result is shown in fig. 10.
Larger TVDI values indicate more severe drought, while the NIBS index characterisation is of opposite significance-larger values indicate less stress on the crop. On the whole, R calculated from all verification sampling points in the research area2At 0.41, the contemporaneous NIBS and TVDI values are clearly negatively correlated in the Huang-Huai-Hai region. In particular, R in a subregion2More than 0.6 of Yanshan region, Heilonghong region and Yuxi region, the Yanshan region has the highest relativity, R2The slope of the trend line of the three regions is relatively close to be about-0.47; the lowest correlation in the sub-regions is the Huang-Huai plain region, R20.44, but still exceeds the overall level.
The experimental result shows that the NIBS indexes of different subareas of the research area keep good correlation with the contemporaneous remote sensing drought monitoring index TVDI, and the NIBS indexes can have the same indication effect as the TVDI in the aspect of the degree of drought at the same moment and have better separation degree, so that the validity of the NIBS index drought indication is proved in spatial distribution.
3. Correlation analysis of NIBS index and soil moisture
Taking grid space reference of NIBS index with spatial resolution of 0.1 degree as statistical basis, and in the grid with soil moisture measurement sampling point distribution, counting the average value of soil moisture recorded in field actual measurement as the soil moisture representative value of the grid. And respectively extracting the calculation results of the NIBS indexes on the corresponding dates through the time records of the field investigation of the soil moisture, wherein the soil moisture measurement time intervals in the same grid participating in statistics are all within one day.
FIG. 11(a) is a plot of NIBS index as measured by a Chenchen stage, a Kelvin, in Hebei; FIG. 11(b) is a graph showing the correlation between soil moisture and contemporary NIBS index in Shandong Jining, Nethereto field survey; FIG. 11(c) is a graph of the correlation between the NIBS index and all the field survey data in Hebei and Shandong. The sample division statistical result shows that the NIBS indexes of the two sample areas are obviously positively correlated with the soil moisture SM, and R is2The numerical values are all around 0.54, wherein the root mean square error of a Shandong sample area is 24.66, and the root mean square error of a Hebei sample area is 10.83; from the overall statistical results, the NIBS index and the soil moisture SM still keep high correlation2An RMSE of 18.21 was found between the two zones of 0.51.
The experiment result shows that the correlation between the NIBS index and the soil moisture in the same period is stable, and the same consistency is shown in both sample areas. Wherein the root mean square error of the Shandong sample area is greater than that of the Hebei sample area, probably because: the long-time spring drought leads the irrigation time of the Huang-Huai-Hai region to be earlier than the historical period, the fluorescence signal of the irrigated winter wheat is higher than the average level, and during investigation, the sample points are selected in the non-irrigated dry land, so that the NIBS index is doped with the fluorescence signals of two types of irrigation and drought stress. During field investigation, the proportion of cultivated land which has been irrigated in the Shandong sample area is obviously larger than that in the Hebei sample area, which causes the stability of the correlation between the NIBS and the SM to be worse than that in the Hebei sample area, and also indicates that the NIBS index is sensitive to irrigation.
4. NIBS-based winter wheat spring drought monitoring
Based on the normalized chlorophyll fluorescence drought index calculation method, the normalized chlorophyll fluorescence index NIBS is calculated day by day in the whole spring (2 months and 15 days to 6 months and 1 day) of the Huang-Huai-Hai region, and a drought monitoring experiment is carried out. Here, key nodes of spring drought and day-to-day change results are displayed and analyzed, and nine-day NIBS index results are selected and shown in fig. 12.
Monitoring results show that the growth condition of winter wheat in a research area of 15 days in 2 months is good, and the NIBS index in most areas is a positive value; NIBS indexes of 3, 5 and 5 days of Shandong, Hebei, Henan east and Anhui south are all reduced to be below zero; when March is low, the drought continuously increases, the NIBS index is reduced to the lowest value in the monitoring interval, but the NIBS values of the east of Henan, Anhui and the north of Jiangsu still keep higher level; the overall distribution trend of April is kept unchanged, but the NIBS index of the area below zero value begins to rise again, and the change trend is consistently continued to the middle of May; the monitoring result of 5 months and 15 days shows that the NIBS indexes of most areas except the south of Anhui province are higher than zero; and 6, 1 day in the 6 th month, finishing the growing season of the winter wheat, finishing harvesting the winter wheat in most regions in the south of the research area, wherein the low value of the NIBS no longer represents the stress influence of the drought, for example, the regions with higher NIBS values before the east of Henan, the north of Anhui and the like are all reduced to be below a zero value after being harvested, the northern un-harvested wheat area still maintains higher NIBS, and the drought basically does not appear.
From the overall trend, the NiBS index can effectively indicate the evolution process from the beginning of the drought, the continuous aggravation to the final death in the spring drought in the Huang-Huai-Hai region and the ending in the middle of May; based on the high time resolution of the TROPOMI sensor and the reconstruction method of the fluorescence data provided by the invention, the NIBS index has the advantage of updating day by day, and taking 5-month 13-5-month 15 as an example, the NIBS index monitoring result can capture the change of the drought in a short time, and has obvious time advantage compared with the traditional mode. The spatial resolution is limited, the indication performance of the NIBS index to drought in the area with low winter wheat planting area proportion is weaker than that in the area with high winter wheat planting area proportion, and the numerical value in the result of the NIBS index is lower in the same period, because the calculation error brought by the mixed pixel is eliminated and weakened, SIF signals corresponding to the area with low area proportion do not participate in the statistics of the characteristic value of the whole fluorescence signal in the research area, but the NIBS index of the NIBS index is still influenced by the type and the occupied proportion of the ground object, so that the result of the NIBS index has larger difference than that of the NIBS index calculated in the area with high winter wheat area proportion.
Seven, conclusion
Studying the hypothesis that the fluorescent signal of winter wheat stressed by drought is weaker than that of healthy crops, a drought monitoring experiment using TROPOMI chlorophyll fluorescence data is developed: by utilizing the equal longitude and latitude rasterization and filtering reconstruction method, the defects of inconsistent spatial resolution and data loss of the original chlorophyll fluorescence remote sensing primary product are overcome, and the normalized chlorophyll fluorescence drought index NIBS is constructed by combining the reconstructed fluorescence data and the winter wheat planting distribution of the research area; by comparing the NIBS index and the TVDI in the same period, the two drought indexes in different ripening areas have good correlation relationship, and meanwhile, soil moisture data and the NIBS index at the corresponding time are highly correlated based on field investigation, so that the NIBS index can effectively indicate the drought stress degree of crops on the spatial distribution; and then, drought monitoring is carried out on the spring of the research area based on the NIBS index, and the result shows that the NIBS index can reflect the evolution process of the drought and prove the effectiveness of the NIBS index from the time change.
The method has the advantages that (1) the longitude and latitude rasterization and reconstruction method for TROPOMI sensor chlorophyll fluorescence inversion products and the like is provided. The method expands the superiority of chlorophyll fluorescence data obtained by TROPOMI sensor inversion on a space-time scale, and the space-time continuous chlorophyll fluorescence data set can provide powerful data source guarantee for drought analysis. (2) The normalized chlorophyll fluorescence index NIBS capable of effectively monitoring drought is constructed. The NIBS index integrates crop distribution data and a reconstructed chlorophyll fluorescence signal, can indicate drought stress degrees of different areas at the same time phase, and can update monitoring results day by day to dynamically track the drought evolution process; compared with the traditional remote sensing drought monitoring index TVDI which relies on eight-day synthetic data products, the NIBS index can monitor the change rule of the drought on a finer time scale.
Meanwhile, the fluorescence signals contained in SIF products are not completely from crops due to TROPOMI imaging characteristics and spatial resolution, which directly causes the defect that mixed pixels exist in NIBS index results and influences monitoring precision. The carbon satellite of the terrestrial ecosystem expected to be launched in China in 2020 carries the super-spectral resolution load and can be specially observed aiming at the terrestrial vegetation SIF. Compared with SIF products obtained by inversion of atmospheric monitoring satellite sensors, the land carbon satellite SIF products have higher spatial resolution and higher signal-to-noise ratio, and the NIBS index based on the land carbon satellite has greater monitoring potential on crop stress and growth states.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications are possible which remain within the scope of the appended claims.

Claims (9)

1. The winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing is characterized by comprising the following steps:
(1) collecting TROPOMI chlorophyll fluorescence data, winter wheat spatial distribution data and soil moisture field investigation data;
(2) on the basis of TROPOMI chlorophyll fluorescence data, carrying out equal longitude and latitude rasterization, linear interpolation and SG filtering reconstruction processing to obtain a solar induced chlorophyll fluorescence SIF data set with continuous space day by day in a research area;
(3) integrating the spatial distribution data of the winter wheat in the step (1) and the solar-induced chlorophyll fluorescence SIF data sets of the reconstructed study area in the step (2) in a spatial continuous day-by-day manner to construct a normalized chlorophyll fluorescence drought index NIBS;
(4) performing precision evaluation on the normalized chlorophyll fluorescent drought index NIBS by using the contemporaneous temperature vegetation drought index TVDI and the soil moisture field investigation data in the step (1);
(5) and dynamically monitoring spring drought in the research area based on the normalized chlorophyll fluorescence drought index NIBS index.
2. The method for monitoring the drought of winter wheat based on TROPOMI chlorophyll fluorescence remote sensing of claim 1, wherein in the step (1):
TROPOMI chlorophyll fluorescence data: a solar induced chlorophyll fluorescence SIF product data set inverted based on a Tropomi sensor carried by a sentinel-5p satellite;
spatial distribution data of winter wheat: by collecting domestic high-grade first data during the growing season of the winter wheat in 2017-2018, combining ground sample data of field investigation in 2018, obtaining spatial distribution data of the winter wheat by using a maximum likelihood classification method, and resampling a classification result to be 1 km;
soil moisture field survey data: in the period from 23 days in 3 months to 28 days in 3 months, field investigation on the soil moisture is carried out in two areas of the Hebei stone village, the Chenge platform, the Shandong Jining and the Neze, and a Time Domain Reflectometer (TDR) is adopted to observe the soil moisture content of different soil layers.
3. The method for monitoring the drought of the winter wheat based on TROPOMI chlorophyll fluorescence remote sensing of the claim 1, wherein in the step (2), the method for rasterizing TROPOMI chlorophyll fluorescence data by equal longitude and latitude is as follows: respectively establishing equal longitude and latitude grids with step lengths of 0.1 degree, 0.05 degree and 0.2 degree in the spatial ranges of 30-43 degrees of north latitude and 110-113 degrees of east longitude in a research area; through traversing longitude and latitude coordinates of the centers of sampling points of non-rasterized sun-induced chlorophyll fluorescence SIF signals, the mean value of SIF signal sequences in each grid is counted, and three solar-induced chlorophyll fluorescence SIF signals with spatial resolution continuously rasterized are obtained.
4. The method for monitoring the drought of the winter wheat based on TROPOMI chlorophyll fluorescence remote sensing of the claim 1, wherein in the step (2), the linear interpolation method comprises the following steps: by analyzing SIF sequence data on the same spatial position, assuming that the change of the sun-induced chlorophyll fluorescence SIF signal is in a linear trend in the signal loss period, and indexing the nearest effective value from the head to the tail of the loss interval to linearly interpolate the sun-induced chlorophyll fluorescence SIF sequence;
SG filtering reconstruction processing: smoothing the SIF sequence signal after linear interpolation by using a Savitzky-Golay algorithm, wherein the calculation method comprises the following steps:
Figure FDA0002330065630000021
in the formula (I), the compound is shown in the specification,
Figure FDA0002330065630000022
representing the reconstructed SIF sequence; y isj+iRepresenting the original SIF data after linear interpolation; ciRepresents a filter coefficient; m is the window width; n is the length of the filter and can be expressed as (2m + 1);
in the filtering process, two parameters need to be set: the window width m, the polynomial fitting times; the larger the window width m is, the smoother the filtered sequence data is; the degree of polynomial fit is set to 2-4, the smaller the degree, the smoother.
5. The method for monitoring the drought of the winter wheat based on TROPOMI chlorophyll fluorescence remote sensing of the claim 4, wherein in the step (3), the method comprises the following steps:
(3-1) a winter wheat crop distribution density calculation method;
and (2) combining the spatial distribution data of the winter wheat in the step (1) to construct a spatial distribution density map of the winter wheat: on the basis of a preprocessed 0.1-degree sun-induced chlorophyll fluorescence SIF data set, the total distribution area of the winter wheat in each equal longitude and latitude grid is statistically analyzed, a winter wheat distribution density graph is generated by taking the crop area as a grid numerical value, wherein the spatial range and the resolution of the equal longitude and latitude grid are consistent with those of a rasterized sun-induced chlorophyll fluorescence SIF signal, namely the north latitude is 30-43 degrees, the east longitude is 110-113 degrees, and the step length is 0.1 degree;
(3-2) construction of normalized chlorophyll fluorescent drought index NIBS
Selecting grids with the distribution density of all the winter wheat being more than 5 on the basis of the winter wheat crop distribution density diagram in the step (3-1), counting sun-induced chlorophyll fluorescence SIF fluorescence signals which are subjected to filtering reconstruction in the same spatial position in the step (2), calculating the maximum value and the minimum value of the grids meeting the statistical conditions, constructing and calculating drought indexes, and providing normalized chlorophyll fluorescence drought indexes NIBS according to the fluorescence signal statistical characteristic values of the winter wheat distribution areas in the research area at the same moment, wherein the specific calculation methods are respectively as follows:
Figure FDA0002330065630000031
in the formula, SIFminThe minimum value of SIF signal meeting the statistical condition of winter wheat distribution at a certain time phase in a research region, SIFmaxThe maximum value of the SIF signal meeting the statistical condition of the distribution of the winter wheat at a certain time phase in a research area.
6. The method for monitoring the drought condition of the winter wheat based on TROPOMI chlorophyll fluorescence remote sensing of the claim 5, wherein in the step (4), the temperature vegetation drought index TVDI integrates vegetation coverage information and land surface temperature information, and can reflect the surface drought condition; the temperature vegetation drought index is obtained by a dry-wet edge fitting equation calculated by normalizing the vegetation index NDVI and the surface temperature LST, and the specific calculation formula is as follows:
Figure FDA0002330065630000032
LSTNDVIimax＝a1+b1NDVIi
LSTNDVIimax＝a2+b2NDVIi
in the formula: LSTNDVIiIs the land surface temperature; LSTNDVIimaxRepresents the highest surface temperature represented by a certain NDVI value, namely a dry edge value; LSTNDVIiminRepresents the lowest surface temperature represented by a certain NDVI value, i.e., the wet edge value; a1, b1 and a2, b2 are coefficients of dry-edge and wet-edge fitting equations, respectively; NDVIi represents a certain normalized vegetation index.
7. The method for monitoring the drought status of winter wheat based on TROPOMI chlorophyll fluorescence remote sensing as claimed in claim 6, wherein the precision evaluation parameter of drought monitoring comprises a determination coefficient R2Root-Mean-Square Error (RMSE); the calculation methods of the decision coefficient and the root mean square error are respectively as follows:
Figure FDA0002330065630000041
Figure FDA0002330065630000042
in the formula:
Figure FDA0002330065630000043
respectively representing TVDI or soil moisture SM values and their corresponding mean values, yiIs an NIBS index value, and m is the number of sampling points;
wherein R is2For describing the correlation between the constructed NIBS index and TVDI and soil moisture, R2The larger, the higher its correlation; and the RMSE is used for describing the stability of the correlation between the NIBS index and the real soil moisture, and the smaller the RMSE value, the higher the stability.
8. The method for monitoring the drought of winter wheat based on TROPOMI chlorophyll fluorescence remote sensing of claim 7, wherein in the step (4): the normalized chlorophyll fluorescence index NIBS is related to the contemporaneous temperature vegetation drought index TVDI and the contemporaneous soil moisture field investigation data, and winter wheat spring drought monitoring can be monitored through the normalized chlorophyll fluorescence drought index NIBS.
9. The method for monitoring the drought of winter wheat based on TROPOMI chlorophyll fluorescence remote sensing of claim 8,
correlation of normalized chlorophyll fluorescence index NIBS with contemporaneous temperature vegetation drought index TVDI: the NIBS indexes of different subareas of the research area are in negative correlation with the contemporaneous remote sensing drought monitoring index TVDI, which shows that the NIBS indexes can have the same indication function of better separation degree as the TVDI in terms of the drought degree at the same moment, thereby proving the effectiveness of the NIBS index drought indication on the spatial distribution;
correlation analysis of the NIBS index with soil moisture: the NIBS index has a clear positive correlation with soil moisture SM.
CN201911332612.XA 2019-12-22 2019-12-22 Winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing Active CN113008843B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911332612.XA CN113008843B (en) 2019-12-22 2019-12-22 Winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911332612.XA CN113008843B (en) 2019-12-22 2019-12-22 Winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing

Publications (2)

Publication Number Publication Date
CN113008843A true CN113008843A (en) 2021-06-22
CN113008843B CN113008843B (en) 2022-02-18

Family

ID=76382910

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911332612.XA Active CN113008843B (en) 2019-12-22 2019-12-22 Winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing

Country Status (1)

Country Link
CN (1) CN113008843B (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113466193A (en) * 2021-06-23 2021-10-01 河南科技大学 Method for measuring relative water content of leaves
CN116124709A (en) * 2022-09-20 2023-05-16 中国水利水电科学研究院 Drought UAV Rapid Monitoring and Discrimination Method Based on Relative Content of Chlorophyll in Winter Wheat
CN116171840A (en) * 2023-01-18 2023-05-30 广州市林业和园林科学研究院 Method for regulating flowering phase of bougainvillea
CN116976516A (en) * 2023-08-01 2023-10-31 中国科学院空天信息创新研究院 Early prediction method for single crop yield in arid region
CN117173565A (en) * 2023-09-01 2023-12-05 中国科学院空天信息创新研究院 Method and system for estimating vegetation coverage of large-area wheat by deep migration learning
CN117593670A (en) * 2023-11-23 2024-02-23 日照朝力信息科技有限公司 Method and system for intelligent investigation and monitoring of forest resources
WO2024061159A1 (en) * 2022-09-20 2024-03-28 中国水利水电科学研究院 Method for rapidly monitoring and determining drought condition of summer corn on basis of chlorophyll content by using unmanned aerial vehicle
CN118883457A (en) * 2024-10-08 2024-11-01 山东大学 A drought remote sensing monitoring method and system integrating multi-source remote sensing data

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103994976A (en) * 2013-11-28 2014-08-20 江苏省水利科学研究院 MODIS data-based agricultural drought remote sensing monitoring method
CN105548113A (en) * 2015-12-31 2016-05-04 浙江大学 Plant physiology monitoring method based on chlorophyll fluorescence and multispectral image
KR20160056361A (en) * 2014-11-10 2016-05-20 한경대학교 산학협력단 Drought tolerance index derived from chlorophyll fluorescence parameters in Maize(Zea Mays L.)
CN107782701A (en) * 2017-09-20 2018-03-09 北京师范大学 A kind of agricultural arid monitoring method of multi- source Remote Sensing Data data
CN108693154A (en) * 2018-04-25 2018-10-23 南京大学 A kind of method of multi-angle observation exact inversion vegetation negative and positive leaf sunlight-induced chlorophyll fluorescence
CN109239027A (en) * 2018-07-19 2019-01-18 江苏大学 A kind of plant physiology water deficiency diagnosis method and system
CN109754120A (en) * 2019-01-08 2019-05-14 中国农业科学院农业资源与农业区划研究所 A Drought Early Warning Method Considering Fluorescence Effect
US20190174692A1 (en) * 2017-10-25 2019-06-13 Advanced Biological Marketing, Inc. Endophytic microbial seed treatment formulations and methods related thereto for improved plant performance
CN110188955A (en) * 2019-05-31 2019-08-30 西南大学 A large-scale autumn grain yield estimation method based on chlorophyll fluorescence remote sensing
CN110222870A (en) * 2019-05-05 2019-09-10 中国农业大学 Assimilate the Regional Fall Wheat yield estimation method of satellite fluorescence data and crop growth model
CN110267525A (en) * 2017-02-15 2019-09-20 索尼公司 Information generation method, information generating apparatus and program

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103994976A (en) * 2013-11-28 2014-08-20 江苏省水利科学研究院 MODIS data-based agricultural drought remote sensing monitoring method
KR20160056361A (en) * 2014-11-10 2016-05-20 한경대학교 산학협력단 Drought tolerance index derived from chlorophyll fluorescence parameters in Maize(Zea Mays L.)
CN105548113A (en) * 2015-12-31 2016-05-04 浙江大学 Plant physiology monitoring method based on chlorophyll fluorescence and multispectral image
CN110267525A (en) * 2017-02-15 2019-09-20 索尼公司 Information generation method, information generating apparatus and program
CN107782701A (en) * 2017-09-20 2018-03-09 北京师范大学 A kind of agricultural arid monitoring method of multi- source Remote Sensing Data data
US20190174692A1 (en) * 2017-10-25 2019-06-13 Advanced Biological Marketing, Inc. Endophytic microbial seed treatment formulations and methods related thereto for improved plant performance
CN108693154A (en) * 2018-04-25 2018-10-23 南京大学 A kind of method of multi-angle observation exact inversion vegetation negative and positive leaf sunlight-induced chlorophyll fluorescence
CN109239027A (en) * 2018-07-19 2019-01-18 江苏大学 A kind of plant physiology water deficiency diagnosis method and system
CN109754120A (en) * 2019-01-08 2019-05-14 中国农业科学院农业资源与农业区划研究所 A Drought Early Warning Method Considering Fluorescence Effect
CN110222870A (en) * 2019-05-05 2019-09-10 中国农业大学 Assimilate the Regional Fall Wheat yield estimation method of satellite fluorescence data and crop growth model
CN110188955A (en) * 2019-05-31 2019-08-30 西南大学 A large-scale autumn grain yield estimation method based on chlorophyll fluorescence remote sensing

Non-Patent Citations (8)

* Cited by examiner, † Cited by third party
Title
G. DUVEILLER, A. CESCATTI: "Spatially downscaling sun-induced chlorophyll fluorescence leads to an improved temporal correlation with gross primary productivity", 《REMOTE SENSING OF ENVIRONMENT》 *
H. WANG ET AL.: "Evaluation of potential crop productivity based on remote sensing and agro-ecological zones around the world", 《GEOCARTO INTERNATIONAL》 *
KÖEHLER ET AL.: "Global retrievals of solar induced chlorophyll fluorescence with TROPOMI: first results and inter-sensor comparison to OCO-2", 《GEOPHYS RES LETT.》 *
SONG ET AL.: "Satellite sun-induced chlorophyll fluorescence detects early response of winter wheat to heat stress in the Indian Indo-Gangetic Plains", 《GLOB CHANGE BIOL.》 *
Z. ZLATEV: "Drought-Induced Changes in Chlorophyll Fluorescence of Young Wheat Plants", 《BIOTECHNOLOGY & BIOTECHNOLOGICAL EQUIPMENT》 *
刘雷震等: "叶绿素荧光及其在水分胁迫监测中的研究进展", 《光谱学与光谱分析》 *
张立福等: "太阳诱导叶绿素荧光的卫星遥感反演方法", 《遥感学报》 *
陈雪洋等: "冬小麦叶面积指数的高光谱估算模型研究", 《测绘科学》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113466193A (en) * 2021-06-23 2021-10-01 河南科技大学 Method for measuring relative water content of leaves
CN116124709A (en) * 2022-09-20 2023-05-16 中国水利水电科学研究院 Drought UAV Rapid Monitoring and Discrimination Method Based on Relative Content of Chlorophyll in Winter Wheat
CN116124709B (en) * 2022-09-20 2023-09-12 中国水利水电科学研究院 Winter wheat drought unmanned aerial vehicle monitoring and distinguishing method based on chlorophyll relative content
WO2024061159A1 (en) * 2022-09-20 2024-03-28 中国水利水电科学研究院 Method for rapidly monitoring and determining drought condition of summer corn on basis of chlorophyll content by using unmanned aerial vehicle
CN116171840A (en) * 2023-01-18 2023-05-30 广州市林业和园林科学研究院 Method for regulating flowering phase of bougainvillea
CN116171840B (en) * 2023-01-18 2023-09-08 广州市林业和园林科学研究院 Method for regulating flowering phase of bougainvillea
CN116976516A (en) * 2023-08-01 2023-10-31 中国科学院空天信息创新研究院 Early prediction method for single crop yield in arid region
CN116976516B (en) * 2023-08-01 2024-03-15 中国科学院空天信息创新研究院 An early prediction method for crop yield in arid areas
CN117173565A (en) * 2023-09-01 2023-12-05 中国科学院空天信息创新研究院 Method and system for estimating vegetation coverage of large-area wheat by deep migration learning
CN117593670A (en) * 2023-11-23 2024-02-23 日照朝力信息科技有限公司 Method and system for intelligent investigation and monitoring of forest resources
CN117593670B (en) * 2023-11-23 2024-05-14 日照朝力信息科技有限公司 Method and system for intelligent investigation and monitoring of forest resources
CN118883457A (en) * 2024-10-08 2024-11-01 山东大学 A drought remote sensing monitoring method and system integrating multi-source remote sensing data

Also Published As

Publication number Publication date
CN113008843B (en) 2022-02-18

Similar Documents

Publication Publication Date Title
CN113008843B (en) Winter wheat drought monitoring method based on TROPOMI chlorophyll fluorescence remote sensing
CN110766299B (en) A Method for Analyzing Watershed Vegetation Change Based on Remote Sensing Data
CN101699315B (en) A monitoring device and method for crop growth uniformity
Wang et al. Summer maize growth under different precipitation years in the Huang-Huai-Hai Plain of China
PROJEÇÃO et al. Climate change projections for the Portuguese viticulture using a multi-model ensemble
CN109685081B (en) Combined change detection method for remote sensing extraction of abandoned land
Sakamoto et al. Detecting spatiotemporal changes of corn developmental stages in the US corn belt using MODIS WDRVI data
Lazoglou et al. Climate change projections for Greek viticulture as simulated by a regional climate model
JP2002360070A (en) Evaluation method of plant vitality
CN108548793B (en) Wheat canopy water content inversion method integrating Nir-Red-Swir spectral characteristics
CN117437538A (en) Tropical rainforest ecosystem space-time pattern feature extraction and prediction method
Zhou et al. Large discrepancies among remote sensing indices for characterizing vegetation growth dynamics in Nepal
CN115062863A (en) Apple flowering phase prediction method based on crop reference curve and accumulated temperature correction
CN106447079A (en) Prediction method for tobacco production of karst mountainous area based on Radarsat-2
Mohammad et al. Evaluating the agricultural drought for Pre-Kharif season in Bangladesh using MODIS Vegetation Health Index
Vanino et al. Earth observation for improving irrigation water management: A case-study from Apulia Region in Italy
Minobe et al. A 1° monthly gridded sea-surface temperature dataset compiled from ICOADS from 1850 to 2002 and Northern Hemisphere frontal variability
Yuan et al. Monitoring of Sugarcane Crop based on Time Series of Sentinel-1 data: A case study of Fusui, Guangxi
Li et al. An improved threshold method to detect the phenology of winter wheat
Hong-wei et al. The Modification of Difference Vegetation Index (DVI) in middle and late growing period of winter wheat and its application in soil moisture inversion
Olsen et al. A method to identify potential cold-climate vine growing sites—A case study from Røsnæs in Denmark
Li et al. Monitoring of sugarcane crop based on time series of sentinel-1 data
Dabrowska-Zielinska et al. Drought Assessment Applying Joined Meteorological and Satellite Data
Falagas et al. On the Corellation of Remote Estimated ET and Leaf Stomatal Conductance
Kumari et al. C-Band RISAT-1 data for crop growth assessment of rice

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20220125

Address after: No. 9 Dengzhuang South Road, Haidian District, Beijing 100094

Applicant after: Aerospace Information Research Institute,Chinese Academy of Sciences

Address before: 100101 Beijing Chaoyang District Andingmen Datun Road No. 20 North

Applicant before: Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences

GR01 Patent grant
GR01 Patent grant