Method for monitoring vegetation drought based on sunlight-induced chlorophyll fluorescence

Abstract

The application provides a vegetation drought monitoring method based on sunlight-induced chlorophyll fluorescence, and steps are as follows: chlorophyll fluorescence, precipitation and surface temperature and other data are pretreated and spatially extracted; the obtained chlorophyll fluorescence, precipitation and surface temperature data are normalized to obtain a precipitation condition index PCI, a temperature condition index TCI and a fluorescence monitoring drought index DFMI; three-dimensional space model construction is carried out by using the PCI, the TCI and the DFMI, the wettest point and the driest point are determined; the Euclidean distance method is used to calculate the distance from any point in the three-dimensional space to the wettest point as a vegetation drought monitoring index, and vegetation drought monitoring is carried out. The application comprehensively considers three factors of precipitation, surface temperature and chlorophyll fluorescence, can overcome the defects of the method for constructing a drought index based on a vegetation index, and effectively reduces the response time of vegetation to drought.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vegetation drought monitoring, in particular to a vegetation drought monitoring method based on sunlight-induced chlorophyll fluorescence. BACKGROUND

[0002] Drought is one of the most common natural disasters in the world, and also one of the natural disasters with the largest social impact and economic loss in the world, which has the characteristics of high occurrence frequency, long action time, wide influence range and large loss. The numerous characteristics of drought make the planning and implementation of drought resistance plan difficult. In order to achieve active drought resistance, the description of drought characteristics is crucial, and the method of using drought index helps to describe the characteristics of drought.

[0003] Currently, the construction methods of drought indices can be divided into two categories: single variable-based drought index construction method and multi-variable-based drought index construction method. The single variable-based drought index construction method is simple, such as PCI, TCI, VCI, DFMI, etc., but the influencing factors of drought are complex and uncertain, so the single variable-based drought index cannot fully describe the characteristics of drought. Multi-variable drought indices, such as TVPDI, TVDI, VHI, etc., can integrate multi-variable information and effectively explain the abnormality of drought-related environment, but the multi-variable drought index construction method is usually based on vegetation index method. The vegetation index based on reflectivity represents the "greenness" of vegetation, and when the leaf chlorophyll content of the vegetation canopy reaches a certain threshold, it will cause the saturation phenomenon of the reflection signal. In addition, some photosynthetic activities cannot cause changes in vegetation reflectivity in a short time, such as changes in stomatal conductance and thermal dissipation proportion adjustment. Therefore, the vegetation index cannot timely and accurately reflect the changes of vegetation under drought stress. Chlorophyll fluorescence is different from the traditional vegetation index based on reflectivity. It is regulated by the absorbed photosynthetically active radiation (APAR) in the process of photosynthesis of plants, which undergoes three pathways: driving photochemistry, dissipating heat, and re-emitting unused parts of energy in the form of fluorescence. If drought causes changes in these functions, it will inevitably lead to changes in SIF, thereby reducing photosynthesis and fluorescence yield. The current drought index based on chlorophyll fluorescence mainly considers the physiological effects of vegetation and the influence of external temperature on vegetation. In 2020, the literature [Z. Zhang, W. Xu, Q. Qin, et al. Downscaling solar-induced chlorophyll fluorescence based on convolutional neural network method to monitor agricultural drought[J]. IEEE T. Geosci. Remote 2020: 1-17.] constructed the temperature fluorescence drought index TFDI through the triangular feature space. In 2021, the literature [Y. Liu, C. Y. Dang, H. Yue, et al. Enhanced drought detection and monitoring using sun-induced chlorophyll fluorescence over Hulun Buir Grassland, China[J]. Science of The Total Environment, 2021, 770.] constructed the SIF health index SHI based on the principle of VHI, both of which only consider chlorophyll fluorescence and land surface temperature. SUMMARY

[0004] In view of the deficiencies in the above background art, the present application proposes a vegetation drought monitoring method based on sunlight-induced chlorophyll fluorescence, which comprehensively considers three factors of precipitation, surface temperature and chlorophyll fluorescence, can overcome the defects of the method of constructing drought index based on vegetation index, and effectively reduces the response time to drought.

[0005] The technical solution of the present application is as follows:

[0006] A vegetation drought monitoring method based on sunlight-induced chlorophyll fluorescence, comprising the following steps:

[0007] Step 1: Preprocess chlorophyll fluorescence data, precipitation data and surface temperature data respectively;

[0008] Step 2: Mask extraction and integration of the preprocessed chlorophyll fluorescence data, precipitation data and surface temperature data into the research area range;

[0009] Step 3: Normalize the chlorophyll fluorescence data, precipitation data and surface temperature data to obtain precipitation condition index PCI, temperature condition index TCI and fluorescence monitoring drought index DFMI;

[0010] Step 4: Construct a three-dimensional space model with PCI, TCI and DFMI as coordinate axes, and determine the wettest point and the driest point;

[0011] Step 5: Calculate the distance from any point in the three-dimensional space to the wettest point by using the Euclidean distance method to obtain the temperature fluorescence precipitation index TFPDI.

[0012] The chlorophyll fluorescence data adopts Dr. Xiao's global ecological group monthly Gosif data, with a spatial resolution of 0.05°; the precipitation data is from the Goddard Earth Sciences Data and Information Services Center, with a spatial resolution of 0.1°; and the surface temperature data set is from MOD11B3 provided by NASA, with a spatial resolution of 1km.

[0013] The data preprocessing method is: reprojecting, resampling, removing outliers and multiplying by a factor of 0.0001 for the chlorophyll fluorescence data; converting the netcdf format precipitation data to tif format, then reprojecting, resampling and calculating the monthly average; using the MRT tool provided by Modis to splice the surface temperature data, then reprojecting, resampling, multiplying by a factor of 0.02, and then subtracting 273.15 to convert the unit to Celsius.

[0014] The method for normalizing the chlorophyll fluorescence data, precipitation data and surface temperature data respectively is as follows:

[0015]

[0016]

[0017]

[0018] In the formula, GPM i is the value of each pixel in GPM in the i-th year of the data year, LST i is the value of each pixel in LST in the i-th year of the data year, SIF i is the value of each pixel in SIF in the i-th year of the data year; GPM max is the maximum value of each pixel in GPM in the data year, GPM min is the minimum value of each pixel in GPM in the data year; LST max is the maximum value of each pixel in LST in the data year, LST min is the minimum value of each pixel in LST in the data year; SIF max is the maximum value of each pixel in SIF in the data year, SIF min is the minimum value of each pixel in SIF in the data year.

[0019] The wettest point is (PCI max , TCI max , DFMI max ), which corresponds to (1, 1, 1) in the three-dimensional space model;

[0020] The driest point is (PCI min , TCI min , DFMI min ), which corresponds to (0, 0, 0) in the three-dimensional space model;

[0021] The drought monitoring index is:

[0022]

[0023] Wherein, (PCI, TCI, DFMI) is any point in the three-dimensional space, and TFPDI is the drought monitoring index; the greater the value of TFPDI, the greater the distance of the point (PCI, TCI, DFMI) from the wettest point in the three-dimensional space, and the higher the drought degree.

[0024] Compared with the prior art, the present application has the beneficial effects that:

[0025] 1) The TFPDI index established by the present application comprehensively integrates data such as land surface temperature, precipitation and chlorophyll fluorescence, and can more fully describe the characteristics of drought compared with drought indices based on a single variable, and overcomes the defects that the reflectivity of the vegetation index based on reflectivity is saturated when the chlorophyll content of the vegetation canopy is high, and that the change in stomatal conductance and other photosynthetic activities cannot cause changes in vegetation reflectivity in a short time.

[0026] 2) The TFPDI index is constructed by introducing the Euclidean space distance principle, which can comprehensively reflect the vegetation drought conditions in the region, and has important significance for vegetation drought monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0028] Figure 1 The three-dimensional space model constructed by the present application.

[0029] Figure 2 The correlation between TFPDI and other drought indices in the Yellow River Basin from 2001 to 2020 of the present application.

[0030] Figure 3 The correlation and significance level spatial distribution of SM and TFPDI from April to October in 2001-2020 of the present application; in the figure, a is the correlation, and b is the significance. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0032] The embodiment of the present application provides a vegetation drought monitoring method based on sunlight-induced chlorophyll fluorescence, and the specific steps are as follows:

[0033] Step one: preprocess the chlorophyll fluorescence data, precipitation data and land surface temperature data respectively; among them, the chlorophyll fluorescence data adopts Dr. Xiao's global ecological group monthly Gosif data, the spatial resolution is 0.05°; the monthly precipitation data set is from the Goddard Earth Sciences Data and Information Services Center, the spatial resolution is 0.1°; the 8-day scale land surface temperature data set is from MOD11B3 provided by NASA, the spatial resolution is 1km.

[0034] The data preprocessing method is: the chlorophyll fluorescence data is re-projected, re-sampled, abnormal values are removed, multiplied by a factor of 0.0001; the netcdf format precipitation data is converted to tif format, then re-projected, re-sampled, the monthly average is calculated; after using the MRT tool provided by Modis to splice the land surface temperature data, re-project, re-sample, multiply by a factor of 0.02, then subtract 273.15, convert the unit to Celsius.

[0035] Step two: the preprocessed chlorophyll fluorescence data, precipitation data and land surface temperature data are respectively extracted by mask and integrated into the research area range; since the chlorophyll fluorescence data reflects the physiological condition of vegetation, the mask extraction can eliminate the lakes in the research area.

[0036] Step three: the chlorophyll fluorescence data, precipitation data and land surface temperature data are normalized respectively to obtain precipitation condition index PCI, temperature condition index TCI and fluorescence monitoring drought index DFMI; the land surface temperature data is processed in reverse normalization, and the chlorophyll fluorescence and precipitation data are processed in forward normalization; the data after normalization is closer to 1, which is more humid.

[0037] The normalization formula is respectively:

[0038]

[0039]

[0040]

[0041] In the formula, GPM i is the value of each pixel in GPM in the i-th year of the data year, LST i is the value of each pixel in LST in the i-th year of the data year, SIF i is the value of each pixel in SIF in the i-th year of the data year; GPM max is the maximum value of each pixel of GPM in the data year, GPM min is the minimum value of each pixel of GPM in the data year; LST max is the maximum value of each pixel of LST in the data year, LST minis the minimum value of each pixel of LST in the month of the data year limit; SIF max is the maximum value of each pixel of SIF in the month of the data year limit, SIF min is the minimum value of each pixel of SIF in the month of the data year limit. GPM is precipitation data of a certain data source, LST is land surface temperature data, and SIF is chlorophyll fluorescence data.

[0042] Step four: construct a three-dimensional space model with PCI, TCI and DFMI as X, Y and Z coordinate axes respectively, as shown in Figure 1 , and determine the wettest point and the driest point; the wettest point is the value of PCI max , TCI max and DFMI max under the wettest condition, corresponding to (1, 1, 1) in the three-dimensional space model; the driest point is (PCI min , TCI min , DFMI min ), corresponding to (0, 0, 0) in the three-dimensional space model. The line between the wettest point and the driest point is the wet-dry edge.

[0043] Step five: calculate the distance from any point in the three-dimensional space to the wettest point using the Euclidean distance method to obtain the drought monitoring index.

[0044] The drought monitoring index is:

[0045]

[0046] wherein (PCI, TCI, DFMI) is any point in the three-dimensional space, and TFPDI is the drought monitoring index; the greater the value of TFPDI, the greater the distance of the point (PCI, TCI, DFMI) from the wettest point in the three-dimensional space, and the higher the degree of drought.

[0047] The correlation analysis between TFPDI and PCI, TCI, VCI, scPDSI, SPEI and DFMI indexes is shown in Figure 2 , and the correlation is about -0.868, -0.590, -0.480, -0.567, -0.739 and -0.647 respectively, wherein the correlation between TFPDI and VCI is the worst. This is because VCI is an index constructed by a single variable NDVI, which reflects the growth condition of vegetation under drought stress through vegetation reflectivity. In the non-growing season, the correlation between TFPDI and VCI is relatively poor.

[0048] As shown in Figure 3As shown, soil moisture can directly reflect the surface dry and wet conditions, and thus can be used to verify the effectiveness of the drought index. After screening, the soil moisture data from FLDAS and TFPDI were selected for monthly correlation analysis. In April and September, the correlation was less than-0.5 in less area, mainly in the eastern part of the Ordos Plateau, the northeastern part of the Loess Plateau, and some areas in the western part of the Loess Plateau; in September, it was mainly in the Inner Mongolia Plateau and the northern part of the Ordos Plateau; the area ratio of the correlation less than-0.5 in May-August and October was more than 60%, covering the Loess Plateau, the Qinghai-Tibet Plateau and the Ordos Plateau, and passing the 0.05 significance level. The correlation and significance of the monthly soil moisture SM and TFPDI index show that the TFPDI index can well reflect the change of soil moisture in space and time.

[0049] The above merely describes preferred embodiments of the present application and is not used to limit the present application, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for monitoring vegetation drought based on sunlight-induced chlorophyll fluorescence, characterized in that, The steps are as follows: Step 1: Preprocess the chlorophyll fluorescence data, precipitation data, and surface temperature data respectively; the chlorophyll fluorescence data uses monthly Gosif data with a spatial resolution of 0.05°; the precipitation data has a spatial resolution of 0.1°. The surface temperature dataset is MOD11B3 with a spatial resolution of 1 km. Step 2: Mask the preprocessed chlorophyll fluorescence data, precipitation data, and surface temperature data and integrate them into the study area. Step 3: Normalize the chlorophyll fluorescence data, precipitation data, and surface temperature data respectively to obtain the precipitation condition index (PCI), temperature condition index (TCI), and fluorescence monitoring drought index (DFMI). Among them, GPM i Let GPM be the value of each pixel in a given month of year i within the data period. max GPM is the maximum value per pixel in the current month within the data period. min This represents the minimum GPM per pixel for the current month within the data period. Step 4: Construct a 3D spatial model using PCI, TCI, and DFMI as coordinate axes, and determine the wettest and driest points; Step 5: Calculate the distance from any point in three-dimensional space to the wettest point using the Euclidean distance method to obtain the Temperature Fluorescence Precipitation Index (TFPDI); Where (PCI, TCI, DFMI) is any point in three-dimensional space, (PCI... max TCI max DFMI max The point (PCI, TCI, DFMI) is the wettest point. The larger the TFPDI value, the greater the distance between the point (PCI, TCI, DFMI) and the wettest point in three-dimensional space, and the higher the degree of drought.

2. The vegetation drought monitoring method based on sunlight-induced chlorophyll fluorescence according to claim 1, characterized in that, The data preprocessing methods are as follows: chlorophyll fluorescence data are reprojected, resampled, outliers are removed, and multiplied by a factor of 0.0001; precipitation data in netcdf format is converted to tif format, then reprojected, resampled, and the monthly average is calculated; surface temperature data is stitched together using the MRT tool provided by Modis, then reprojected, resampled, multiplied by a factor of 0.02, and then 273.15 is subtracted to convert the units to degrees Celsius.

3. The vegetation drought monitoring method based on sunlight-induced chlorophyll fluorescence according to claim 1, characterized in that, The method for normalizing chlorophyll fluorescence data, precipitation data, and land surface temperature data respectively is as follows: In the formula, LST i SIF represents the value of each pixel in the LST of a certain month in the i-th year within the data period. i LST represents the value of each pixel in the SIF for a given month of the i-th year within the data period. max The maximum value of LST for each cell in the current month within the data period. min The minimum LST value for each cell in the current month within the data period; SIF max The SIF value for each cell within the current month of the data period. min This represents the minimum SIF value for each pixel in the current month within the data period.

4. The vegetation drought monitoring method based on sunlight-induced chlorophyll fluorescence according to claim 1, characterized in that, The wettest point is (PCI) max TCI max DFMI max ), which corresponds to (1, 1, 1) in the three-dimensional space model; The driest point is (PCI) min TCI min DFMI min ), which corresponds to (0, 0, 0) in the three-dimensional space model.

Images