Hongze lake total nitrogen inversion method based on Landsat 8 OLI

By constructing a TN remote sensing inversion model based on Landsat 8 OLI in the waters of Hongze Lake, the problem that the existing technology cannot accurately and widely monitor the total nitrogen concentration in Hongze Lake is solved, and high-precision and low-cost water TN monitoring is achieved. Suitable for other lakes.

CN118937236APending Publication Date: 2024-11-12HUAIYIN TEACHERS COLLEGE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411163915.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The existing total nitrogen concentration estimation method of Hongze Lake cannot be monitored in one go, accurately and extensively, and cannot meet the monitoring needs of water quality parameters.

Method used

The total nitrogen inversion method of Hongze Lake based on Landsat 8 OLI is adopted. By uniformly laying sampling points in the waters of Hongze Lake, the measured TN data is obtained, and the TN remote sensing inversion model is constructed using the water surface remote sensing reflectivity of Landsat 8 OLI to realize TN monitoring of spatiotemporal distribution.

Benefits of technology

It realizes one-time and large-scale TN monitoring, with the comprehensive advantages of short monitoring cycle, high accuracy and low cost, and is suitable for other lakes with small differences in climate levels and terrain.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118937236A_ABST
    Figure CN118937236A_ABST
Patent Text Reader

Abstract

The invention discloses a Hongze lake total nitrogen inversion method based on Landsat 8OLI. The method comprises the following steps: uniformly arranging sampling points in a water area of the Hongze lake, and obtaining actually measured TN data of the sampling points in different periods; carrying out satellite-ground matching on all the sampling points and Landsat 8OLI to obtain water surface remote sensing reflectivity of Landsat 8OLI related wavebands in different periods; proportionally dividing the matched point pairs into a modeling data set and a verification data set; constructing a TN remote sensing inversion model suitable for the Landsat 8OLI by using the water surface remote sensing reflectivity of the Landsat 8OLI related wave band of the modeling data set; and evaluating the model precision of the TN remote sensing inversion model by using the verification data set. According to the method, the Hongze lake TN is estimated in an inversion mode based on the Landsat 8OLI satellite image, the TN can be accurately monitored at a time in a large range, operation is easy and convenient, and timeliness is high.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of environmental science and remote sensing monitoring technology, and in particular to a total nitrogen inversion method of Hongze Lake based on Landsat 8 OLI. Background Art

[0002] The definition of total nitrogen (TN) in lakes and reservoirs is: the total concentration of ammonia nitrogen, nitrite nitrogen, nitrate nitrogen and organic nitrogen in the water. TN in lakes and reservoirs is a necessary nutrient salt for algae growth, which can promote the massive growth and reproduction of phytoplankton and organisms. It is an important indicator for evaluating the eutrophication of water bodies and measuring the quality of water. It has been widely used because it can effectively evaluate the eutrophication of water bodies. According to the definition of TN in lakes and reservoirs, TN does not have its own obvious absorption curve characteristics like chlorophyll a, but is jointly determined by various forms of inorganic and organic nitrogen in the water body, such as nitrate nitrogen, ammonia nitrogen, Kjeldahl nitrogen, etc. in the water body, which causes phytoplankton and organisms in the water body to reproduce in large numbers, quickly consume dissolved oxygen in the water, and deteriorate the water quality.

[0003] However, with the growth of population, the acceleration of urbanization, and the rapid development of industries such as industry, aquaculture, and animal husbandry, the ecological security of lakes and reservoirs is facing huge threats. The TN concentration in lakes and reservoirs is high, and the water quality is deteriorating. Therefore, it is urgent to realize the monitoring and evaluation of TN in lakes and reservoirs quickly, in real time, and on a large scale. The existing TN concentration estimation method of Hongze Lake cannot monitor TN in a one-time, accurate, and large-scale manner, and cannot meet the monitoring needs of TN as a water quality parameter. Therefore, it is necessary to propose a total nitrogen inversion method for Hongze Lake based on Landsat 8 OLI to solve the above problems. Summary of the invention

[0004] The present invention provides a total nitrogen inversion method for Hongze Lake based on Landsat 8 OLI, so as to solve the problem that the existing Hongze Lake TN concentration estimation method cannot monitor TN in a one-time, accurate and large-scale manner, and cannot meet the monitoring demand for TN, a water quality parameter of the water body.

[0005] The present invention provides a method for inverting total nitrogen in Hongze Lake based on Landsat 8 OLI, comprising:

[0006] Sampling points were evenly distributed in the waters of Hongze Lake to obtain the measured TN data of the sampling points at different periods;

[0007] All sampling points were matched with Landsat 8 OLI to obtain matching point pairs, and the water surface remote sensing reflectance of the relevant bands of Landsat 8 OLI at different periods was obtained;

[0008] Divide the matched point pairs into a modeling data group and a verification data group in proportion;

[0009] Using the data of the modeling data set and the water surface remote sensing reflectance of the relevant bands of Landsat 8 OLI, a TN remote sensing inversion model suitable for Landsat 8 OLI is constructed;

[0010] The model accuracy of the TN remote sensing inversion model is evaluated using the data of the verification data set.

[0011] Furthermore, sampling points were evenly distributed in the waters of Hongze Lake to obtain the measured TN data of the sampling points at different times, including:

[0012] The entire Hongze Lake was selected as the study area. Sampling points were evenly distributed when the weather was clear and cloudless, the water surface was calm, and the Landsat 8 OLI satellite was passing by. Water samples were collected within 0.5m of the surface of the lake. The total nitrogen concentration was determined by alkaline potassium persulfate digestion-ultraviolet spectrophotometry. TN data of different sampling points were collected, and the latitude and longitude information of the sampling points were recorded at the same time. The water samples were stored in a refrigerator at 4°C.

[0013] Furthermore, all sampling points are matched with Landsat 8 OLI to obtain matching point pairs, and the water surface remote sensing reflectance of relevant bands of Landsat 8 OLI at different periods is obtained, including:

[0014] According to the latitude and longitude information of all sampling points, Landsat 8 OLI images that passed within 48 hours were obtained. After resampling and atmospheric correction preprocessing, the water surface remote sensing reflectance in the visible light-near infrared band corresponding to the lake water sampling points was extracted, and abnormal invalid sampling points were screened out and removed, and a valid Landsat image array was made with the matching TN data.

[0015] Furthermore, the matched point pairs are divided into a modeling data group and a verification data group in proportion, including:

[0016] Four fifths of the Landsat image valid array data are divided into modeling data groups for building a TN remote sensing inversion model suitable for Landsat 8 OLI;

[0017] The remaining one-fifth of the data in the effective array of Landsat images is divided into a verification data group for verifying the accuracy of the TN remote sensing inversion model applicable to Landsat 8 OLI, and ensuring that the data are evenly distributed in the water area of ​​Hongze Lake and contain sampling points of different periods.

[0018] Furthermore, a TN remote sensing inversion model suitable for Landsat 8 OLI is constructed by using the data of the modeling data set and the water surface remote sensing reflectance of the relevant bands of Landsat 8 OLI, including:

[0019] With TN data as the dependent variable, the band ratios obtained by multiplying and dividing the single-band and multi-band data of different bands of Landsat 8 OLI were used as independent variables, and correlation analysis was performed in turn;

[0020] Select the single band or band ratio with the largest correlation coefficient as the pre-selected TN inversion model;

[0021] The pre-selected TN inversion models were applied to the validation data set respectively;

[0022] The TN estimation model with the best evaluation result was selected to obtain the spatiotemporal distribution of TN in Hongze Lake and the inversion model of TN in the water area of ​​Hongze Lake.

[0023] The present invention has the following beneficial effects: a total nitrogen inversion method for Hongze Lake based on Landsat 8 OLI of the present invention can obtain TN of the entire study area at one time and on a large scale by studying TN through remote sensing technology, and has comprehensive comparative advantages such as short monitoring period, high overall accuracy, low cost, etc., and can be completed by a notebook or an ordinary computer, the operation is relatively simple, and historical data can be obtained; the present invention obtains real-time Landsat 8 OLI satellite image data of Hongze Lake and constructs a TN inversion model of Hongze Lake water body based on the reflectivity of the satellite image and the measured TN data, and has strong applicability to other lakes with smaller climate levels and terrain differences. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 This is the distribution map of sampling points in Hongze Lake;

[0026] Figure 2 This is the inversion model map of Hongze Lake TN based on Landsat 8 OLI satellite image;

[0027] Figure 3 This is the accuracy evaluation map of Hongze Lake TN constructed based on Landsat 8 OLI satellite images;

[0028] Figure 4 This is the spatial and temporal distribution map of TN in Hongze Lake;

[0029] Figure 5 This is a correlation analysis diagram for single band and band combination. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0031] The embodiment of the present invention provides a method for inverting total nitrogen in Hongze Lake based on Landsat 8 OLI, comprising:

[0032] Step 1: Evenly distribute sampling points in the waters of Hongze Lake and obtain the measured TN data of the sampling points at different periods.

[0033] Specifically, the entire Hongze Lake was selected as the study area. Sampling points were evenly distributed when the weather was clear and cloudless, the water surface was calm, and the Landsat 8 OLI satellite was passing by. Water samples were collected within 0.5 m of the lake surface. The total nitrogen concentration was determined by alkaline potassium persulfate digestion-ultraviolet spectrophotometry. TN data of different sampling points were collected, and the latitude and longitude information of the sampling points were recorded at the same time. The water samples were stored in a refrigerator at 4°C.

[0034] Step 2: All sampling points are matched with Landsat 8 OLI to obtain matched point pairs, and the water surface remote sensing reflectivity of the relevant bands of Landsat 8 OLI at different periods is obtained.

[0035] Specifically, according to the longitude and latitude information of all sampling points, Landsat 8 OLI images that passed within 48 hours were obtained. After resampling and atmospheric correction preprocessing, the water surface remote sensing reflectance in the visible light-near infrared band corresponding to the lake water sampling points was extracted, and abnormal invalid sampling points were screened out and removed, and a valid Landsat image array was made with the matching TN data.

[0036] Step three: divide the matched point pairs into a modeling data group and a verification data group in proportion.

[0037] Specifically, four-fifths of the data in the effective array of Landsat images are divided into a modeling data group for building a TN remote sensing inversion model suitable for Landsat 8 OLI; the remaining one-fifth of the data in the effective array of Landsat images is divided into a verification data group for verifying the accuracy of the TN remote sensing inversion model suitable for Landsat 8 OLI, and ensuring that the data are evenly distributed in the water area of ​​Hongze Lake and that the data contain sampling points from different periods.

[0038] Step 4: Using the data of the modeling data set and the water surface remote sensing reflectance of the relevant bands of Landsat 8 OLI, a TN remote sensing inversion model suitable for Landsat 8 OLI is constructed.

[0039] Specifically, TN data was taken as the dependent variable, and the band ratios obtained by multiplying and dividing the single-band and multi-band data of different bands of Landsat 8 OLI were used as independent variables, and correlation analysis was performed respectively; the single band or band ratio with the largest correlation coefficient was selected as the pre-selected TN inversion model; the pre-selected TN inversion models were applied to the verification data group respectively; the TN estimation model with the best evaluation result was selected, so as to obtain the spatiotemporal distribution of TN in Hongze Lake and the inversion model of TN in the water area of ​​Hongze Lake.

[0040] Step five: using the data of the verification data set to evaluate the model accuracy of the TN remote sensing inversion model.

[0041] The following describes in detail the total nitrogen inversion method of Hongze Lake based on Landsat 8 OLI of the present invention, taking Hongze Lake as the study area.

[0042] See also Figures 1 to 5 The present invention selects Hongze Lake as the study area, and conducts sampling during the non-ice period from 2018 to 2021 when the weather is clear and cloudless and the water surface is calm. The sampling points are evenly arranged, and the total nitrogen concentration is determined by alkaline potassium persulfate digestion-ultraviolet spectrophotometry. TN at different points is collected, and the latitude and longitude information of the sampling points is recorded at the same time. The water samples are stored in a refrigerator at 4°C, and a total of 138 valid sampling points are obtained in all samplings.

[0043] According to the latitude and longitude information of the sampling points obtained from all field sampling measurements, the Landsat8 OLI images that passed within 48 hours were obtained and preprocessed with resampling and atmospheric correction, and the water surface remote sensing reflectivity in the visible light-near infrared band corresponding to the lake water sampling points was extracted, and the abnormal invalid sampling points were screened and removed, and a Landsat valid array was made with the matching TN. The remote sensing image data source used in the present invention is the OLI image of the Landsat satellite. The OLI image data in the Landsat series sensors used in the present invention is the land remote sensing reflectivity product released by the United States Geological Survey (USGS), and the product data uses the LEDAPS method to perform atmospheric correction on the original data provided by Landsat; the data can be downloaded from the US GS official website, and can also be accessed and obtained through the Google Earth Engine website.

[0044] Four-fifths of the valid array sample points made by Landsat are used to build the TN inversion model of Hongze Lake, and the remaining one-fifth is used to verify the accuracy of the TN inversion model of Hongze Lake, and the data are evenly distributed in the water area of ​​Hongze Lake, and the data contain sampling points of different periods. Among the 138 valid TN sample points obtained, there are 78 valid sample points that can be matched to the Landsat 8 OLI satellite image passing within the time window of 48 hours, of which 62 point pairs are used for model construction, and the other 16 independent point pairs are used for model verification.

[0045] The estimation model of TN based on Landsat 8 OLI satellite images was obtained: TN was used as the dependent variable, and the band ratios obtained by multiplying and dividing the single-band and multi-band data of different bands of Landsat 8OLI were used as the independent variables. Correlation analysis was performed in turn, and the single band or band ratio with the largest correlation coefficient was selected as the pre-selected TN inversion model. The single bands are B2, B3, B4, and B5, and the band combinations are B2 / B3, B2 / B4, B2 / B5, B3 / B4, B3 / B5, and B4 / B5. Among them, the correlation coefficient at B2 / B3 is the highest, which is 0.83. The evaluation of the correlation results of single bands and band combinations is shown in Figure 5 Then, the pre-selected models were applied to the validation data set, and the TN estimation model with the best evaluation results was selected to obtain the spatiotemporal distribution of TN in Hongze Lake, and to obtain the inversion model of TN in the water area of ​​Hongze Lake. The final TN estimation model based on Landsat8 OLI satellite images is shown in the formula:

[0046] TN=4.4033×OLI(Blue / Green)^(2.7569)

[0047] In the formula, Blue and Green are the blue band and green band of Landsat remote sensing satellite images, respectively, corresponding to B2 (Blue) and B3 (Green) of the OLI sensor. The specific model construction and estimation effect evaluation can be found in Figure 2 .

[0048] The above only expresses the preferred embodiments of the present invention, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, and these modifications and improvements are all within the scope of protection of the present invention.

Claims

1. A method for inverting total nitrogen in Hongze Lake based on Landsat 8OLI, characterized in that: include: Sampling points were evenly distributed in the waters of Hongze Lake to obtain the measured TN data of the sampling points at different periods; All sampling points were matched with Landsat 8OLI to obtain matching point pairs, and the water surface remote sensing reflectivity of the relevant bands of Landsat 8OLI at different periods was obtained; Divide the matched point pairs into a modeling data group and a verification data group in proportion; Using the data of the modeling data group and the water surface remote sensing reflectance of the relevant bands of Landsat 8OLI, a TN remote sensing inversion model suitable for Landsat 8OLI is constructed; The model accuracy of the TN remote sensing inversion model is evaluated using the data of the verification data set.

2. The method for inverting total nitrogen in Hongze Lake based on Landsat 8OLI according to claim 1, characterized in that: Sampling points were evenly distributed in the waters of Hongze Lake to obtain the measured TN data of the sampling points at different times, including: The entire Hongze Lake was selected as the study area. Sampling points were evenly distributed when the weather was clear and cloudless, the water surface was calm, and the Landsat 8OLI satellite was passing by. Water samples were collected within 0.5m of the surface of the lake. The total nitrogen concentration was determined by alkaline potassium persulfate digestion-ultraviolet spectrophotometry. TN data of different sampling points were collected, and the latitude and longitude information of the sampling points were recorded at the same time. The water samples were stored in a refrigerator at 4°C.

3. The method for inverting total nitrogen in Hongze Lake based on Landsat 8OLI according to claim 1, characterized in that: All sampling points are matched with Landsat 8OLI to obtain matching point pairs, and the water surface remote sensing reflectivity of Landsat 8OLI related bands at different periods is obtained, including: According to the longitude and latitude information of all sampling points, Landsat 8OLI images that passed within 48 hours were obtained. After resampling and atmospheric correction preprocessing, the water surface remote sensing reflectance in the visible light-near infrared band corresponding to the lake water sampling points was extracted, and abnormal invalid sampling points were screened out and removed, and a valid Landsat image array was made with the matching TN data.

4. The method for inverting total nitrogen in Hongze Lake based on Landsat 8OLI according to claim 1, characterized in that: Divide the matched point pairs into modeling data group and verification data group in proportion, including: Four fifths of the Landsat image valid array data are divided into modeling data groups for building a TN remote sensing inversion model suitable for Landsat 8OLI; The remaining one-fifth of the data in the effective array of Landsat images is divided into a verification data group for verifying the accuracy of the TN remote sensing inversion model applicable to Landsat 8OLI, and ensuring that the data are evenly distributed in the water area of ​​Hongze Lake and that the data contain sampling points of different periods.

5. The method for inverting total nitrogen in Hongze Lake based on Landsat 8OLI according to claim 1, characterized in that: Using the data of the modeling data group and the water surface remote sensing reflectance of the relevant bands of Landsat 8OLI, a TN remote sensing inversion model suitable for Landsat 8OLI is constructed, including: With TN data as the dependent variable, the band ratios obtained by multiplying and dividing the single-band and multi-band data of different bands of Landsat 8OLI were used as independent variables, and correlation analysis was performed in turn; Select the single band or band ratio with the largest correlation coefficient as the pre-selected TN inversion model; The pre-selected TN inversion models were applied to the validation data set respectively; The TN estimation model with the best evaluation result was selected to obtain the spatiotemporal distribution of TN in Hongze Lake and the inversion model of TN in the water area of ​​Hongze Lake.