Analysis Method, System and Storage Medium for Reconstructing Paleoclimate of the Cenozoic Era

Abstract

The present invention provides an analysis method, system and storage medium for reconstructing paleoclimate in the Cenozoic era, belonging to the technical field of past global change analysis, and comprising the following steps: Step 1, identification and statistics of detrital minerals quartz, feldspar and rock debris on a sedimentary section; Step 2, obtaining the values of F / Qt and Qm / Qp and the corresponding depths on the sedimentary section, and obtaining the change curves of F / Qt and Qm / Qp with depth on the sedimentary section; Step 3, averaging the original data of the F / Qt and Qm / Qp values on the sedimentary section to obtain their respective total average values Of and Oq, and respectively calculating the average values mf and nf of the parts below Of and above Of, as well as the average values mq and nq of the parts below Oq and above Oq; Through the above method, it can be seen that the present invention summarizes the climate types in the sedimentary evolution process, further establishes a climate model, and serves for predicting the trend of climate change, having the advantages of economy, convenience, high efficiency and science.

Description

Technical Field

[0001] The present invention belongs to the technical field of past global change analysis, and particularly relates to an analysis method, system and storage medium for reconstructing Cenozoic paleoclimate by using detrital minerals in sediments. Background Art

[0002] Using paleoclimate proxy indicators in sedimentary strata to reconstruct the past paleoclimate change process is an important prerequisite for understanding the laws and mechanisms of climate change and predicting future climate change trends.

[0003] Paleoclimate proxy indicators such as fossils, geochemistry, and geophysics in sedimentary strata have played an important role in paleoclimate reconstruction. However, the horizons of sedimentary strata containing fossils are limited, and it is difficult to obtain continuous paleoclimate change information; the testing costs of geochemical and geophysical indicators are high, increasing the difficulty of obtaining continuous paleoclimate information. Feldspar and quartz are common minerals in terrigenous clastic sedimentary rocks, which are easy to obtain and identify. They can not only easily solve the problem of barren strata, but also take less time, cost less, and have high credibility. In the 1970s, the famous sedimentologist Pettijohn proposed that the quartz / feldspar ratio (Q / F) in sedimentary debris can reflect the changes of paleoclimate to a certain extent. Many subsequent studies have applied Q / F to paleoclimate reconstruction research.

[0004] Theoretically, under certain provenance and tectonic backgrounds, the differences in the detrital components of terrigenous sedimentary rocks are often controlled by the changes in weathering related to warm and humid - dry and cold. However, due to the constraints of sedimentary minerals by various factors, how to effectively extract and interpret the paleoclimate information contained therein has always been a long - standing problem that plagues researchers. Currently, the paleoclimate information is mainly judged by directly extracting the contents of feldspar and quartz and the feldspar / quartz ratio. This results in subjective tendencies or objective uncertainties in the interpretation of the paleoclimate process, causing deviations in the self - consistency of research results and poor consistency with other evidence.

[0005] Therefore, the problems and defects of the existing technology are: the self - consistency of the analysis results of the existing technology and the consistency with other evidence are poor; and it is only a direct analysis of the original data of the contents of feldspar and quartz and the feldspar / quartz ratio, without clear and feasible methods and ways to apply them to climate analysis and climate change prediction; a clear and feasible method for completely interpreting the paleoclimate information of sedimentary minerals is urgently needed to be clarified.

[0006] Based on this, the present invention provides an analysis method, system and storage medium for reconstructing Cenozoic paleoclimate to solve the above - mentioned technical problems. Summary of the Invention

[0007] To solve the above - mentioned existing problems, the present invention provides an analysis method, system and storage medium for reconstructing Cenozoic paleoclimate.

[0008] The present invention provides an analysis method for reconstructing the paleoclimate of the Cenozoic era, including the following steps:

[0009] Step 1: Identification and statistics of detrital minerals quartz, feldspar, and rock fragments on the sedimentary section;

[0010] Step 2: Obtain the values of F / Qt and Qm / Qp and their corresponding depths on the sedimentary section, and obtain the variation curves of F / Qt and Qm / Qp with depth;

[0011] F / Qt is the total amount of feldspar / total amount of quartz, and Qm / Qp is the total amount of single-crystal quartz / total amount of polycrystalline quartz;

[0012] Step 3: Average the original data of the F / Qt and Qm / Qp values on the sedimentary section to obtain their respective total average values O f and O q , and calculate the average values m f and n f of the parts below O f and above O f , respectively, as well as the average values m q and n q of the parts below O q and above O q ;

[0013] Step 4: Use wavelet transform of Matlab software to perform spectral analysis on the F / Qt value and Qm / Qp data, obtain spectrograms at different scales, divide the variation characteristics of high, medium, and low frequencies according to the mirror symmetry characteristics of the F / Qt and Qm / Qp spectrograms, and then use the spectrograms obtained by wavelet transform to perform stage division on the parameters of the F / Qt value and Qm / Qp value, and obtain the stage characteristics of the parameters in the high, medium, and low frequency bands;

[0014] Step 5: Use the Savitzky-Golay method to perform denoising filtering on the original data of the F / Qt value and Qm / Qp value with adjacent several points as the window, remove the high-frequency part, and obtain the medium and low frequency variation trend lines of the F / Qt value and Qm / Qp value;

[0015] Step 6: Use the CONISS function to perform time series clustering analysis on the original data of the F / Qt value and Qm / Qp value, subdivide the parameters on the sedimentary section into different variation stages according to the clustering level, and perform averaging of each short-term stage on the original data of the F / Qt value and Qm / Qp value obtained for each stage, respectively, to obtain the average values S fi and S qi of the two short-term stages, and draw the corresponding depth bar chart;

[0016] Step 7: According to the average value O f of the F / Qt value, m f, n f and the average value of Qm / Qp, O q , mq, nq are used to draw the warm and humid - dry and cold quadrant boundaries, establish the semi - quantitative quadrant paleoclimate analysis criteria for F / Qt and Qm / Qp, and average the values of each short - term stage, S fi and S qi are compared with the average values Of, mf, nf and Oq, mq, nq of F / Qt and Qm / Qp, and the change curves of each short - term stage on the sedimentary section are drawn;

[0017] Step Eight: Determine the semi - quantitative relationship between the average value of each stage and the quadrant boundary value, and clarify the significant climate characteristics of the short - term stage;

[0018] Step Nine: Combine the characteristics of the frequency - wave spectrum of time - frequency analysis at scale point A to analyze the medium - term and long - term cycle characteristics of climate change; establish the climate types on the sedimentary section, analyze the laws of climate change, and use them for future climate change prediction.

[0019] Furthermore, O f = [(F / Qt)1+(F / Qt)2+……(F / Qt) n / n, where O f is the total average value of F / Qt, and n is the number of samples;

[0020] m f = [(F / Qt)1+(F / Qt)2+……(F / Qt) i / i, where m f is the average value of the samples with F / Qt less than O f , and i is the number of samples with F / Qt less than O f ;

[0021] n f = [(F / Qt)1+(F / Qt)2+……(F / Qt) j / j, where n f is the average value of the samples with F / Qt greater than O f , and i is the number of samples with F / Qt greater than O f ;

[0022] O q = [(Qm / Qp)1+(m / Qp)2+……(Qm / Qp) n / n, where O q is the total average value of Qm / Qp, and n is the number of samples;

[0023] m f = [(Qm / Qp)1+(m / Qp)2+……(Qm / Qp) i / i, where m qF / Qt is less than 0 q The average value of the sample, i is Qm / Qp less than O f The number of samples;

[0024] n f =[(Qm / Qp)1+(m / Qp)2+……(Qm / Qp) j ] / j, where n q F / Qt is greater than 0 q The average value of the sample, i is Qm / Qp greater than O f The number of samples.

[0025] Further, O f =[(F / Qt)1+(F / Qt)2+……(F / Qt) n ] / n, where O f is the total mean value of F / Qt, n is the number of samples;

[0026] m f =[(F / Qt)1+(F / Qt)2+……(F / Qt) i ] / i, where m f F / Qt is less than 0 f The average value of the sample, i is F / Qt less than O f The number of samples;

[0027] n f =[(F / Qt)1+(F / Qt)2+……(F / Qt) j ] / j, where n f F / Qt is greater than 0 f The average value of the sample, i is F / Qt greater than O f The number of samples;

[0028] O q =[(Qm / Qp)1+(m / Qp)2+……(Qm / Qp) n ] / n, where O q is the total average value of Qm / Qp, n is the number of samples;

[0029] m f =[(Qm / Qp)1+(m / Qp)2+……(Qm / Qp) i ] / i, where m q F / Qt is less than 0 q The average value of the sample, i is Qm / Qp less than O f The number of samples;

[0030] n f =[(Qm / Qp)1+(m / Qp)2+……(Qm / Qp) j / j, where n q is the average value of samples where F / Qt is greater than 0 q and i is the number of samples where Qm / Qp is greater than 0 f .

[0031] Furthermore, the stage division refined by the chronological clustering analysis in step six is divided into several short-term climate change cycles.

[0032] Furthermore, the determination of the scale point A and the scale point B includes:[[]]

[0033] Taking the conversion point from symmetry to weakening of symmetry that appears in the intermediate frequency part as the scale point A;

[0034] Taking the conversion point from weakening of symmetry to complication of symmetry that appears in the intermediate-high frequency part as the scale point B.

[0035] Furthermore, the CONISS function's chronological clustering analysis of the original data of F / Qt and Qm / Qp includes:[[]]

[0036] The clustering division point is G. Taking the G value as the division criterion, the chronological clustering is grouped to obtain the division scheme N;

[0037] When the division criterion is G or less than the G value, comparing the spectral stage division scheme M and the clustering grouping scheme N, if N is included in M and N is a refinement of M, then determine G as the division point for the chronological clustering analysis required this time, and the division scheme N as the division result required this time.

[0038] Furthermore, the climate change types in step six include: warm and humid, relatively warm and humid, semi-arid and cold, dry and cold, four types.

[0039] Another object of the present invention is to provide an analysis system for reconstructing paleoclimate using F / Qt and Qm / Qp data for implementing the analysis method of reconstructing paleoclimate using F / Qt and Qm / Qp data. The analysis system for reconstructing paleoclimate using F / Qt and Qm / Qp data includes:[[]]

[0040] A spectrogram acquisition module;

[0041] A spectral stage division scheme module;

[0042] A data monotonicity determination module;

[0043] A relative change in content module;

[0044] A curve processing module;

[0045] A climate change model determination module.

[0046] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which when executed by a processor causes the processor to perform the following steps of the method for analyzing paleoclimate by reconstructing F / Qt and Qm / Qp data:

[0047] Step 1: Obtain the F / Qt and Qm / Qp of sediment samples at different depth sampling points on the sediment profile and the corresponding depth values; obtain the variation curves of F / Qt and Qm / Qp with depth on the sediment profile; average the original data of F / Qt and Qm / Qp on the sediment profile to obtain their respective total averages; perform spectral analysis on the two groups of data using wavelet transform to obtain spectrograms at different scales;

[0048] Step 2: According to the mirror symmetry analysis of F / Qt and Qm / Qp, identify the variation characteristics of high, medium, and low frequencies, and determine scale point A and scale point B; use the color transformation position as the boundary point or the maximum / minimum value of the color as the boundary point, and use the change of the spectral value at scale point B of the spectrogram of time-frequency analysis to perform stage division on F / Qt and Qm / Qp to obtain the spectral stage division scheme M;

[0049] Step 3: Perform time series clustering analysis on the original data of F / Qt and Qm / Qp using the CONISS function; use the Savitzky-Golay method to perform denoising filtering on the original data of LOI with adjacent several points as the window to remove the high-frequency part and obtain the medium and low-frequency variation trend lines of F / Qt and Qm / Qp; use the variation characteristics of each segment in the division scheme N of the variation trend line to determine the monotonicity of the organic matter content and carbonate content data in each segment;

[0050] Step 4: According to the principle that changing from dry and cold to warm and humid is the ascending semi-cycle, and changing from warm and humid to dry and cold is the descending semi-cycle, identify the short-term climate semi-cycles in the division scheme N; segment and average the original data of F / Qt and Qm / Qp within each short-term climate semi-cycle to obtain the average values of each short-term climate semi-cycle, and draw a depth histogram;

[0051] Step 5: Compare the corresponding relationship between the average values of each stage of F / Qt and Qm / Qp and the total average value, and analyze the relative change of the warm and humid degree;

[0052] Step 6: Determine the climate type of the short-term semi-cycle based on the relative changes of F / Qt and Qm / Qp; combine the characteristics of the frequency-wave spectrum at scale point A of the time-frequency analysis to analyze the medium-term cycle characteristics and long-term cycle characteristics of climate change; establish the climate type on the sediment profile, analyze the law of climate change; determine the climate change model for the corresponding period to predict the future climate change trend.

[0053] Compared with the prior art, the beneficial effects of the present invention are:

[0054] Starting from the original data of sedimentary minerals, from the perspective of signal analysis, the present invention uses time-frequency analysis and time-series analysis to divide the original data into different internally self-consistent cycles. Inside each cycle, noise reduction filtering and statistical methods are used to analyze the characteristics and variation laws of F / Qt and Qm / Qp. According to the principle that the decrease of F / Qt and the increase of Qm / Qp are related to warm and humid climate change, the two parameters restrict and confirm each other, effectively solving the objective uncertainty problem in the process of interpreting paleoclimate information of sedimentary minerals feldspar and quartz, obtaining the continuous change process of warm and humid-dry and cold on the section, summarizing the climate types in the sedimentary evolution process, further establishing a climate model, and serving for predicting the trend of climate change, having the advantages of economy, convenience, high efficiency and science.

[0055] The present invention provides a method for analyzing the warm and humid-dry and cold change characteristics of the paleoenvironment implied by F / Qt and Qm / Qp in sediments, analyzing the climate types and variation laws they represent, and providing basic technical support for promoting global change research and predicting climate change.

[0056] The present invention can clearly analyze the amplitude and degree in the process of relative change of paleoclimate proxy indicators on the section; can scientifically, relatively simply and clearly divide the climate change stages, which plays an important role in analyzing the regularity of paleoclimate change, helps to quickly interpret the results of paleoclimate reconstruction, and promotes the effective development of global comparison work.

[0057] In the current stage of climate evolution of human society, with global warming and increasing extreme climates, sudden temperature drops, cold snaps, freezing rain, blizzards, heatwaves, droughts and out-of-season phenomena have a huge interference and threat to agricultural and pastoral production, transportation, industrial production and the personal safety of residents. Understanding and predicting the future trend of global climate change is the basic task of the sustainable development of human society, and predicting short- and medium-term climate change is the top priority. The analysis method and system for reconstructing paleoclimate using F / Qt and Qm / Qp data provided by the present invention are research methods for realizing the study from the past to the present, clarifying the climate change law in the geological history period, and being able to provide a model and basis for predicting the current global climate change. The popularization of the present invention can accelerate and supplement the establishment of the global climate change model. With the continuous improvement and supplement of the global climate change model, humans will be able to continuously improve the accuracy and efficiency of short- and medium-term climate prediction, reduce social costs and reduce the loss of people's lives and property, thus generating great social benefits and commercial value.

[0058] In the present invention, the feldspar / quartz in detrital sediments is an index of traditional chemical weathering intensity, and its weathering intensity is positively correlated with the degree of warm and humid climate. Quartz grains can be divided into single-crystal quartz (Qm) and polycrystalline quartz (Qp). After undergoing physical and chemical weathering, polycrystalline quartz will be decomposed into single-crystal quartz to a certain extent. This process is closely related to the water medium and is restricted by warm and humid climate conditions. The present invention directly performs wavelet transform on the data to extract the paleoclimate information therein, changing the subjectivity of the interpretation of paleoclimate information caused by directly interpreting the F / Q value by predecessors. At the same time, the present invention also supplements the parameters of single-crystal quartz and polycrystalline quartz (Qm / Qp) to assist in the interpretation of paleoclimate information, improving the objective uncertainty problem of the previous paleoclimate information. It fills the long-term gap in the cross-innovation of sedimentary mineralogy and paleoclimatology by domestic and foreign paleoclimatologists.

[0059] Starting from the sedimentary cycle, the present invention comprehensively studies the two sets of quantities of F / Qt and Qm / Qp in the same objective sedimentary cycle, establishes a semi-quantitative paleoclimate analysis standard, reduces the frequency of the occurrence of the problem of multiple solutions in data interpretation, and promotes the process of applying detrital minerals to paleoclimate research.

[0060] For the F / Qt and Qm / Qp data of continuous sediments, the present invention obtains sedimentary cycles at different levels through wavelet transform and time series CONISS cluster analysis, solves the uncertainty in the process of paleoclimate analysis according to the principle that the external control factors are the same in the same sedimentary cycle, and at the same time, starting from the mean value of the data itself, solves the subjectivity in the process of paleoclimate analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention.

[0062] Figure 1 It is a schematic diagram of the analysis method for reconstructing paleoclimate from F / Qt and Qm / Qp data provided by the embodiment of the present invention.

[0063] Figure 2 Schematic diagram of the depth change curves of F / Qt and Qm / Qp in the Dalianhe Formation of the Yilan Basin provided by the embodiment of the present invention;

[0064] Figure 3 Spectrum diagram of wavelet transform of the original F / Qt and Qm / Qp data in the Dalianhe Formation of the Yilan Basin provided by the embodiment of the present invention;

[0065] Figure 4 Schematic diagram of the stage division of the F / Qt and Qm / Qp parameters in the Dalianhe Formation of the Yilan Basin based on the spectrum diagram provided by the embodiment of the present invention;

[0066] Figure 5Schematic diagram of the low-frequency change trend lines of F / Qt and Qm / Qp in the Dalianhe Formation of the Yilan Basin provided by the embodiments of the present invention;

[0067] Figure 6 Schematic diagram of the stage division of the time series clustering of the original data of F / Qt and Qm / Qp in the Dalianhe Formation of the Yilan Basin provided by the embodiments of the present invention;

[0068] Figure 7 Histogram of the average depth of the time series clustering stage of the original data of F / Qt and Qm / Qp in the Dalianhe Formation of the Yilan Basin provided by the embodiments of the present invention;

[0069] Figure 8 Schematic diagram of the average value change curve of the time series clustering stage of the original data of F / Qt and Qm / Qp in the Dalianhe Formation of the Yilan Basin provided by the embodiments of the present invention;

[0070] Figure 9 Schematic diagram of the short-term, medium- and long-term climate change cycles and their characteristics reflected by F / Qt and Qm / Qp in the Dalianhe Formation of the Yilan Basin provided by the embodiments of the present invention. Detailed implementation manners

[0071] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the following embodiments further elaborate on the present invention. Of course, the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0072] Embodiment 1

[0073] Refer to Figure 1 As shown, the analysis method for reconstructing the paleoclimate of the Cenozoic includes the following steps:

[0074] Step 1: Identification and statistics of detrital minerals quartz, feldspar and rock fragments on the sedimentary section;

[0075] Step 2: Obtain the values of F / Qt and Qm / Qp and the corresponding depths on the sedimentary section, and obtain the change curves of F / Qt and Qm / Qp with depth on the sedimentary section;

[0076] F / Qt is the total amount of feldspar / total amount of quartz, and Qm / Qp is the total amount of single-crystal quartz / total amount of polycrystalline quartz;

[0077] Step 3: Average the original data of the F / Qt and Qm / Qp values on the sedimentary section to obtain their respective total averages O f and O q , and calculate the averages m f and n f of the parts below O f and above O f , as well as below O q and above O qAverage value m of the part q and n q ;

[0078] O f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) n / n, where O f is the total average value of F / Qt, and n is the number of samples;

[0079] m f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) i / i, where m f is the average value of the samples where F / Qt is less than O f , and i is the number of samples where F / Qt is less than O f ;

[0080] n f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) j / j, where n f is the average value of the samples where F / Qt is greater than O f , and i is the number of samples where F / Qt is greater than O f ;

[0081] O q = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) n / n, where O q is the total average value of Qm / Qp, and n is the number of samples;

[0082] m f = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) i / i, where m q is the average value of the samples where F / Qt is less than O q , and i is the number of samples where Qm / Qp is less than O f ;

[0083] n f = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) j / j, where n q is the average value of the samples where F / Qt is greater than O q , and i is the number of samples where Qm / Qp is greater than O f .

[0084] Furthermore, O f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) n / n, where O fis the total average value of F / Qt, and n is the number of samples;

[0085] m f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) i / i, where m f is the average value of the samples with F / Qt less than O f and i is the number of samples with F / Qt less than O f ;

[0086] n f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) j / j, where n f is the average value of the samples with F / Qt greater than O f and i is the number of samples with F / Qt greater than O f ;

[0087] O q = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) n / n, where O q is the total average value of Qm / Qp, and n is the number of samples;

[0088] m f = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) i / i, where m q is the average value of the samples with F / Qt less than O q and i is the number of samples with Qm / Qp less than O f ;

[0089] n f = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) j / j, where n q is the average value of the samples with F / Qt greater than O q and i is the number of samples with Qm / Qp greater than O f ;

[0090] Step 4: Use the wavelet transform of Matlab software to perform spectral analysis on the F / Qt values and Qm / Qp data, obtain the spectrograms at different scales, divide the variation characteristics of high, medium, and low frequencies according to the mirror symmetry characteristics of the F / Qt and Qm / Qp spectrograms, and then use the spectrograms calculated by the wavelet transform to perform stage division on the F / Qt values and Qm / Qp values, and obtain the stage characteristics of the parameters in the high, medium, and low frequency bands;

[0091] Step 5: Use the Savitzky-Golay method to perform denoising filtering on the original data of F / Qt values and Qm / Qp values with several adjacent points as the window, remove the high-frequency part, and obtain the medium- and low-frequency change trend lines of F / Qt values and Qm / Qp values;

[0092] Step 6: Use the CONISS function to perform time-series clustering analysis on the original data of F / Qt values and Qm / Qp values. According to the clustering level, subdivide the parameters on the sediment profile into different change stages, and respectively average the original data of F / Qt values and Qm / Qp values obtained in each stage to obtain the average values S fi and S qi of each short-term stage, and draw the corresponding depth histogram;

[0093] The stage division refined by time-series clustering analysis is several short-term climate change cycles;

[0094] The CONISS function performs time-series clustering analysis on the original data of F / Qt and Qm / Qp, including:

[0095] The clustering division point is G. Using the G value as the division criterion, group the time-series clustering to obtain the division scheme N;

[0096] When the division criterion is G or less than the G value, compare the spectral stage division scheme M and the clustering grouping scheme N. If N is included in M and N is a refinement of M, then determine G as the division point for the time-series clustering analysis required this time, and the division scheme N is the division result required this time.

[0097] The climate change types include: warm and humid, relatively warm and humid, semi-arid and cold, and arid and cold four types

[0098] Step 7: According to the average value O f , m f , n f of F / Qt value and the average value O q of Qm / Qp value, mq, nq, draw the warm and humid-arid and cold quartering boundaries, establish the semi-quantitative quartering paleoclimate analysis standard for F / Qt and Qm / Qp, and compare the average values S fi and S qi of each short-term stage with the average values Of, mf, nf and Oq, mq, nq of F / Qt and Qm / Qp, and draw the change curves of each short-term stage on the sediment profile;

[0099] Step 8: The semi-quantitative relationship between the average value of each stage and the quartering boundary value to clarify the significant climate characteristics of the short-term stage.

[0100] Step 9: Combine the characteristics of the frequency-wave spectrum at scale point A in time-frequency analysis to analyze the medium-term and long-term cycle characteristics of climate change; establish the climate types on the sedimentary section, analyze the laws of climate change, and use them for future climate change prediction.

[0101] The determination of the scale point A and the scale point B includes:

[0102] Take the conversion point from symmetry to weakening of symmetry in the intermediate frequency part as the scale point A;

[0103] Take the conversion point from weakening of symmetry to complication of symmetry in the intermediate-high frequency part as the scale point B.

[0104] Based on the original data of sediment F / Qt and Qm / Qp, use time-frequency analysis and time series analysis to divide the original data into different cycles that are internally self-consistent. Within each cycle, use noise reduction filtering and statistical methods to analyze the characteristics and variation laws of F / Qt and Qm / Qp within each cycle, and at the same time obtain the changing process of paleoenvironmental characteristics - the degree of warmth and humidity from continuous qualitative to semi-quantitative, obtain the climate types on the sedimentary section, and construct a climate model.

[0105] Specific case 1:

[0106] Specific case 1 is the continuous high-precision terrigenous clastic rock sediment drilling of the Dalianhe Formation in the Yilan Basin in Northeast China. The section from 899.2m to 708.2m is the conglomerate, coal-bearing and oil shale section (E 1-2 d 1 ); the middle part from 708.2m to 590.3m is the oil shale section (E 1- 2d 2 ); the upper part from 590.3m to 132.0m is the sandstone-shale section, composed of interbeds of sandstone, siltstone and mudstone (E 1-2 d 3 ).

[0107] The specific implementation steps are as follows:

[0108] Step 1: Obtain the samples at sampling points at different depths on the sedimentary section, grind thin sections, conduct microscopic identification and statistics of feldspar F, single crystal quartz (Qm), polycrystalline quartz (Qp) and lithic fragments (Lt), and obtain the sediment F / Qt and Qm / Qp values and their corresponding depth values di (cm) at sampling points at different depths on the sedimentary section.

[0109] After collecting section samples at 5 cm intervals, make thin sections and conduct microscopic identification, with more than 300 particles identified and statistically analyzed.

[0110] Step 2: Obtain the curves of the F / Qt and Qm / Qp values changing with depth on the sedimentary section. As Figure 2 shown.

[0111] Step 3: Average the original data of F / Qt and Qm / Qp values on the sedimentary section to obtain their respective total average values O f and O q , and calculate the average values m f and n f for the parts below O f and above O f , respectively, as well as the average values m q and n q for the parts below O q and above O q .

[0112] Step 4: Use the wavelet transform of Matlab software to perform spectral analysis on the two groups of data to obtain spectrograms at different scales. According to the principle that in wavelet transform, a large scale a shows the characteristics of the low-frequency part and a small scale a shows the characteristics of the high-frequency part, and based on the mirror symmetry characteristics of the F / Qt and Qm / Qp spectrograms, divide the variation characteristics of high, medium, and low frequencies, as Figure 3 shown.

[0113] In this specific case, the different scales are two scales of a = 450 and a = 100, where a ≥ 450 belongs to the low frequency; 100 ≤ a ≤ 450 belongs to the medium-high frequency, and a < 100 belongs to the high frequency.

[0114] In this specific case, the spectrograms of the F / Qt value and the Qm / Qp value are basically the same when the wavelet transform scale a ≥ 450, that is to say, the spectral changes in the medium-low frequency part (a = 450 - 1000) are basically the same, representing the consistency of the control factors of the F / Qt value and the Qm / Qp value at the medium-low frequency scale. However, when 100 ≤ a ≤ 450, although the spectra of the F / Qt value and the Qm / Qp value have a certain similarity, differences have appeared at some depths, especially when the scale a = 100, and a more complex corresponding relationship begins to appear in their spectrograms. This reflects that at the high frequency scale, the control factors of the F / Qt value and the Qm / Qp value are significantly different.

[0115] Step 5: Use the spectrograms obtained by wavelet transformation to perform stage division on the parameters of the F / Qt value and the Qm / Qp value, and obtain the stage characteristics of the parameters in the high, medium, and low frequency bands, as Figure 4 shown.

[0116] In the embodiment of the present invention, the change boundary of color is used. When a = 450, three parts can be identified. The first part corresponds to the Ypresian stage, and the second and third parts correspond to the Lutetian and Bartonian stages. When a = 100, the F / Qt value can identify ten clear stages of 1 - 3, 5, 7 - 10, 13, 16, and the Qm / Qp value can identify fourteen clear stages of 1 - 6, 8, 11, 12, 14 - 18 (as Figure 4as shown

[0117] Step 6: Use the Savitzky-Golay method to perform denoising filtering on the original data of F / Qt values and Qm / Qp values with several adjacent points as a window, remove the high-frequency part, and obtain the medium- and low-frequency change trend lines of the F / Qt values and Qm / Qp values, as Figure 5 shown

[0118] Use the Savitzky-Golay method to perform denoising filtering on the original data of F / Qt values and Qm / Qp values with 50 adjacent points as a window to remove the high-frequency part, and obtain the medium- and low-frequency change trend lines of the F / Qt values and Qm / Qp values.

[0119] Step 7: Use the CONISS function to perform time-series clustering analysis on the original data of F / Qt values and Qm / Qp values, and subdivide the parameters on the section into different change stages according to the clustering level, as Figure 6 shown

[0120] Subdivide the sampled well section into 19 short-term stages with a clustering level of 1.1. When the clustering level is less than 1.1, the boundary divided by the time-series analysis is consistent with the boundary divided by the time-frequency analysis scale a = 100. Therefore, the 19 short-term stages are included within the 18 time-frequency analysis stages, and the boundaries are basically the same, which are the best short-term stages in this specific case.

[0121] Step 8: Perform averaging for each short-term stage on the original data of F / Qt values and Qm / Qp values obtained respectively, and obtain the average values S fi and S qi of the two for each short-term stage, and draw the corresponding depth histogram, as Figure 7 shown

[0122] Perform stage averaging on the original data of F / Qt values and Qm / Qp values for the 19 short-term stages respectively. If the average value of F / Qt includes the abnormally high value at the bottom, it is 0.22. According to the lithological characteristics, the abnormally high value should be discarded. After discarding the abnormally high value, obtain the average value S fi of the F / Qt value on the section for each cycle and the corresponding depth histogram. Similarly, obtain the average value Sqi of Qm / Qp on the section for each cycle and the corresponding depth histogram.

[0123] Step 9: According to the average value O f , m f , n f of the F / Qt value and the average value O q , mq, nq of the Qm / Qp value, draw the warm-wet - dry-cold quartering boundary. Establish the semi-quantitative quartering paleoclimate analysis standard for F / Qt and Qm / Qp, as Figure 7 shown

[0124] Based on the average values of short - term semi - cycles F / Qt and Qm / Qp and the relative changes of O f , m f , n f and Oq, mq, nq to determine the climate type; when the F / Qt value is less than mf, it is defined as dry - cold; when it is between mf and Of, it is defined as semi - dry - cold; when it is between Of and nf, it is defined as semi - warm - humid; when it is greater than nf, it is defined as warm - humid. When the Qm / Qp is less than mq, it is defined as dry - cold; when it is between mq and Oq, it is defined as semi - dry - cold; when it is between Oq and nq, it is defined as semi - warm - humid; when it is greater than nq, it is defined as warm - humid.

[0125] The warm - humid - dry - cold quartering boundaries of F / Qt and Qm / Qp on the section are Of = 0.2, mf = 0.17, nf = 0.25; Oq = 11.81, mq = 9.46, nq = 14.46.

[0126] Step 10: Plot the change curves of each short - term stage on the sediment section, as Figure 8 shown.

[0127] Step 11: Based on the semi - quantitative relationship between the average value of each stage and the quartering boundary value, clarify the significant climate characteristics of the short - term stage. As Figure 8 .

[0128] Taking the corresponding relationship between the average value of each stage of the F / Qt value and the Qm / Qp value and the total average value, the average value lower than the total average value, and the average value higher than the total average value as the semi - quantitative benchmark, analyze the relative changes of the F / Qt value and the Qm / Qp value, and clarify the significant characteristics of 19 short - term stages.

[0129] Short - term climate stage characteristics of F / Qt: (1) Dry - cold; (2) Relatively warm - humid; (3) Relatively warm - humid; (4) Relatively warm - humid; (5) Warm - humid; (6) Relatively warm - humid; (7) Relatively warm - humid; (8) Dry - cold; (9) Relatively dry - cold; (10) Warm - humid; (11) Relatively dry - cold; (12) Dry - cold; (13) Relatively warm - humid; (14) Relatively warm - humid; (15) Relatively warm - humid; (16) Warm - humid; (17) Warm - humid; (18) Relatively dry - cold; (19) Warm - humid.

[0130] Short - term climate stage characteristics of Qm / Qp: (1) Dry - cold; (2) Relatively dry - cold; (3) Warm - humid; (4) Relatively dry - cold; (5) Warm - humid; (6) Relatively dry - cold; (7) Relatively warm - humid; (8) Dry - cold; (9) Relatively dry - cold; (10) Warm - humid; (11) Dry - cold; (12) Dry - cold; (13) Relatively dry - cold; (14) Relatively warm - humid; (15) Relatively warm - humid; (16) Warm - humid; (17) Relatively warm - humid; (18) Relatively dry - cold; (19) Warm - humid.

[0131] Step 12: Divide the stages refined by time - series clustering into 19 short - term climate change cycles, as Figure 9 shown.

[0132] According to the above principle, the change from dry and cold to warm and humid is defined as the ascending half cycle, and the change from warm and humid to dry and cold is defined as the descending half cycle. The refined stage of time series clustering is divided into 19 short-term climate change cycles, and the climate characteristics of each cycle are the same as the significant climate characteristics of the short-term stage.

[0133] Step 13: Combine the medium and low frequency change trends, and summarize the characteristics of the medium-term climate cycle on the profile according to each short-term climate half cycle.

[0134] The (1) dry and cold, (2) relatively warm and humid, (3) relatively warm and humid, and (4) relatively warm and humid of the short-term time series climate cycle constitute the medium-term ascending Cycle A. During this period, the climate shows a change from dry and cold to warm and humid; the (5) warm and humid, (6) relatively warm and humid, and (7) relatively warm and humid constitute the medium-term descending Cycle B. During this period, the climate gradually changes from warm and humid to dry and cold; the (8) dry and cold, (9) relatively dry and cold, and (10) warm and humid constitute the medium-term ascending Cycle C. During this period, the climate briefly reverses from dry and cold to warm and humid; the (11) relatively dry and cold, (12) dry and cold, (13) relatively warm and humid, (14) relatively warm and humid, (15) relatively warm and humid, (16) warm and humid, and (17) warm and humid constitute the medium-term ascending Cycle D. During this period, the climate is in a long-term change from dry and cold to warm and humid; the short-term climate cycle (18) relatively dry and cold and (19) warm and humid constitute the medium-term ascending Cycle E. During this period, the climate continues to change towards warm and humid.

[0135] Step 14: Combine the medium and low frequency change trends, and summarize the characteristics of the long-term climate cycle on the profile according to each short-term climate half cycle.

[0136] The medium-term climate cycles A and B constitute a complete long-term climate cycle. During this period, the climate changes from dry and cold to warm and humid, and then to dry and cold; the medium-term cycles C, D, and E constitute the long-term ascending half cycle. The climate is in a process of fluctuating warming, but the degree of warm and humid does not exceed that of the previous climate cycle.

[0137] Step 15: Establish the climate types on the sedimentary profile and analyze the laws of climate change.

[0138] There are four climate types in the Eocene of the Yilan Basin: humid and hot, relatively humid and hot, dry and cold, and semi-dry and cold.

[0139] Step 16: Analyze the mechanism of the paleoclimate indication significance of sedimentary minerals feldspar and quartz, and propose a climate change model for this period to provide a basis for predicting the local climate change trend.

[0140] The lower part of the profile of the present invention is dry and cold, and then the degree of warm and humid gradually increases; the earliest stage of the Middle Lutetian reaches the warmest and wettest state, and then the degree of warm and humid decreases; the earliest stage of the Bartonian reaches the driest and coldest state, and then the degree of warm and humid rebounds. This paleoclimate change pattern provides an important paleoclimate basis for the paleoclimate change in the Eocene of the Yilan Basin and an important sedimentary mineralogical basis for predicting the current climate change trend.

[0141] The long-term climate cycle analysis revealed in the Yilan area of Northeast China shows that the paleoclimate in the Ypresian stage pulsatingly warmed from a relatively low level through multiple oscillations and reached the warmest and wettest state in the Lutetian stage. The paleoclimate in the early Lutetian stage was in a warm and wet state for a relatively long time, and the degree of warmth and wetness decreased and transformed into dry and cold in the later stage. In the early Bartonian stage, the paleoclimate was in a dry and cold state, and the degree of dryness and cold reached the driest and coldest state. After the early Bartonian stage, through three warmings, it gradually changed from the driest and coldest state to warm and wet.

[0142] This process is in very good agreement with the evolution process of the climate type reconstructed from the loss on ignition in this area in the early stage. The climate change trends reconstructed by the two proxy indicators in the Eocene are consistent with the overall trends of temperature changes in Northeast China and the world. However, it essentially solves the problem that Quan et al. (2012) could not improve the continuous accuracy of paleoclimate due to the discontinuity of fossils. At the same time, the easier accessibility of continental drill holes compared to marine drill holes will also bring new vitality to the reconstruction of paleoclimate in continental sedimentary strata. This is a problem that is difficult to solve by other methods at present.

[0143] Another object of the present invention is to provide an analysis system for reconstructing paleoclimate using F / Qt and Qm / Qp data for implementing the analysis method for reconstructing paleoclimate using F / Qt and Qm / Qp data. The analysis system for reconstructing paleoclimate using F / Qt and Qm / Qp data includes:

[0144] A spectrogram acquisition module, configured to acquire the F / Qt and Qm / Qp of sediment at different depth sampling points on a sedimentary section and the corresponding depth values; acquire the change curves of F / Qt and Qm / Qp with depth on the sedimentary section; average the original data of F / Qt and Qm / Qp on the sedimentary section to obtain their respective total average values; perform spectral analysis on the two groups of data using wavelet transform to obtain spectrograms at different scales;

[0145] A spectral stage division scheme module, configured to identify the change characteristics of high, medium, and low frequencies and determine scale point A and scale point B according to the mirror symmetry analysis of the F / Qt and Qm / Qp spectrograms; use the position of color change as the boundary point or the maximum / minimum value of color as the boundary point, and perform stage division of F / Qt and Qm / Qp using the change of the spectral value at scale point B of the spectrogram obtained by time-frequency analysis to obtain the spectral stage division scheme M;

[0146] The data monotonicity determination module is used to perform time-series clustering analysis on the original F / Qt and Qm / Qp data using the CONISS function; use the Savitzky-Golay method to perform denoising filtering on the original F / Qt and Qm / Qp data with adjacent points as the window, remove the high-frequency part, and obtain the mid-low frequency change trend lines of F / Qt and Qm / Qp; use the change characteristics of the trend lines in each segment of the division scheme N to determine the monotonicity of the F / Qt and Qm / Qp data in each segment.

[0147] The relative change of content module is used to determine the relationship between F / Qt and Qm / Qp and their respective boundary values based on the value range of the paleoclimate quartile boundary; plot the trend change curves of F / Qt and Qm / Qp on the sedimentary profile; evaluate the relative change of F / Qt and Qm / Qp based on the corresponding relationship between the average values of each stage of F / Qt and Qm / Qp and the total average value.

[0148] The curve processing module is used to determine the climate type of the short-term semi-cycle based on the relative change of F / Qt and Qm / Qp; according to the semi-quantitative change curve of the paleo-warm and humid degree of each short-term semi-cycle climate type; perform smoothing filtering on the F / Qt and Qm / Qp curves to obtain their respective warm and humid degree trend lines; analyze the consistency of the changes of the two parameters on the same scale.

[0149] The climate types of the short-term semi-cycle include:

[0150] (1) When the F / Qt value is less than mf, it is defined as dry and cold, between mf and Of as semi-dry and cold, between Of and nf as semi-warm and humid, and greater than nf as warm and humid.

[0151] (2) When Qm / Qp is less than mq, it is defined as dry and cold, between mq and Oq as semi-dry and cold, between Oq and nq as semi-warm and humid, and greater than nq as warm and humid.

[0152] The climate change model determination module is used to analyze the characteristics of the mid-term cycle of climate change by combining the characteristics of the time-frequency analysis frequency spectrum at the scale point A; analyze the characteristics of the long-term cycle of climate change by synthesizing the overall change characteristics of the warm and humid degree; establish the climate type on the sedimentary profile, analyze the laws of climate change; and determine the climate change model corresponding to the period.

[0153] During the Eocene (56 - 33.9 Ma), the global climate first warmed and then cooled, transitioning from a greenhouse to an icehouse. During this period, there were a series of extreme climate events and typical climate types such as the Paleocene-Eocene Thermal Maximum (PETM) and the Eocene Climate Optimum (EECO). Currently, humans are in a stage of climate evolution with global warming and increasing extreme climates, which has strong comparability with the Eocene. Therefore, the study of Eocene climate evolution will provide research types and comparison bases for the current global climate change caused by the increase in greenhouse gases.

[0154] Continuous sedimentary clastic mineral identification and statistical experiments were carried out on the borehole core data of the Eocene Dalihe Formation in the Yilan Basin, Northeast China. Time-frequency time-series analysis was used to divide the F / Qt and Qm / Qp data of the Eocene continuous sediments in the Yilan Basin into cycles, and the Eocene Dalihe Formation was divided into 3 long-term half-cycles, 5 middle-term half-cycles, and 19 short-term half-cycles.

[0155] Then, Pearson correlation analysis, noise reduction filtering, and statistical methods were used to analyze the characteristics and variation laws of F / Qt and Qm / Qp data. The discrimination criteria are as follows: when the F / Qt value is less than mf, it is defined as dry and cold; when it is between mf and Of, it is defined as semi-dry and cold; when it is between Of and nf, it is defined as semi-warm and humid; when it is greater than nf, it is defined as warm and humid. When Qm / Qp is less than mq, it is defined as dry and cold; when it is between mq and Oq, it is defined as semi-dry and cold; when it is between Oq and nq, it is defined as semi-warm and humid; when it is greater than nq, it is defined as warm and humid.

[0156] Four climate types, namely warm and humid, relatively warm and humid, semi-dry and cold, and dry and cold, were identified in the Eocene of the Yilan Basin. Based on continuous high-precision sampling of the profile, the continuous climate change process in the Eocene of the Yilan area was identified. It is considered that the medium- and long-term change trend of the warm and humid degree in the Eocene of Northeast China is to increase first, reach the highest level in the entire Eocene in the Lutetian stage, then gradually decrease, reach the lowest level in the early Bartonian stage, and then rise. In particular, it was found that F / Qt and Qm / Qp as a whole show a mirror image relationship. The higher the degree of mirror image, the more synchronous the inference results, and the more credible the inference results. Among them, F / Q is generally sensitive to the warm and humid process; Qm / Qp is sensitive to the dry and cold process. This study will provide an important supplement to the paleoclimate change in the Eocene of Yilan and Northeast China, and also provide a scientific method and analysis means for reconstructing paleoclimate from sedimentary minerals in other regions, and can provide an important geological basis for predicting future climate change trends.

[0157] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which when executed by a processor causes the processor to execute the following steps of the analysis method for reconstructing paleoclimate from F / Qt and Qm / Qp data:

[0158] Step 1: Obtain the F / Qt and Qm / Qp of the sediments at different depth sampling points on the sedimentary profile and the corresponding depth values; obtain the change curves of F / Qt and Qm / Qp with depth on the sedimentary profile; average the original data of F / Qt and Qm / Qp on the sedimentary profile to obtain their respective total average values; use wavelet transform to perform spectral analysis on the two groups of data to obtain spectrograms at different scales;

[0159] Step 2: According to the mirror symmetry analysis of F / Qt and Qm / Qp, identify the variation characteristics of high, medium, and low frequencies, and determine scale points A and B; use the color transformation position as the boundary point or the maximum / minimum value of the color as the boundary point, and use the change of the spectral value at scale point B in the spectrogram of time-frequency analysis to divide the stages of F / Qt and Qm / Qp, obtaining the spectral stage division scheme M.

[0160] Step 3: Use the CONISS function to perform time-series clustering analysis on the original data of F / Qt and Qm / Qp; use the Savitzky-Golay method to perform denoising filtering on the original data of LOI with adjacent several points as the window, remove the high-frequency part, and obtain the medium and low-frequency change trend lines of F / Qt and Qm / Qp; use the change characteristics of each segment in the division scheme N of the change trend line to determine the monotonicity of the organic matter content and carbonate content data in each segment.

[0161] Step 4: According to the principle that changing from dry and cold to warm and humid is the ascending half-cycle, and changing from warm and humid to dry and cold is the descending half-cycle, identify the short-term climate half-cycles in the division scheme N; segment and average the original data of F / Qt and Qm / Qp within each short-term climate half-cycle to obtain the average value of each short-term climate half-cycle, and draw a depth histogram.

[0162] Step 5: Compare the corresponding relationship between the average values of each stage of F / Qt and Qm / Qp and the total average value, and analyze the relative change of the warm and humid degree.

[0163] Step 6: Determine the climate type of the short-term half-cycle based on the relative changes of the F / Qt and Qm / Qp; combine the characteristics of the frequency-wave spectrum of time-frequency analysis at scale point A to analyze the mid-term cycle characteristics and long-term cycle characteristics of climate change; establish the climate type on the sedimentary section, analyze the law of climate change; determine the climate change model corresponding to the period for predicting future climate change trends.

[0164] The above is only a preferred embodiment of the present invention, and it is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. Analytical method for reconstructing paleoclimate of the Cenozoic era, characterized in that, It includes the following steps: Step 1: Identification and statistics of detrital minerals quartz, feldspar, and rock fragments on the sediment profile; Step 2: Obtain the values of F / Qt and Qm / Qp and their corresponding depths on the sediment profile, and obtain the variation curves of F / Qt and Qm / Qp with depth; F / Qt is the total amount of feldspar / the total amount of quartz, and Qm / Qp is the total amount of single-crystal quartz / the total amount of polycrystalline quartz; Step 3: Average the original data of F / Qt and Qm / Qp values on the sediment profile to obtain their respective total average values O f and O q , and calculate the average values m f and n f for the parts below O f and above O f , respectively, as well as the average values m q and n q for the parts below O q and above O q ; Step 4: Use the wavelet transform of Matlab software to perform spectral analysis on the F / Qt values and Qm / Qp data, obtain spectrograms at different scales, divide the high, medium, and low-frequency variation characteristics according to the mirror symmetry characteristics of the F / Qt and Qm / Qp spectrograms, and then use the spectrograms obtained by wavelet transform to perform stage division on the F / Qt values and Qm / Qp values parameters to obtain the stage characteristics of the parameters in the high, medium, and low-frequency bands; Step 5: Use the Savitzky-Golay method to perform denoising filtering on the original data of F / Qt values and Qm / Qp values with adjacent several points as the window, remove the high-frequency part, and obtain the medium and low-frequency variation trend lines of F / Qt values and Qm / Qp values; Step 6: Use the CONISS function to perform time series clustering analysis on the original data of F / Qt values and Qm / Qp values. According to the clustering level, subdivide the parameters on the sediment profile into different change stages, and respectively average the original data of F / Qt values and Qm / Qp values obtained in each stage to obtain the average values S fi and S qi , and draw the corresponding depth histogram; Step Seven: According to the average value O of F / Qt f , m f , n f and the average value O of Qm / Qp q , mq, nq, draw the warm-wet and dry-cold quartering boundaries, establish the semi-quantitative quartering paleoclimate analysis criteria for F / Qt and Qm / Qp, and compare the average values S fi and S qi of each short-term stage with the average values Of, mf, nf and Oq, mq, nq of F / Qt and Qm / Qp, and draw the variation curves of each short-term stage on the sedimentary profile; Step 8: The semi-quantitative relationship between the average value of each stage and the quartering boundary value to clarify the significant climate characteristics of the short-term stage; Step 9: Combine the characteristics of the frequency-wave spectrum at scale point A in the time-frequency analysis, analyze the characteristics of the medium-term cycle and long-term cycle of climate change; establish the climate type on the sediment profile, analyze the law of climate change, and use it for future climate change prediction.

2. The analysis method for reconstructing paleoclimate of the Cenozoic era according to claim 1, wherein O f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) n / n, where O f is the total average value of F / Qt, and n is the number of samples; m f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) i / i, where m f is the average value of samples with F / Qt less than O f and i is the number of samples with F / Qt less than O f ; n f = [(F / Qt)1 + (F / Qt)2 + ……(F / Qt) j / j, where n f is the average value of the samples with F / Qt greater than O f and i is the number of samples with F / Qt greater than O f ; O q = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) n / n, where O q is the total average value of Qm / Qp, and n is the number of samples; m f = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) i / i, where m q is the average value of samples with F / Qt less than O q and i is the number of samples with Qm / Qp less than O f ; n f = [(Qm / Qp)1 + (m / Qp)2 + ……(Qm / Qp) j / j, where n q is the average value of samples with F / Qt greater than O q and i is the number of samples with Qm / Qp greater than O f .

3. The analysis method for reconstructing the paleoclimate of the Cenozoic era according to claim 2, characterized in that The stage division refined by the time series clustering analysis in Step 6 is several short-term climate change cycles.

4. The method for analyzing and reconstructing paleoclimate of the Cenozoic era according to claim 3, wherein The determination of the scale point A and scale point B includes: Take the conversion point from symmetry to weakening of symmetry in the medium-frequency part as scale point A; Take the conversion point from weakening of symmetry to complication of symmetry in the medium-high frequency part as scale point B.

5. The analysis method for reconstructing the paleoclimate of the Cenozoic era according to claim 4, characterized in that, The CONISS function performs time series clustering analysis on the F / Qt and Qm / Qp original data, including: The clustering division point is G, and using the G value as the division standard, group the time series clustering to obtain the division scheme N; When the division standard is G or less than the G value, compare the spectral stage division scheme M and the clustering grouping scheme N. If N is included in M and N is a refinement of M, then determine G as the division point for the time series clustering analysis required this time, and the division scheme N as the division result required this time.

6. The analysis method for reconstructing paleoclimate of the Cenozoic era according to claim 5, characterized in that, The climate change types in Step 6 include: warm and humid, relatively warm and humid, semi-arid and cold, dry and cold four types.

7. An analysis system for implementing the analysis method of reconstructing the paleoclimate of the Cenozoic era as described in any one of claims 1-6, characterized in that, The analysis system includes: Spectrogram acquisition module; Spectral stage division scheme module; Data monotonicity determination module; Relative content change module; Curve processing module; Climate change model determination module.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the processor executes the method according to any one of claims 1-6.

Images