Quantitative reconstruction of paleoclimate based on spore-pollen data

Abstract

The application provides a quantitative recovery method of paleoclimate based on spore and pollen data, which comprises the following steps: step 1, establishing a basic database of paleo-pollen; step 2, establishing a comprehensive database of ecological and climatic conditions of paleo-pollen; step 3, reconstructing a paleoclimate distribution model tending to normal distribution; step 4, checking the distribution shape of the average distribution data model; step 5, normalizing the data model; and step 6, recovering the characteristics of the paleoclimate conditions. The quantitative recovery method of paleoclimate based on spore and pollen data has strong adaptability and expandability, can be combined with the discrimination of the source of the early spore and pollen, adjust the scale of the confidence interval based on the proportion of the original and external spore and pollen, and further eliminate the interference of the external spore and pollen with different degrees on the recovery of the paleoclimate. The method has clear and clear principles, quantifies the recovery of the paleoclimate, and has high accuracy and operability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas exploration and development, and particularly relates to a method for quantitatively restoring paleoclimate based on sporopollen data. BACKGROUND

[0002] As an important carrier of oil and gas enrichment, sedimentary strata provide different types of reservoirs (such as sandstone, mudstone and carbonate reservoirs) for oil and gas enrichment, and also provide hydrocarbon source material for oil and gas enrichment. Undoubtedly, how the sedimentary strata are formed has always been an important content of oil and gas geological exploration research. Previous sedimentary research shows that the formation of sedimentary strata is controlled by two external factors, i.e. tectonic activity and climate change. Tectonic activity is easy to quantitatively describe due to its long duration and intense activity, and there are many research methods to support the related research on the formation of sedimentary strata. In comparison, it is difficult to quantitatively restore climate change due to its frequent fluctuations, and it is difficult to study the formation process of sedimentary strata under the constraint of climate.

[0003] At present, the method for quantitatively restoring climate is mainly the coexistence analysis method based on sporopollen data. However, there are still some limitations in quantitatively restoring paleoclimate by sporopollen coexistence analysis method. For example: (1) The coexistence analysis method needs a certain amount of data to quantitatively restore paleoclimate, and the accuracy of quantitatively restoring climate by the coexistence analysis method is low in the study area with few drilling cores and few sporopollen surviving climate; (2) The coexistence analysis method is based on the principle of biological coexistence to obtain all biological coexistence climate intervals in a certain period of time to quantitatively describe the climate in this period. The interval of the restored paleoclimate parameters such as paleotemperature and paleoprecipitation is limited. The paleoclimate parameters such as paleotemperature and paleoprecipitation in the geological history are continuously fluctuating, and the data distribution characteristics of the paleoclimate parameters are not clear; (3) Exogenous sporopollen interferes with the restoration of paleoclimate, and it is difficult to effectively eliminate the influence of exogenous sporopollen.

[0004] Therefore, how to accurately and comprehensively quantitatively restore paleoclimate is a technical problem to be solved at present.

[0005] In the Chinese patent application with the application number CN201810016055.X, a method for quantitatively characterizing paleo-hydrocarbon generation environment is disclosed, which comprises the following steps: step A, establishing a quantitative relationship between sporopollen and paleoclimate by characterizing sporopollen index; step B, quantitatively describing paleo-water salinity by characterizing trace elements; step C, quantitatively describing the redox condition of paleo-water medium by characterizing biomarkers; and step D, quantitatively determining the paleo-hydrocarbon generation environment level by superimposed analysis of the three factors of paleoclimate, paleo-water salinity and paleo-water medium redox condition. The purpose of the present application is to provide a method for quantitatively characterizing paleo-hydrocarbon generation environment, which can determine the level of paleo-productivity, analyze the development of hydrocarbon source rock, and provide a feasible technical system for the prediction and research of high-quality hydrocarbon source rock distribution and the realization of high-efficiency storage target.

[0006] In the Chinese patent application with the application number: CN201710255338.5, a method for constructing a sedimentary model under the constraint of a paleo-dry climate in a terrestrial basin is disclosed, especially for the case where multiple sedimentary facies types coexist under the constraint of a dry climate in a terrestrial lake basin. The invention aims to address the relative humid and dry environmental conditions under the constraint of a paleo-dry climate in a terrestrial lake basin, and to construct a sedimentary model under the constraint of a paleo-dry climate, thereby supplementing the relevant theory of climate and sedimentary facies in a sedimentary basin, and providing guidance for the sedimentary filling pattern and favorable sand body prediction in a terrestrial lake basin. The technical method comprises the following steps: step 1: determining the isochronal sedimentary top and bottom interfaces under the constraint of a paleo-dry climate; step 2: refining the paleo-dry climate fluctuations using geochemical data; step 3: establishing the relationship between the paleo-dry climate and the sedimentary facies based on lithological combinations; and step 4: establishing a sedimentary model under the constraint of a paleo-dry climate.

[0007] In the Chinese patent application with the application number: CN201910034491.4, a method for determining the development range of favorable weathered crust reservoirs in complex lithologic paleo-buried hills is disclosed, which comprises the following steps: 1) performing paleo-climate analysis on the strata superimposed on the complex lithologic paleo-buried hills to restore the paleo-climate background of the paleo-buried hills during each period when the paleo-buried hills were buried by superimposition; 2) determining the strata deposited under dry climate conditions and the strata deposited under humid climate conditions superimposed on the complex lithologic paleo-buried hills; the weathered crust of the complex lithologic paleo-buried hills superimposed by the strata deposited under dry climate conditions is a potential location for the development of favorable reservoirs in the paleo-buried hills, and the weathered crust of the complex lithologic paleo-buried hills superimposed by the strata deposited under humid climate conditions is difficult to develop favorable weathered crust reservoirs; and 3) removing the parts where the internal siltstone and mudstone layers of the paleo-buried hills are exposed on the surface of the paleo-buried hills from the weathered crust of the complex lithologic paleo-buried hills superimposed by the strata deposited under dry climate conditions, and the remaining parts are the development range of the favorable weathered crust reservoirs in the complex lithologic paleo-buried hills.

[0008] The above prior art has great differences from the present invention and cannot solve the technical problems we want to solve, therefore we have invented a new method for quantitatively restoring paleo-climate based on pollen data. SUMMARY

[0009] The purpose of the present invention is to provide a method for quantitatively restoring paleo-climate based on pollen data, which can restore the data distribution characteristics of paleo-temperature, paleo-rainfall and other climate parameters in the geological history period, and further quantitatively restore the paleo-climate in that period.

[0010] The purpose of the present invention can be achieved by the following technical measures: a method for quantitatively restoring paleo-climate based on pollen data, which comprises:

[0011] Step 1: Establish a basic database of palynology;

[0012] Step 2: Establish a comprehensive database of palynology ecological climate conditions;

[0013] Step 3: Reconstruct a palaeoclimate distribution model tending to normal distribution;

[0014] Step 4: Check the distribution shape of the average distribution data model;

[0015] Step 5: Regularize the data model;

[0016] Step 6: Restore the characteristics of palaeoclimate conditions.

[0017] The object of the application can also be achieved by the following technical measures:

[0018] In step 1, a coring well with complete stratigraphic record in the study area is selected, and the palynology data of the target layer are counted to determine the types of palynology developed in the target layer, the frequency of different types of palynology and the total number of palynology, and a basic database of palynology is established.

[0019] In step 2, based on the types of palynology in the basic database of palynology, the corresponding existing closest relative species of palynology are determined, and according to the palaeoclimate parameters required for the growth of the existing closest relative species in the database, the palynology indicating palaeoclimate parameters are determined, and a comprehensive database of palynology ecological climate conditions is established.

[0020] In step 2, the palynology indicating palaeoclimate parameters include annual average temperature, average temperature of the coldest month, average temperature of the hottest month, annual average precipitation, average precipitation of the wettest month, and average precipitation of the driest month.

[0021] In step 3, based on the comprehensive database of palynology ecological climate conditions, the average distribution of palaeoclimate data model is formed by taking the climate condition interval of different types of spores and its frequency as constraints.

[0022] In step 4, based on the average distribution data model formed, the data distribution shape is determined to be which one of normal distribution, negative skewness, positive skewness, steep peak state or slow peak state according to left-right symmetry and peak steepness.

[0023] In step 4, if the graphical method cannot effectively determine, the distribution state of the data model is further determined by calculating three parameters of kurtosis K, skewness S and standard error SE; the calculation formulas of the three parameters are as follows:

[0024]

[0025]

[0026]

[0027] where K is kurtosis; S is skewness; SE is standard error; SD is standard deviation; n is sample size, x i The expression data model of the i-th value, is the average value of the data model sample.

[0028] In step 4, the high peak of the left skewness distribution is right-biased, and the long tail gradually extends from the right to the left end, and S<0, and the positive skewness is the opposite; the discrimination of the peak state is based on the kurtosis K, K>0 is the steep peak state, and K<0 is the slow peak state; the discrimination of the skewness and the peak state degree is based on K, S and SE parameters, when the absolute value of K or S is greater than 1.96 times of the standard error, it has obvious significant skewness and peak state characteristics, when the absolute value of K or S is 2-3 times of the standard error, the skewness and the peak state reach the medium degree, and when the absolute value of K or S exceeds 3 times of the standard error, the skewness and the peak state reach the high degree.

[0029] In step 5, based on the judgment of the shape distribution of the data model in step 4, the corresponding regularization method is selected to regularize the data model; if the data model is judged to be in accordance with the normal distribution in step 4, it can directly jump to step 6 to generate a random data set to obtain the climate condition interval, the central tendency and the fluctuation degree.

[0030] In step 5,

[0031] Light to moderate skewness and peak state data adopts square root transformation,

[0032] High degree skewness and peak state data adopts logarithmic transformation, y=lny′ or lgy′

[0033] Large fluctuation data at both ends adopts inverse transformation,

[0034] where y is the data set after regularization, y′ is the data set to be regularized;

[0035] The square root transformation variable requires y′≥0, the logarithmic and inverse transformation requires y′>0, the square root transformation and the logarithmic transformation are only effective for positive skewness, if the regularized data set meets the above requirements, it can be directly regularized.

[0036] In step 5, if the data does not meet the requirements, linear conversion or reverse conversion processing is required before selecting the corresponding transformation mode:

[0037] Square root transformation, y′=x max -x;

[0038] Logarithmic and inverse transformation, y′=x max -x+1

[0039] In the formula, y' is the processed data, x max is the maximum value in the data model, and x is the value in the data model; after processing according to the selected regular method according to the shape characteristics of the data set distribution, the processed data is drawn, the histogram and probability density curve of the corrected data set are calculated, the K, S and SE parameters are calculated, and it is judged whether it is approximated to the normal distribution; if there is still a certain degree of skewness, secondary regularization can be performed, and after the corrected data set is approximated to the normal distribution, the next step is entered.

[0040] In step 6, based on the regularized corrected data model, the upper and lower limits of the 90% confidence interval are calculated, and if the data set that has not been regularized has met the normal distribution, the 90% confidence interval can be directly used as the climate condition fluctuation interval; the 90% confidence interval of the regularized data set is calculated reversely, and the regularized process restores the upper and lower limits of the interval to the interval that meets the original data characteristics.

[0041] In step 6, if the original data model is not linearly or inversely converted, the data restoration methods corresponding to different regularization methods are as follows:

[0042] Square root transformation, x=y 2 ;

[0043] Logarithmic transformation, x=e y x=10 y

[0044] If the original data model is linearly or inversely converted, the data restoration methods corresponding to different regularization methods are as follows:

[0045] Square root transformation, x=x max -y 2 ;

[0046] Logarithmic transformation, x=x max -e y +1x=x max -10 y +1

[0047] The fluctuation interval of the climate condition can be obtained after restoration calculation.

[0048] Step 6 also includes generating a random data set approximated to the normal distribution according to the mean and standard deviation of the regularized model, which is consistent with the total number of spores, obtaining the original random data set through restoration calculation, calculating the mean and standard deviation of the original random data set, and showing the concentration trend and fluctuation degree of the climate condition.

[0049] The quantitative palaeoclimate recovery method based on spore-pollen data of the present application, based on limited palaeo-pollen data which does not cover all palaeoclimate conditions, restores the long-time-scale climate change interval by reconstructing a palaeoclimate distribution model conforming to or approximating to a normal distribution to restore the long-time-scale climate change interval. The method is based on the palaeoclimate distribution model, and the 90% confidence interval is used to recover the palaeoclimate change interval, effectively eliminating the interference of a small amount of foreign spores on the palaeoclimate recovery. The method has strong adaptability and expansion, and can be combined with the previous discrimination of spore source, based on the proportion of local and foreign spores, to adjust the scale of the confidence interval in time, and then eliminate the interference of foreign spores of different degrees on the palaeoclimate recovery. The principle of the method is clear and easy to understand, and the palaeoclimate recovery is quantified, and has high accuracy and operability. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 The flow chart of a specific embodiment of the quantitative palaeoclimate recovery method based on spore-pollen data of the present application;

[0051] Figure 2 The column chart and probability density curve chart of the MAT average distribution data model reconstructed in specific embodiment 1 of the present application;

[0052] Figure 3 The column chart and probability density curve chart of the MAT data model regularized in specific embodiment 1 of the present application;

[0053] Figure 4 The column chart and probability density curve chart of the MAP average distribution data model reconstructed in specific embodiment 2 of the present application;

[0054] Figure 5 The column chart and probability density curve chart of the MAP data model regularized in specific embodiment 2 of the present application. DETAILED DESCRIPTION

[0055] It should be noted that the following detailed description is exemplary in nature and is intended to provide further description of the present application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0056] It is to be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or groups thereof.

[0057] In botany, the climatic conditions of the original species and its closest relatives exhibit strong consistency. The extant closest relatives of archaeophores can indicate the climatic conditions of archaeophores, reflecting climatic conditions during geological history. Geological history spans a large timescale, resulting in a wide range of climatic conditions with significant fluctuations and overlaps. However, most random changes in nature conform to or approximate a normal distribution. Based on the climatic intervals and frequencies reflected by archaeophores during geological history, a statistical model conforming to or approximates a normal distribution can be constructed to obtain the distribution of climatic conditions over long timescales in geological history. The 90% confidence interval in the model represents a reliable range of long-term climatic changes. This invention adopts the following technical solution:

[0058] (1) Statistical analysis was performed on the ancient pollen data of the target layer to determine the total pollen content and the frequency of different types of pollen in the target layer.

[0059] (2) Determine the closest existing kinship species corresponding to different types of archaeolians, ascertain the living ecological conditions of the corresponding existing kinship species, and determine the ecological and climatic conditions indicated by different types of archaeolians.

[0060] (3) Based on the ecological and climatic conditions indicated by ancient pollen, and constrained by the frequency of ancient pollen occurrence, the statistical model that approximates the normal distribution is reconstructed through average distribution and regularization correction. The 90% confidence interval of the statistical model for different ecological and climatic conditions is obtained, and the 90% confidence interval of the statistical model is restored to the original data characteristics as the ecological and climatic conditions of the target layer period.

[0061] like Figure 1 As shown, Figure 1 This is a flowchart of the paleoclimate quantitative reconstruction method based on pollen data of the present invention, which specifically includes the following steps:

[0062] Step 1: Establish a basic database of ancient pollen.

[0063] Core wells with complete stratigraphic records were selected from the study area. The paleontological data of the target strata were statistically analyzed to determine the types of paleontology developed in the target strata, the frequency of occurrence of different types of paleontology, and the total number of paleontology, thus establishing a basic paleontological database (Table 1).

[0064] Table 1. Basic Paleopollen Database of Target Stratum from X1 Core Well

[0065]

[0066]

[0067] Step 2: Establish a comprehensive database of ancient pollen ecoclimatic conditions

[0068] Based on the types of fossil spores in the basic database of fossil spores, the corresponding present nearest relative species of fossil spores were determined through online Palaeoflora Database, Climbot Database and previous research results of spore pollen. According to the growth required paleoclimate parameters of present nearest relative species in the database, such as mean annual temperature (MAT), mean temperature of the coldest month (CMT), mean temperature of the warmest month (WMT), mean annual precipitation (MAP), mean precipitation of the wettest month (HMP) and mean precipitation of the driest month (LMP), the paleoclimate parameters indicated by fossil spores were determined, and the comprehensive database of paleoecological and climatic conditions of fossil spores was established (Table 2).

[0069] Table 2 Comprehensive database of paleoecological and climatic conditions of fossil spores in the target interval of X1 core

[0070]

[0071]

[0072] Step 3: Reconstruction of paleoclimate distribution model tending to normal distribution

[0073] Based on the comprehensive database of paleoecological and climatic conditions of fossil spores, the paleoclimate data model with average distribution was formed by taking the climate condition interval of different types of spores and its frequency as constraints. Here, the parameter MAT is taken as an example to detail the method:

[0074] First, the frequency of the end member temperature at both ends of the MAT interval of different types of spores is taken, and then the temperatures in the interval with the same frequency are traversed and superimposed with an increment of 0.1. The data model tending to normal distribution is established by the average distribution method. Taking Ulmipollenites sp as an example, the average distribution method is as follows: the MAT interval is -1.2-24.3℃, and the frequency is 67. The frequency of the end member temperature at both ends of -1.2 and 24.3℃ is 67. The 254 temperatures in the interval of -1.1-24.2℃ are traversed and superimposed with an increment of 0.1℃, and the frequency of each temperature is 67. In this way, the MAT data distribution model tending to normal distribution is established by using the average distribution method to process the MAT interval of different types of spores.

[0075] Principle:

[0076] Different types of pollen climate condition interval overlap, using the existing interval frequency to average distribution of interval values to construct data distribution model can restore the maximum degree of climate conditions span of long time scale in geological history on the basis of existing data. Since the average distribution of data model needs to be formed to the maximum extent to approach normal distribution, when doing average distribution of each interval, it is not through the interval and frequency to form the corresponding arithmetic sequence (interval end element as the sequence end element value, interval frequency as the sequence term number). Taking the MAT interval of Ulmipollenites sp as an example, this method forms an arithmetic sequence with a difference of 0.39 from -1.2℃ to 24.3℃ and a term number of 67 as the data distribution model. The above average distribution method weakens the constraint of interval frequency and highlights the control effect of overlapping interval. If you want to approach normal distribution, high frequency and overlapping interval should have high frequency in the reconstructed array, that is, you should pay attention to the double constraints of frequency and overlapping interval. The above method cannot make the model tend to be normal distribution well. The practical method in step 3 traverses all interval arrays with an increment of 0.1 to ensure the uniformity and integrity of interval data. On this basis, the frequency of all interval arrays is the corresponding interval frequency, which can ensure the high frequency of high frequency and overlapping interval data and make the data distribution model effectively approach normal distribution.

[0077] Step 4: Check the shape of the average distribution data model

[0078] Based on the average distribution data model formed, the data model can be graphed by using the probability density column and curve drawing module in various software, such as the matpotlib module in python language or the drawing module in spss. According to the left and right symmetry and the steepness of the peak state, determine which one of normal distribution, negative skewness, positive skewness, steep peak state or slow peak state is the data distribution form. If the graphing cannot effectively determine, the three parameters of kurtosis K, skewness S and standard error SE can be calculated. The calculation formulas of the three parameters are as follows:

[0079]

[0080]

[0081]

[0082] In the formula, K is the kurtosis; S is the skewness; SE is the standard error; SD is the standard deviation; n is the sample number, x i The i-th value in the data model is expressed as is the average value of the data model sample.

[0083] A large number of sample data models can be directly calculated by calling the embedded algorithms in the pandas and numpy libraries in the python language or using the corresponding modules in SPSS, and then the distribution state of the data model is determined.

[0084] Principle:

[0085] The left skewness frequency distribution has a high peak to the right, and the long tail gradually extends to the left from the right, and S<0, and the positive skewness is the opposite. The discrimination of the peak state is mostly based on kurtosis K, K>0 for steep peak state, and K<0 for gentle peak state. The discrimination of the degree of skewness and peakness can be based on K, S and SE parameters, and when the absolute value of K or S is greater than 1.96 times of the standard error, it has obvious significant skewness and peakness characteristics, and when the absolute value of K or S is 2-3 times of the standard error, the skewness and peakness reach a moderate degree, and when the absolute value of K or S exceeds 3 times of the standard error, the skewness and peakness reach a high degree.

[0086] Taking the MAT data as an example, the bar chart and probability distribution curve are drawn Figure 2 ), it can be seen that the data model has the characteristics of negative skewness, and the K, S, SE parameters of the data model are calculated to further determine the distribution state of the data model, K=-0.60, S=-0.36, SE=0.13. K and S are less than 0, and the absolute value of both is less than 1.96 times of SE, and the data model has low degree of negative skewness and gentle peakness characteristics.

[0087] Step 5: data model regularization

[0088] Based on the judgment of the shape distribution of the data model in step 4, the corresponding regularization method is selected to regularize the data model. If the data model is determined to be in accordance with the normal distribution in step 4, it can directly jump to step 6 to generate a random data set to calculate the 90% confidence interval, mean and standard deviation, and obtain the climate condition interval, central tendency and fluctuation degree.

[0089] Light to moderate skewness and peakness data adopts square root transformation,

[0090] High degree of skewness and peakness data adopts logarithmic transformation, y=lny′ or lgy′

[0091] Large fluctuation at both ends of the data adopts inverse transformation,

[0092] In the formula, y is the data set after regularization, and y′ is the data set that needs to be regularized. The square root transformation variable requires y′≥0, and the logarithmic and inverse transformations require y′>0. Square root transformation and logarithmic transformation are only effective for positive skewness. If the regularized data set meets the above requirements, it can be directly regularized. If the data does not meet the requirements, linear conversion or reverse conversion processing should be performed before selecting the corresponding transformation method:

[0093] Square root transformation, y' = x max -x;

[0094] Log and inverse transformation, y' = x max -x + 1

[0095] where y' is the processed data, x max is the maximum value in the data model, and x is the value in the data model. According to the shape characteristics of the data set distribution, the corresponding regularization method is selected for processing, and after processing the processed data, the histogram and probability density curve of the corrected data set are drawn, the K, S, and SE parameters are calculated, and it is judged whether it is close to the normal distribution. If there is still a certain degree of skewness, secondary regularization can be performed, and after the corrected data set is close to the normal distribution, the next step is entered.

[0096] Taking the MAT data as an example, the data model shows low negative skewness and slow peak characteristics, and the square root transformation is selected for processing. Since the data is negatively skewed and has negative values, it is processed by the inverse conversion process, and the conversion is converted into a regularizable data set by the formula y' = x max -x. After the conversion data set is processed by the square root transformation, the corrected data set is obtained, the corrected data set histogram and probability density interval Figure 3 ) are drawn, and the K, S, and SE parameters are calculated. The corrected data set K = -0.54, S = -0.19, and SE = 0.01. The graphical judgment shows that the corrected data is basically symmetrical, K and S data are reduced to a certain extent, SE is close to a minimum value due to the large sample size, and K / SE and S / SE do not support the judgment of the distribution shape. The data set has approached the normal distribution.

[0097] Step 6: Restore the characteristics of paleoclimate conditions

[0098] Based on the regularized corrected data model, the numpy library in python language is called to calculate the average value and standard deviation of the data model, and then the upper and lower limits of the 90% confidence interval are calculated by using the stats.norm.interval (confidence = 90%, loc = average value, scale = standard deviation) command in the scipy library, or the confidence interval module in SPSS can be directly realized by inputting the confidence. If the data set has not been regularized and meets the normal distribution, the 90% confidence interval can be directly used as the climate condition fluctuation interval. The 90% confidence interval of the regularized data set reverses the regularization process to restore the interval upper and lower limits to meet the original data characteristics. If the original data model is not linear or inverted, the data restoration method corresponding to different regularization methods is as follows:

[0099] Square root transformation, x = y 2 ;

[0100] Logarithmic transformation, x = e y x = 10 y

[0101] If the original data model is linearly or inversely converted, the corresponding data restoration methods of different regularization methods are as follows:

[0102] Square root transformation, x = x max -y 2 ;

[0103] Logarithmic transformation, x = x max -e y +1x = x max -10 y +1

[0104] After the restoration calculation, the fluctuation interval of the climate condition can be obtained. Further, according to the mean value and standard deviation of the regularization model, the random data set conforming to the approximate normal distribution corresponding to the total number of spores can be generated by calling the random. normal function, the original random data set is obtained by restoration calculation, and the mean value and standard deviation of the original random data set are calculated to show the concentration trend and fluctuation degree of the climate condition.

[0105] Taking the MAT data as an example, the average value of the data model after regularization is 3.85, the standard deviation is 1.1, the 90% confidence interval is 2.04-5.66, x max = 29.2℃, through the formula x = x max -y 2 The fluctuation interval of MAT is -2.84-25.03℃. Further, a random data set with a sample number of 224 is generated by using the mean value (3.85) and the standard deviation (1.1), and the original random data set is obtained by the formula x = x max -y 2 The average value of MAT is 13.18℃, and the standard deviation is 8.44.

[0106] Example 1:

[0107] The research object is a set of N group sedimentary rocks of X1 well in A depression of Paleogene. The spore and pollen data are counted to establish a basic database of paleo-pollen (Table 1), and then the existing nearest relative species of different types of spores and pollen are determined by using the Palaeoflora Database database and the previous spore and pollen research results, and the living climate condition interval (such as MAT) of different types of paleo-pollen is found out, and then a comprehensive database of paleo-pollen ecological climate condition is established (Table 2). With the MAT interval and frequency of different types of spores and pollen as constraints, the data distribution model is established by average distribution, and the histogram and probability density curve are drawn. Figure 2). The average distribution data model distribution shape was determined by graphical judgment of MAT, combined with K, S, and SE parameters, the regularization method was determined, the data model was corrected, and it was graphed Figure 3 ). The data model was determined to have approached normal distribution by Figure 3 , combined with K, S, and SE parameters, and the 90% confidence interval of the regularized data model was calculated to be 2.04-5.66 based on the mean (3.85) and standard deviation (1.1) of the regularized data model, and the corresponding method was selected to restore the upper and lower limits of the interval, and the MAT climate condition interval was obtained to be -2.84-25.03°C. Finally, a random data set with a sample size of 224 was generated based on the mean (3.85) and standard deviation (1.1), which was restored to the original random data set, and the average value of the MAT was calculated to be 13.18°C and the standard deviation was 8.44.

[0108] Example 2

[0109] The research object was the MAP (annual average precipitation) of the N group sedimentary rock layer of X1 well, and the database establishment process was described in the embodiments and example 1 and will not be repeated (Tables 1 and 2). The average distribution was established to build a data model, which was imaged, and K = -0.35, S = 0.58, and SE = 1.28 were calculated, and the model had slight positive skewness and slow peak characteristics Figure 4 ). The square root transformation was selected for regularization, and the data had no negative value and could be directly transformed, and then the regularized model was graphed, K = -0.53, S = -0.06, and SE = 0.02 were calculated, and the large sample size caused K / SE and S / SE to have no discriminant effect, and the skewness was obviously improved and had approached normal distribution Figure 5 ). The average value of the regularized data model was calculated to be 36.72 and the standard deviation was 11.27, the 90% confidence interval of the regularized data model was calculated to be 18.19-55.26, and the climate fluctuation interval was obtained to be 330.91-3053.41 mm by directly squaring the restoration. Based on the mean (36.72) and standard deviation (11.27), a random data set with a sample size of 224 was generated, which was restored to the original random data set, and the average value of the MAP was calculated to be 1491.86 mm and the standard deviation was 887.30.

[0110] Example 3

[0111] The research object was the CMT (average temperature of the coldest month), WMT (average temperature of the hottest month), HMP (average precipitation of the wettest month), and LMP (average precipitation of the driest month) of the N group sedimentary rock layer of X1 well, and the restoration results are shown in Table 3.

[0112] Table 3 Quantitative restoration data table of paleoclimate parameters of the target layer of X1 coring well

[0113] Climate parameter MAT (°C) CMT (°C) WMT (°C) MAP (mm) LMP (mm) HMP (mm) Mean 13.18 -1.32 22.36 1491.86 64.95 242.72 Standard deviation 8.44 15.62 5.80 887.30 42.33 138.39 95% confidence interval -2.84-25.03 -25.72-21.17 12.24-30.82 330.91-3053.41 7.12-146.66 56-460.05

[0114] Finally, it should be noted that the above only for the preferred embodiments of the present application, and is not intended to limit the present application, although with reference to the foregoing embodiments of the present application has been described in detail, for those skilled in the art, it still can be modified, or part of the technical features of the equivalent replacement. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application, should be included within the scope of the present application.

[0115] In addition to the technical features described in the specification, are known to those skilled in the art.

Claims

1. A method for quantitative paleoclimate reconstruction based on pollen data, characterized in that, This paleoclimate quantitative reconstruction method based on pollen data includes: Step 1: Establish a basic database of ancient pollen; Step 2: Establish a comprehensive database of ancient pollen ecological and climatic conditions; Step 3: Reconstruct a paleoclimate distribution data model that tends towards a normal distribution; Step 4: Check the distribution shape of the reconstructed data model; Step 5: Regularize the reconstructed data model; Step 6: Reconstruct paleoclimate characteristics; In step 4, based on the established average distribution data model, the data distribution pattern is determined to be normal, negatively skewed, positively skewed, steeply kurtotic, or gently kurtotic based on left-right symmetry and kurtosis steepness. In step 5, based on the determination of the shape distribution of the data model in step 4, the appropriate regularization method is selected to regularize the data model; if the data model in step 4 is determined to conform to a normal distribution, the process jumps directly to step 6 to generate a random dataset and calculate the 90% confidence interval, mean and standard deviation to obtain the climate condition interval, central tendency and degree of fluctuation.

2. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 1, characterized in that, In step 1, core wells with complete stratigraphic records in the study area are selected, and the paleontology data of the target strata are statistically analyzed to determine the types of paleontology developed in the target strata, the frequency of occurrence of different types of paleontology, and the total number of paleontology, thereby establishing a basic paleontology database.

3. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 1, characterized in that, In step 2, based on the types of ancient pollen in the basic database, the closest existing kinship species corresponding to the ancient pollen are determined. According to the paleoclimate parameters required for the growth of the closest existing kinship species in the database, the paleoclimate parameters indicating the ancient pollen are identified, and a comprehensive database of ancient pollen ecological climate conditions is established.

4. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 3, characterized in that, In step 2, the paleoclimate parameters indicated by the ancient pollen include the annual average temperature, the average temperature of the coldest month, the average temperature of the hottest month, the annual average precipitation, the average precipitation of the wettest month, and the average precipitation of the driest month.

5. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 1, characterized in that, In step 3, based on the comprehensive database of ancient pollen ecological climate conditions, a paleoclimate data model with an average distribution is formed, constrained by the ranges and frequencies of different types of pollen climate conditions.

6. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 1, characterized in that, In step 4, if graphical representation is insufficient, the distribution of the data model is further determined by calculating three parameters: kurtosis K, skewness S, and standard error SE. The formulas for calculating these three parameters are as follows: In the formula, K is kurtosis; S is skewness; SE is standard error; SD is standard deviation; n is the sample size, x i Describe the i-th value in the data model. This represents the average value of the data model samples.

7. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 6, characterized in that, In step 4, the peak of the left-skewed frequency distribution is skewed to the right, while the long tail gradually extends from the right to the left, and S < 0, while the positively skewed distribution is the opposite. The kurtosis is determined based on the kurtosis K, where K > 0 indicates a steep kurtosis and K < 0 indicates a gentle kurtosis. The degree of skewness and kurtosis is determined based on the parameters K, S, and SE. When the absolute value of K or S is greater than 1.96 times the standard error, there are obvious skewness and kurtosis characteristics. When the absolute value of K or S is 2-3 times the standard error, the skewness and kurtosis reach a moderate degree. When the absolute value of K or S exceeds 3 times the standard error, the skewness and kurtosis reach a high degree.

8. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 7, characterized in that, In step 5, When the absolute value of K or S in the distributed data model is between 1.96 and 3 times the standard error, a square root transformation is applied. When the absolute value of K or S in the distributed data model is greater than three times the standard error, a logarithmic transformation is applied, y = lny′ or lgy′. In the formula, y is the dataset after regularization, and y′ is the dataset that needs to be regularized; For square root transformation, the variable y′≥0, and for logarithmic and reciprocal transformations, y′>0. Square root transformation and logarithmic transformation are only effective for positively skewed data. If the dataset needs to be regularized to meet the above requirements, regularization should be performed directly.

9. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 8, characterized in that, In step 5, if the data does not meet the requirements, a linear or inverse transformation of the data model is required before selecting the corresponding transformation method: Square root transformation, y′=x max -x; Logarithmic transformation, y′=x max -x+1 In the formula, y′ represents the processed data, and x max Let x be the maximum value in the data model and x be the numerical value in the data model. Based on the distribution shape characteristics of the dataset, select the appropriate regularization method for processing. After processing the data, draw the histogram and probability density curve of the corrected dataset, calculate the K, S, and SE parameters, and determine whether it approximates a normal distribution. If there is still a certain degree of skewness, perform secondary regularization. After the corrected dataset approximates a normal distribution, proceed to the next step.

10. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 9, characterized in that, In step 6, based on the data model after regularization correction, the upper and lower limits of the 90% confidence interval are calculated. If the dataset conforms to a normal distribution without regularization, the 90% confidence interval is directly used as the climate condition fluctuation interval. For the 90% confidence interval of the regularized dataset, the regularization process is reversed to restore the upper and lower limits of the interval to the interval that conforms to the characteristics of the original data.

11. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 10, characterized in that, In step 6, if the original data model has not undergone linear or inverse transformation, the data restoration methods corresponding to different regularization methods are as follows: Square root transformation, x = y 2 ; Logarithmic transformation, x = e y x = 10 y If the original data model has undergone linear or inverse transformations, the data restoration methods corresponding to different regularization methods are as follows: Square root transformation, x = x max -y 2 ; Logarithmic transformation, x = x max -e y +1 x=x max -10 y +1 After the calculation is restored, the fluctuation range of climate conditions is obtained.

12. The method for quantitative paleoclimate reconstruction based on pollen data according to claim 11, characterized in that, Step 6 also includes generating a random dataset that approximates a normal distribution and matches the total number of pollen based on the mean and standard deviation of the regularization model, obtaining the original random dataset through restoration calculation, and calculating the mean and standard deviation of the original random dataset to show the central tendency and degree of fluctuation of climate conditions.

Images