(1. Guangzhou Marine Geological Survey Guangzhou 510760; 2. China Geo University (Beijing) Beijing 100083)
Introduction to the first author: jinliang, male, 197 1 born, senior engineer, 1995 graduated from the department of information engineering and geophysics of Chengdu University of Technology, majoring in applied geophysics, mainly engaged in natural gas hydrate investigation and research.
In this paper, Jason inversion technique is used to calculate the P-wave velocity of line A on the northern slope of the South China Sea, and combined with the analysis of various hydrate occurrence information such as BSR, amplitude blank zone and waveform polarity inversion, the velocity characteristics of hydrate metallogenic belt are comprehensively studied. The results show that high-speed anomaly under low-speed background is an important feature of natural gas hydrate occurrence. High-speed anomalies are generally distributed in strips parallel to the seabed; Inside the high-speed anomaly, the speed is constantly changing. Generally, the velocity is the highest in the center of the abnormal body, and gradually decreases from the center to the edge, which is reflected in the hydrate zone, and the hydrate saturation gradually decreases from the center to the edge of the ore body. The research results of this paper further show that high-precision velocity analysis is not only helpful to discover hydrate occurrence, but also to further determine hydrate enrichment horizons.
Jason inversion technique; Natural gas hydrate; velocity analysis
1 preface
Natural gas hydrate is a caged structure compound composed of water-ice crystal skeleton and natural gas molecules adsorbed in it under low temperature and high pressure environment, which is widely distributed in seabed and frozen soil zone. Temperature and pressure are the most important factors for the formation and preservation of natural gas hydrate (Wang Hongbin et al., 2004). The field investigation and study of natural gas hydrate show that the high-resolution seismic exploration method is an effective method for the investigation and evaluation of natural gas hydrate. Seismic inversion technology has always been the core technology in seismic exploration. Its purpose is to infer the distribution of underground wave impedance, velocity, porosity and other parameters by using seismic reflection data, so as to estimate the parameters of gas-bearing layer of natural gas hydrate, predict the distribution of natural gas hydrate and provide reliable basic data for natural gas hydrate exploration. Commonly used seismic inversion techniques include Jason, Strata, Seislog and Islamic State, among which Jason inversion technique is widely praised for its high resolution in predicting gas-bearing hydrate layers, mainly including two methods with and without well constraints (Liao et al., 2002).
Abnormal velocity is one of the important conditions to judge whether natural gas hydrate exists. Combining BSR (Submarine Simulated Reflector) characteristics, waveform polarity characteristics, amplitude characteristics and AVO characteristics has become the main means to judge whether there is a gas hydrate layer (Dou Shi et al., 1999). A large number of experimental data show that the velocity of hydrate is close to that of ice, but higher than that of water. Compared with water-bearing or free gas-bearing sediments, the density of hydrate-bearing sediments decreases and the sound velocity increases. The formation speed of hydrate-bearing sediments is often higher than the general formation speed, and the lower part of hydrate-bearing sediments is filled with water or gas, which leads to the negative velocity anomaly at the bottom interface of hydrate. Therefore, formation velocity inversion is a geophysical sign of hydrate existence. The sound velocity of hydrate-bearing formation is related to the hydrate content, and the higher the hydrate content, the higher the sound velocity. In terms of velocity, BSR is the interface between the overlying high-speed hydrate-bearing formation and the underlying low-speed aquifer or gas-bearing formation. The seismic velocity of P-wave in shallow marine sediments is generally1600 ~1800 m/s. If there is hydrate, the seismic velocity will be greatly increased, reaching1850 ~ 2500 m/s. If there is a free gas layer below the hydrate layer, the seismic velocity can be sharply reduced by 200 ~ 500 m/s. The change trend of formation speed of hydrate layer is typical of Mesoyaha gas field in Siberia. The data show that after hydrate is formed in the original water-bearing sand layer, the propagation speed of P-wave will increase from 1850m/s to 2700m/s.. In cemented sandstone layer, this speed will increase from 3000 m/s to 3500 m/s. The logging results of 570 stations in the deep-sea drilling plan show that when the water-bearing sandstone layer enters the hydrate-bearing sandstone layer, the density decreases from1.79g/cm3 to1.19g/cm3, and the acoustic wave propagation speed increases from1700m/sec to 3600m/sec.
VSP logging data of ODP889 station in Cascadia sea area show that the bottom boundary of hydrate is a strong negative velocity interface, and the velocity drops sharply from 1900m/s of hydrate deposit to 1580m/s of free gas layer. VSP logging is seismic logging, which is less affected by drilling factors, so it is considered that VSP logging truly reflects the velocity change at the bottom boundary of hydrate deposits (Chen Jianwen et al., 2004).
From 200/kloc-0 to 2004, Guangzhou Marine Geological Survey of the Ministry of Land and Resources conducted a high-resolution seismic survey of natural gas hydrate over 10000 km on the northern slope of the South China Sea. In this study, Jason inversion technology is used to analyze the seismic velocity data in the northern slope of the South China Sea, and on the basis of delineating the distribution range of BSR, the velocity characteristics of each sedimentary layer in the slope area are studied. Finally, the relationship between velocity value and hydrate is analyzed and discussed.
2 Method principle
The density of natural gas hydrate (0.9g/cm3) is close to that of seawater, and the content of free gas is very limited, which determines that the difference of wave impedance of BSR is mainly caused by velocity. The characteristic of velocity inversion technology is to extrapolate and interpolate all seismic event axes under the control of seismic interpretation horizon without well constraint, so as to complete wave impedance inversion, thus overcoming the limitation of seismic resolution, optimally approaching logging resolution and keeping good lateral continuity of inversion results. The main principles of velocity inversion technology are as follows: ① A series of reflection coefficients with sparse characteristics are obtained by maximum likelihood deconvolution; ② Wave impedance is obtained by maximum likelihood inversion; ③ Calculate the velocity with wave impedance. The main advantage of this method is that it can obtain broadband reflection coefficient, and it is a model-based inversion with many modeling methods. By comparing and analyzing the established models, the geological model is more reasonable and the inversion results are more reliable (Hao Yinquan et al., 2004).
The starting point of wave impedance inversion method is that the reflection coefficient of underground is sparsely distributed, that is, the reflection coefficient of stratum is composed of a series of strong axes superimposed on Gaussian background. The specific inversion is to extract reflection coefficients from seismic traces according to the sparse principle, and generate synthetic seismic records through wavelet convolution. The residual error between synthetic seismic record and original seismic trace is used to correct the reflection coefficient, and a new reflection coefficient sequence is obtained, and then the wave impedance is obtained. The specific steps are as follows:
It is assumed that the reflection coefficient of the stratum consists of the reflection of the large reflection interface and the small reflection of the Gaussian background, and a minimum objective function is derived according to this assumption (An Hongwei et al., 2002):
Geological research in the South China Sea. 2006
Where: R(K) is the reflection coefficient of the first sampling point, m is the number of reflection layers, l is the total number of samples, n is the square root of the noise variable, and λ is the likelihood value of the given reflection coefficient.
Maximum likelihood inversion is the process of deriving broadband wave impedance by converting reflection coefficient. If the reflection coefficient formula R(t) is obtained by maximum likelihood deconvolution, the wave impedance is:
z(I)= z(I- 1)×( 1+R(I))/R( 1-I)(2)
Using the relationship between wave impedance and velocity:
v=Z(i)/ρ (3)
You can get the speed value. Where ρ is the formation density, which can be obtained by inversion of regional logging data combined with survey line gravity data.
In the above process, in order to obtain a reliable estimate of reflection coefficient, wave impedance information can be input separately as a constraint condition to obtain the most reasonable velocity model. On the one hand, the velocity inversion result is a broadband reflection sequence, wave impedance and velocity data, and low-frequency components are added to make the inversion result reflect the velocity variation law more accurately; On the other hand, it has many quality control methods, such as the selection of monitoring wavelet, the continuous tracking of in-phase axis, the judgment of the accuracy of inversion results and the related analysis of various intersection displays. Therefore, velocity inversion can be used to invert any phase of seismic profile, and any in-phase axis velocity data can be obtained at each CDP point, and a highly continuous velocity profile can be obtained by using the two-dimensional velocity tomography inversion method of reflected waves. If the seismic lines are dense enough, the velocity volume image can be obtained by using three-dimensional velocity inversion.
3 implementation process
3. 1 Establishment of initial model
Under the guidance of geological laws, the sedimentary characteristics and sedimentary cycles are analyzed by using seismic and logging data; Establish the relationship between rock and electricity, and compare the sand layer group and single sand layer; The reflection wave attribute of each oil-bearing sand group is extracted from the seismic profile, and the relationship between seismic genus and ore body is established, so as to realize the comprehensive prediction of the plane distribution thickness of ore body by earthquake and logging, and the extrapolation prediction of interlayer ore bodies is carried out. Establishing an initial velocity field; Seismic inversion is carried out under the constraint of seismic attributes, and the thickness of interlayer sub-layer ore bodies is inversed. Whether the calibration of subdivision inversion horizon is correct or not directly affects the accuracy of inversion results. Therefore, it is very important to study wavelet extraction, energy spectrum characteristics, signal-to-noise ratio, spectrum and reflection coefficient in the inversion process (Yan Kuibang et al., 2004). The technical route flow is as shown in figure 1:
3.2 Acquisition of initial velocity field
To obtain the initial velocity field, the velocity spectrum must be explained first. Whether the interpretation and value of velocity spectrum are reasonable will directly affect the calculation accuracy of root mean square velocity. The specific steps are as follows:
1) velocity spectrum interpretation starts from the profile with simple geological conditions, good quality of reflection layer, strong energy cluster and less interference, and draws the superposition velocity-reflection time curve, which gradually expands outward;
2) According to the reflection characteristics of seismic profile, judge whether the velocity extreme point is correct, and select the extreme point with the largest reading energy cluster. Eliminate the energy cluster of interference wave, so as to get the superposition velocity of effective wave;
3) Compare the adjacent velocity spectra, and check and correct by comparing the shape of the velocity spectrum curve with the velocity extreme value of the same reflection layer.
4) Pick up a set of data every 40 CDP, and use the reflection dip data on the seismic profile for correction to get the root mean square velocity (jinliang et al., 2006).
Figure 1 Circuit Flow Diagram of Velocity Inversion Technology
Figure 1 Flow chart of technical route speed inversion
3.3 Wavelet extraction
When extracting wavelet, the energy should be concentrated on the main lobe of wavelet, which is consistent with the shape of seismic wavelet. If the extracted wavelet is close to zero phase, the energy decays rapidly from the peak to both sides, and the waveforms on both sides of the peak are symmetrical; In the analysis of wavelet energy spectrum characteristics, energy should be concentrated in the main frequency range of seismic waves; When there are well data, we should control the quality of automatic correlation between wavelet and seismic wave. It is an important aspect of inversion to ensure that the wavelet energy spectrum is consistent with the seismic wave energy spectrum, and the peak value of wavelet energy spectrum is consistent with the peak value of seismic wave spectrum. First, understand the deviation between synthetic records and seismic records. By analyzing the deviation between synthetic records and seismic records, Jason's reflection coefficient deviation and energy spectrum deviation are further corrected to reduce the deviation between synthetic records and seismic records. Then, by analyzing the deviation between the reflection coefficient and the seismic data, corresponding measures are taken to correct it, so that the reflection coefficient of the stratum is consistent with that of the synthetic record. Then the signal-to-noise ratio is analyzed, so that the signal-to-noise ratio after inversion is improved to the greatest extent. Through a series of quality control measures, the calibration accuracy of synthetic records and seismic records of various reservoirs has been greatly improved.
The credibility of velocity inversion cannot be completely determined by inversion method, but the key lies in the quality of seismic records and the amplitude fidelity of the pre-inversion processing flow. Another influence factor is that the numerical simulation results are more accurate, which is related to the calculation method, wavelet picking and geological structure model. As for the sensitivity of inversion results, it is mainly judged by fitting error and convergence speed. If the given initial model is correct, that is, consistent with the actual geological structure, the fitting error is small and the convergence speed is fast. Because the work in this paper is limited by the actual situation and there is no actual logging data verification, the accuracy and precision of inversion velocity will be affected to some extent.
4 Velocity profile characteristics
Comprehensive analysis using a variety of special seismic images is the key technology for seismic data interpretation of natural gas hydrate. At present, the corresponding seismic characteristics of natural gas hydrate, such as BSR, amplitude blank zone, waveform polarity inversion, abnormal velocity, wave impedance and AVO, are generally used to comprehensively analyze whether the sediments contain hydrate. High-precision interval velocity analysis is helpful to determine hydrate-rich horizons. The abnormal structure of velocity and amplitude is a special image formed by the interaction between hydrate and underlying free gas, which shows a "uplift and depression" structure on the profile, and multiple layers overlap to form an obvious vertical "bright spot". This special imaging structure is more suitable for finding hydrates in undeformed hydrate basins, and can be used to quantitatively estimate the number of hydrates in hydrate basins and analyze the detailed velocity structure above and below BSR.
Fig. 2 integration profile of track a of the survey line on the northern slope of the south China sea
Fig. 2 Integral Profile of Line A on the North Slope of the South China Sea
Fig. 2 is an integral profile of seismic reflection trace of line A on the northern slope of the South China Sea. It can be seen from the figure that there is a strong amplitude reflection wave in the middle and lower right corner of the profile about 350ms away from the seabed, which is roughly parallel to the seabed reflection wave and inclined to the stratum, and has obvious BSR characteristics. In terms of waveform polarity, both submarine reflection wave and BSR show strong amplitude and double peaks, submarine reflection wave shows blue-red-blue, while BSR shows red-blue-red, which indicates that BSR shows negative reflection in-phase axis relative to the seabed, that is, the so-called polarity inversion (opposite to submarine reflection). The polarity of the reflected wave is determined by the reflection coefficient of the reflection interface, which is related to the wave impedance difference between the two sides of the interface. In fact, the seabed and BSR are both strong wave impedance surfaces, and the seabed is the interface between seawater and surface sediments. The upper part is a low-speed layer, the lower part is a relatively high-speed layer, and the reflection coefficient is positive. BSR is the interface between hydrate-bearing layer and lower layer (or gas-bearing layer). The upper layer is a high-speed layer (the hydrate metallogenic belt is a relatively high-speed body), and the lower layer is a relatively low-speed layer (if it contains free gas, the velocity is low), and the reflection coefficient is negative, thus causing the phenomenon that the polarity of BSR and seabed reflection waves are opposite (Sha Zhibin et al., 2003). Fig. 3 is a P-wave velocity profile inverted by velocity inversion method, which obviously shows a relatively high-speed geological body approximately parallel to the seabed, and its position is just above BSR. The P-wave velocity of high-speed geological body is about 2000 ~ 2400 m/s, the P-wave velocity of its upper low-speed layer is about 1500 ~ 1800 m/s, and the P-wave velocity of its lower low-speed layer is about1500 ~1900 m/s.
Fig. 3 P-wave velocity profile of slope line A in the northern South China Sea calculated by velocity inversion method.
Fig. 3 p velocity profile of line a on the north slope of the south China sea calculated by velocity inversion
As can be seen from Figure 3, the internal velocity of the hydrate metallogenic belt changes, indicating that the hydrate distribution is uneven, parallel to the seabed, with the highest central velocity and gradually decreasing from the center to the edge. There are three low-speed zones and high-speed zones approximately parallel to the seabed: ① The relatively low-speed zone between the seabed and the high-speed body is a water-saturated zone; ② Hydrate metallogenic belt; ③ Low velocity zone under the hydrate metallogenic belt. The low velocity zone under the hydrate metallogenic belt has no obvious low velocity characteristics in the velocity profile, so it is inferred that there may be no free gas or low gas saturation under the hydrate metallogenic belt.
5 conclusion
The formation of hydrate requires not only certain temperature and pressure conditions, but also a lot of hydrocarbon gas and enough water. This requires the formation to have high porosity and permeability. Unconsolidated sedimentary rocks have high porosity and permeability and physical conditions for hydrate formation. The seismic wave velocity of loose sedimentary rocks with this characteristic is low, while the seismic wave velocity of hydrate-bearing strata increases. This has formed a hydrate metallogenic belt as a high-speed geological body under the background of low speed. In addition, the formation of hydrate is controlled by temperature and pressure. Generally, the isothermal surface and the isobaric surface are approximately parallel to the seabed, so the relatively high-speed geological bodies are approximately parallel to the seabed under the background of low speed, which is the characteristic of the hydrate metallogenic belt (Liu et al., 2003).
Through the calculation of P-wave velocity of A-line on the northern slope of the South China Sea, combined with the comprehensive analysis of BSR, amplitude blank zone identification and waveform polarity inversion, we can further understand the velocity characteristics of hydrate metallogenic belt: it is revealed that the high-speed anomaly in hydrate metallogenic belt is generally distributed in a strip parallel to the seabed, and the internal velocity of high-speed anomaly is constantly changing, generally the highest in the anomaly center, and the velocity gradually decreases from the center to the edge. This phenomenon is reflected in the hydrate metallogenic belt, and the hydrate distribution is uneven. Analyzing the detailed velocity structure above and below BSR is an important means to comprehensively interpret hydrate seismic data. High-precision velocity analysis is helpful to determine the rich horizon of hydrate, and it is more suitable to find the occurrence state of hydrate, and the hydrate resources can be estimated accordingly.
refer to
An Hongwei, Li Zhengwen, Li Renfu, et al. 2002. Application of sparse pulse wave impedance inversion in XY oilfield development. Petroleum geophysical exploration, 4 1 (1): 56 ~ 60.
Yan, Wu Zhiqiang, etc. 2004. Geophysical identification mark of natural gas hydrate. Marine geodynamics, 6: 9 ~ 12.
Hao Yinquan, Pan Mao and Li Zhongquan. 2004. Application of Jason multi-well constrained inversion technology in oil and gas reservoir prediction. Journal of Chengdu University of Technology, 3 1 (3): 297 ~ 300.
Jinliang, Wang Hongbin and Guo Yiqun. 2006. Study on seismic velocity of natural gas hydrate on the northern slope of South China Sea [J]. Modern Geology, 20 (1):123 ~129.
Liao, Ma Bo, et al. 2002. Prediction of sand body and gas-bearing property with Jason software. Natural gas exploration and development, 25 (3): 34 ~ 42.
Liu,,, et al. 2005. Study on seismic response of natural gas hydrate-application example of China Nanhai HD 152 line. Modern geology, 19 (1): 33 ~ 38.
Sha Zhibin, Yang Muzhuang, Liang Jinqiang, et al. 2003. Characteristics of BSR reflection wave and its application in gas hydrate identification. Geological research of South China Sea,15 (1): 55 ~ 61.
Dou Shi, Zheng Junwei. 1999. Present situation and prospect of research and development of natural gas hydrate in the world. Progress in Earth Science, 14: 330 ~ 339.
Wang Hongbin, jinliang, Gong Yuehua, et al. 2005. Calculation of submarine heat flow on the northern slope of South China Sea based on seismic data of natural gas hydrate. Modern geology, 19 (1): 67 ~ 73.
Yan Kuibang, Li Dongmei, Wu Koizumi. 2004. Application of Jason inversion technology in lithology identification. Petroleum Geophysical Exploration, 43 (1): 54 ~ 58.
Zhang Optics, Huang Yongyang, Chen Bangyan, editor. 2003. Seismology of natural gas hydrate in sea area. Beijing: Ocean Press.
Zhang optics, Wen Pengfei. 2000. Study on seismic characteristics of methane hydrate in the South China Sea, Proceedings of the first Guangdong Young Scientists Forum, China Science and Technology Press.
Application of Jason inversion technology in gas hydrate velocity analysis
Liang Jin 1 Wang Hongbin 1, 2 Liang Jinqiang 1
(1. Guangzhou Marine Geological Survey, Guangzhou, 5 107602. China Geo University (Beijing), Beijing, 100083)
Abstract: The P-wave velocity of the seismic profile on the northern slope of the South China Sea is calculated by Jason inversion method. According to the calculation results and the information of natural gas hydrate, including BSR, amplitude blank and polarity inversion, the velocity characteristics of natural gas hydrate layer are studied in detail. The research shows that the high-speed anomaly under the background of low speed is an important feature of the existence of natural gas hydrate; High-speed anomalies are usually distributed in strips parallel to the seabed; Within the anomaly with high velocity, the velocity changes gradually, the highest velocity appears at the center of the anomaly, while the lower velocity appears at the edge of the anomaly, which indicates that the saturation of natural gas hydrate gradually decreases from the center to the edge. The above results show that high-resolution velocity analysis is not only helpful to find hydrate points, but also helpful to estimate the enrichment layer of natural gas hydrate.
Keywords: Jason inversion technology gas hydrate velocity analysis