|
|
 |
|||||||||||||||||
|
|||||||||||||||||
 |
|||||||||||||||||
| JOURNAL HOME | HELP | CONTACT PUBLISHER | SUBSCRIBE | ARCHIVE | SEARCH | TABLE OF CONTENTS |
SPECIAL SECTION: Hydrogeophysics |
Southwest Research Institute
South Florida Water Management District
Schlumberger Water Services
Corresponding author: jparra{at}swri.edu
High-resolution crosswell seismic reflection data were acquired at a Port Mayaca aquifer test site about 30 miles west of the Atlantic Ocean and approximately one mile east of the eastern boundary of Lake Okeechobee in Martin County, Florida (Figure 1). Measurements were taken between monitoring wells MF-37 and EXPM-1 at an interwell separation of 1300 ft using a Z-Seis piezoceramic X series source and a 10-level hydrophone system. Multiple source and detector measurements were taken in the interval from 400 to 1700 ft (Figure 2). The objectives of the survey were to map flow-unit variability in the region between the two wells, to assess whether the high-resolution seismic survey could be resolved, detect zones of high-water production, and to map the matrix porosity/permeability.
To reach these objectives, we integrated geophysical well-log data to generate impedance, permeability, and porosity images of the interwell region as well as to provide an interpretation with FMI resistivity images and petrophysical data. We used selected P-wave velocity and density logs and the crosswell reflection data with the band-limited inversion algorithm to generate impedance images. The algorithm allows constraint of the impedance inversion to match the well-log impedance in the vicinity of both wells. Away from the wells, impedance is constrained by lateral continuity. To relate the impedance changes to the hydrological characteristics of the aquifer, we constructed crossplots of impedance at the borehole locations versus the porosity and permeability well logs. These plots were selected on the basis of the lithological information. Regression equations relating impedance to flow properties at the boreholes were used to produce permeability and porosity images at the crosswell seismic scale. Finally, we produced spectral analyses of the zero vertical-offset crosswell data to identify high-attenuation zones that might be related to heterogeneous permeable zones.
Petrophysics
The lithologic and geophysical logs and packer test results from well MF-37 indicate moderate water production capacity of the upper Floridan aquifer from 755 to 1300 ft. The aquifer's intermediate confining unit (from 146 to 755 ft) is formed by clay, silt, and mudstone, while the aquifer system is formed by wackestone and packstone in the interval from 750 to 1050 ft, and wackestone and mudstone below 1050 ft (Figure 3).
|
The top of the Floridan Aquifer System (FAS) is part of the Arcadia Formation in the region of 755–790 ft. The FAS consists of moderately indurated packstone and grainstone units containing shell fragments and phosphate sands and silts. The dual induction, sonic, and caliper logs indicated a competent low porosity unit at 755–790 ft, where the resistivity increased from 12 to 40 
-m.
A sharp formation contact between the Arcadia Formation and the Suwannee limestone at 790 ft was identified by a change in lithology from dark gray to yellowish-gray packstone. This discontinuity was associated with a decrease in formation resistivity and sonic velocity (Figure 4). It is intercepted at 750 ft by well MF-37 and at 790 ft by well EXPM-1.
|
Seismic inversion
We inverted the incident reflection seismogram (Figure 2) for impedance using the VP and density logs and a band-limited method. We constrained the impedance using the velocity and density logs of wells EXPM-1 and MF-37, and we performed the control inversion using linear programming. The inversion was done trace by trace, and required reprocessing including data checks and possible corrections. The resulting impedance image (Figure 4) shows the variation of the pore structure and other features more clearly than the reflection or tomography data alone. The derived impedance from sonic and density logs was superimposed on the impedance image to illustrate crosswell seismic resolution versus well-log resolution.
In the regions slightly above 800 ft and 1090 ft (Figure 4), we observed a light blue continuous background that corresponded to a low-impedance formation; superimposed on this background are more rigid heterogeneities represented by green features. Slightly above 900 ft, we observed a low-impedance zone (low velocity and density) that is continuous between both wells and softer in the region near well EXPM-1. Below this zone there is an increase in impedance that can be observed in both well logs. This 25-ft zone is formed by two separate structures (green) that converge into one unit 800 ft from well MF-37. At this point there are still discontinuities that slightly separate the two structures. The presence of this single unit can be observed in the impedance well-log signature in well EXPM-1, which is represented by a single maximum impedance. The third important feature observed in the impedance image is the high-impedance unit (in red) above 800 ft. This is a continuous zone that can be considered a low-permeability confining unit or permeability barrier. We used the impedance distribution with the porosity and permeability well logs to generate porosity and permeability images, respectively.
Porosity image
The porosity logs correlate nicely with several horizontal streaks (resistive units) in both wells (Figure 5). This comparison shows that the resistive units are connected between the wells. While the FMI image logs are capable of capturing these features, the porosity image does not distinguish them, but it does reveal lateral variations in matrix porosity. The highly porous interval at 800–900 ft is continuous, with porosities ranging from 0.38 to 0.5 volume fraction (Figure 5). Near well EXPM-1, a high-porosity anomaly (red) correlates with a high-conductivity FMI interval at 890–910 ft. The high FMI conductivity intervals are controlled by saturated porous zones containing brackish water. Such conditions were observed in water samples from the wells and are more pronounced in well EXPM-1 than well MF-37.
|
In the carbonate, the secondary porosity is much higher than in the rest of the formation; this may be due to the brittle limestone conditions (Figure 5). However, this zone was not identified as a higher water-production flow unit, which suggests that the secondary porosity peaks in this 25-ft interval are near the region surrounding the borehole only, and not interconnected with the matrix in the interwell region. The same characteristics are observed in well EXPM-1. However, in the region below 930 ft, we observe a more uniform FMI total-porosity log that extends all the way to 1090 ft. In this interval the secondary porosity makes a strong contribution to water production, particularly in well MF-37.
Permeability image
To further investigate matrix interconnectivity, we analyzed permeability based on the FMI data. The permeability was estimated using the Timur-Coates equation and data sets from both wells. This equation estimates secondary porosity-based permeability by considering the estimated fractions of macroporosity (large pores or vugs) and matrix porosity from the FMI data. The permeability log captures the fluid-flow conduits that cannot be resolved with the sonic data. We attempted to correlate the resulting permeability log with the impedance, but we found that the two logs do not correlate. This was not surprising because P-wave velocity data have much lower resolution than permeability data based on the FMI data. The derived permeability image (Figure 6) was integrated with the permeability logs (including the Timur-Coates permeability) from both wells.
|
Since water production appears to be low in the 900–930 ft interval, it is possible that higher-permeability values captured by the FMI permeability logs are localized in the region near the borehole and do not extend to well EXPM-1. The two zones of high-water production are better visualized by red in the permeability image than any other image (Figure 6). Variability of matrix permeability allows integration of the high-resolution well-log information by evaluating how secondary porosity can contribute to high-water production in the interwell region intervals from 800 to 900 ft and 930 to 1050 ft.
Spectral analysis
We extracted zero vertical-offset waveforms from the cross-well seismic data set to identify zones of high attenuation (Figure 7a). In the interval from 900 to 950 ft, the waveforms are practically absent (strongly attenuated), and the corresponding spectral plot shows energy losses (Figure 7b). A comparison of the waveforms and the logs illustrates that the high attenuation corresponds to strong variability in the porosity log from 900 to 950 ft. In this interval, the FMI image possesses a horizontal sequence of black and white streaks. The white streaks are thin resistive units, and the black streaks are highly conductive features. The resistive units are tight limestone or dolomite, and the highly conductive features are porous and saturated with brackish water. The highly attenuated zone affected processing of the Stoneley wave signal and prevented the use of permeability logs for generating the permeability image region between wells EXPM-1 and MF-37 (i.e., the formation's heterogeneous conditions were not suitable for extracting a reliable permeability log from the sonic data).
|
To make use of the high-resolution permeability logs, we compared the porosity/permeability images with the permeability logs based on FMI at a 3-inch resolution. These plots showed zones captured by the crosswell data as well as those that were not captured. Correlation of logs from both wells showed a degree of continuity in the small features. The plots helped us identify connected zones between the wells and determine how the permeable heterogeneities are embedded in the formation matrix near the wells. Our interpretation allowed us to estimate matrix permeability and porosity variations at the interwell scale.
The high-resolution FMI logs revealed secondary porosity features associated with high-permeability zones and high-water production in the FAS at intervals from 800 to 900 ft and from 930 to 1090 ft. The crosswell images revealed variability in matrix porosity and permeability, which led us to conclude that interconnectivity exists between the secondary and primary porosities in the interwell region. However, in the interval from 900 to 930 ft, the FMI images indicated that the secondary porosity does not have good interconnectivity with the matrix porosity. Additional indicators of this are that water production is lower in this interval, and the interwell image is heterogeneous and discontinuous.
Crosswell seismic reflection data give information on the intrinsic properties of a formation, but cannot resolve small features such as conduits and fractures seen in the FMI data. The FMI data can capture such features at a resolution of from 1 to 3 inches, which is better than can be accomplished with sonic log data. For example, in the FMI (Figure 5), the dark features observed in well EXPM-1 correspond to lower electrical resistivity and to fluid-filled pores/vugs/fractures or clay-reach zones.
The zero vertical-offset data exhibit zones of high attenuation that correlate with the FMI data. In particular, formation variability at the borehole wall is reflected in high-attenuation characteristics. This suggests that wave attenuation is sensitive to the presence of conduits. In fact, the macroporosity, which is a combination of fractures, vugs, collapse features, conduits and other large pore structures, is visible with the FMI. In the interval from 900 to 950 ft, these features contribute to scattering attenuation that can be observed in the zero vertical-offset waveforms and spectra. This heterogeneous low-velocity zone is delineated as a green feature in the crosswell permeability and impedance images. In the zone of high permeability or high water production, the zero vertical-offset waves are not strongly attenuated, and any contribution to the waveform characteristics is due to the intrinsic permeability or fluid-flow effects. These scattering and intrinsic effects can be explained by modeling the wave phenomena in the interwell region.
Conclusion
High-resolution seismic impedance delineated the upper confining unit contact intercepted by wells MF-37 and EXPM-1 at depths of 750 and 790 ft, respectively. This impedance image captures the variability of the soft and stiff zones in the interwell region. The impedance data with permeability and porosity well logs allow the generation of porosity and permeability images in the interwell region. These images visualize the flow units of high water production and low-permeable zones or permeability barriers. On the other hand, FMI images provide information on the secondary porosity distribution at the borehole scale not detected with any other log. The sequences of conductive and resistive bands captured by the FMI images are in the stiff zones of the aquifer. The combined interpretation of FMI and crosswell images allows an improved understanding as to the origin of the high-permeability spikes developed near the well that are not significant conduits.
Suggested reading
Hydrogeologic Investigation of the Floridan Aquifer System, Port Mayaca Site, Martin County, Florida by Bennett and Recrenwald (Preliminary Report, SFWMD, West Palm Beach, 2002). "Recovery of the acoustic impedance from reflection seismograms" by Oldenburg et al. (GEOPHYSICS, 1983), "Permeability and porosity images based on NMR, sonic, and seismic reflectivity: Application to a carbonate aquifer" by Parra et al. (TLE, 2003). "Permeability and porosity images based on P-wave surface seismic data: Application to a south Florida aquifer" by Parra et al. (Water Resources Research, 2006).
Acknowledgments:
The authors acknowledge the South Florida Water Management District for its support. We especially thank Michael Bennett of the SFWMD for his constructive feedback.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| JOURNAL HOME | HELP | CONTACT PUBLISHER | SUBSCRIBE | ARCHIVE | SEARCH | TABLE OF CONTENTS |
