METHOD AND SYSTEM FOR MEASURING SUBSIDENCE

Abstract

A method for measuring subsidence and/or uprise on a field, comprises the steps of: deploying at least one cable on a solid surface; collecting inline tilt data from numerous tilt sensors deployed along each cable (100); and performing a statistical analysis on the tilt data to determine changes in curvature on the solid surface. Preferably, the statistical method involves computing a cumulative inline and/or cross-line tilt, whereby random errors cancel and systematic changes add. In addition, regression and/or interpolation may provide a quantitative estimate of curvature etc.

Claims

1-10. (canceled)

11. A method for measuring subsidence and/or uprise on a field, comprising the steps of: deploying at least one cable on a solid surface; collecting inline tilt data from numerous tilt sensors deployed along each cable; and performing a statistical analysis on the tilt data to determine changes in curvature on the solid surface.

12. The method according to claim 11, wherein the statistical analysis involves computing a cumulative inline tilt as a sum of collected tilt data from tilt sensors disposed along one cable.

13. The method according to claim 12, further comprising the step of adding several cumulative inline tilts.

14. The method according to claim 11, wherein the statistical analysis involves computing a cumulative cross-line tilt as a sum of collected tilt data from tilt sensors disposed along one cross-line extending perpendicular to several essentially parallel cables.

15. The method according to claim 14, further comprising the step of adding several cumulative cross-line tilts.

16. The method according to claim 11, further comprising the step of repeating the steps at predetermined intervals.

17. The method according to claim 11, further comprising the step of performing a regression analysis on the tilt data in order to obtain an estimate of a curvature on the solid surface.

18. The method according to claim 11, wherein a sign of tilt data is conserved to provide a difference between subsidence and uplift.

19. A system measuring subsidence and/or uprise on a field, comprising several cables with seismic stations arranged at regular intervals, each seismic station comprising a tilt sensor and the cables being arranged essentially parallel in an array, wherein each seismic station is connected through the array at base station and an umbilical to a control unit for performing the method of claim 11.

20. The system according to claim 19, wherein the solid surface is a seafloor above a subsurface formation to be monitored.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The invention will be explained by means of examples and reference to the drawings, in which:

[0022] FIG. 1 illustrates basic principles of the invention.

DETAILED DESCRIPTION

[0023] FIG. 1 shows a system 1 comprising a control unit 2 providing power and two-way communication to a seismic array 5 through an umbilical 3 and a base station 4. The seismic array 5 is deployed at a solid surface, e.g. a seafloor, and comprises cables 100 running essentially in the horizontal direction denoted x. A cable 10 on the surface essentially perpendicular to the cable 100, i.e. in the direction denoted y, connects the cables 100 to the base station 4.

[0024] Each cable 100 provides several seismic stations 110, 140 with power and communication. Seismic stations 110, 140 are placed along the entire length of each cable 100, but most of them are omitted from FIG. 1 for clarity of illustration. The direction along cables 100 is termed inline, and the horizontal direction perpendicular to the cables is termed cross-line. Typically, the inline distance between sensor stations 110, i.e. along the cables 100 are 50 m. The cross-line distance, i.e. between cables 100 is typically in the range 200-500 m. For simplicity of illustration, all cables 100 run in the x-direction. Dashed lines 20 through the sensor stations 110 in the y-direction are not physical connections, but illustrates that the seismic array 5 may be mapped to a polygonal mesh representing the solid surface. If desired, e.g. for computational purposes, the quadrilateral mesh may be represented by a triangular mesh in a known manner. In either case, the seismic stations 110 or 140 are located at corners of the mesh.

[0025] FIG. 1 also illustrates consequences of subsidence such that the cables 100 sinks to new positions illustrated by dashed lines 101. More particularly, a point 150 on the solid surface subsides a distance dz in the vertical direction z. The shift dz at vertex 150 will shift an adjacent seismic station 140 downward, e.g. to the position illustrated by the dotted circle below seismic station 140. The shift dz also increases the tilt 141 in the inline direction at seismic sensor 140 by an angle . A corresponding downward shift is shown at vertex 151 of the grid, and a change of tilt in the cross-line direction y is illustrated by an angle .

[0026] For useful subsidence measurements, vertical displacement less than 10 cm should be detectable. Thus, 50 m between seismic stations in the inline direction corresponds to an angle <arctan(10.sup.2/50)=0.2. Similarly, a cross-line spacing of 200 m corresponds to <0.06 and a cross-line spacing of 500 m corresponds to <0.02.

[0027] It is possible to detect subsidence by mapping a polygonal mesh to the solid surface, e.g. the seafloor above a formation, and monitoring the mesh in a time-lapse sequence. In this case, tilt sensors within the seismic stations 110, 140 could provide spatial derivatives in the x and y-directions. If each tilt sensor is able to detect tilt changes less than 0.06 and the spacing of the cables 100 is less than 200 m, then the edges of the mesh are easily determined. In addition or alternatively, the corners of the mesh may be determined by pressure sensors capable of detecting pressure changes less than approximately 10.sup.1m/(10 m/bar)=0.01 bar.

[0028] However, the tilt sensors within the seismic stations 110, 140 are generally not designed with the accuracy discussed above. Similarly, some or all seismic stations 110, 140 may lack pressure sensors with the required sensitivity and/or means to filter noise in pressure data due to waves on the surface.

[0029] However, it may be possible to use statistical analysis to cancel out presumed stochastic variations in accuracy of the tilt sensors already present in the seismic stations 110, 140. If so, it will also be possible to provide those seismic stations 110, 140 that do not already have tilt sensors with relatively inexpensive tilt sensors, typically based on MEMS accelerometers.

[0030] Returning to FIG. 1, it is seen that the tilt 141 at sensor 140 is changed due to the greater curvature in the x-z plane after subsidence, i.e. after the downward shift dz at vertex 150. Thus, if tilt is measured as a deviation from the horizontal direction as indicated by arrow 141, the sum of all tilts in the x-z plane taken along the dotted line 101 will be greater than the same sum taken along the solid line 100. Furthermore, this assumption holds even if a real cable 100 deviates from the x-z plane, i.e. curves slightly in the x-y plane. In other words, a sum of deviations in the inline direction is equivalent to a sum in the x-direction. Depending on the implementation of the tilt sensors, this difference may or may not obviate a scalar product between a measured tilt and a unit vector in the x-direction, or a similar trigonometric computation, to obtain the tilt direction in the x-z plane. The sum of tilts along one cable 100 will be termed a cumulative inline tilt in the following.

[0031] From FIG. 1 it is also apparent that similar changes in curvature due to subsidence occur at the cable 100 running through vertex 151, and in other cables. A sum of cumulative inline tilts of several or all cables 100 is expected to be an even better measure of change of curvature, i.e. presence of subsidence, as the summed tilt difference grows systematically if the solid surface has subsided, while random inaccuracies in the tilt sensors continue to cancel each other.

[0032] A similar argument applies to the cross-line direction. The angle implies a greater tilt in the y-z plane, which is equivalent to the cross-line direction. The sum of tilts along one cross-line 10, 20 is termed a cumulative cross-line tilt, and a sum of cumulative cross-line tilts of several or all cross-lines 10, 20 is expected to provide a better indication of subsidence than each individual cumulative cross-line tilt.

[0033] In short, the sum of cumulative inline tilts, possibly added to the sum of cumulative cross-line tilts, provides a fast and accurate indication of the presence of subsidence. Obviously, the presence of an uprise could be determined in the same manner.

[0034] Alternatively or additionally, there may be a desire to map the solid surface by means of inexpensive tilt sensors rather than just determine the presence of subsidence or uplift as discussed above. It is readily seen that regression analysis or known interpolation techniques can be employed inline and cross-line to obtain estimates for edges of the polygonal mesh, and hence quantitative estimates for curvature etc., using the ideas discussed above.

[0035] So far, the basic observed variable, i.e. tilt, has been described as deviation from a horizontal plane, i.e. the x-y plane in FIG. 1, for ease of explanation. However, conventional tilts, i.e. deviation from a vertical, work equally well, as displacing all angles by 90 or /2 changes the sums, but does not change the basic ideas. Furthermore, any basic variable measuring the different curvature of inlines 100 and 101 and/or the cross-lines can be used without changing the basic ideas of obtaining cumulative sums inline and/or cross-line, and then summing the cumulative sums. Thus, the term tilt as used herein should be understood as any such basic variable that can be derived from tilt sensor measurements, and is not limited to angular deviation from a horizontal as in the previous example.

[0036] The direction of tilt must of course be preserved in order to detect a difference between subsidence and uplift, whereas a sum involving squared basic variables may be employed if only subsidence or only uplift are of interest. Also, partial sums may be used if some part of the solid area is prone to uplift and other parts are prone to subsidence. Selecting suitable basic variables and constructing appropriate sums are considered well within the capabilities of one skilled in the art knowing the present disclosure and knowing the application at hand.

[0037] Thus, while the invention has been described by way of examples, the scope of the invention is determined by the accompanying claims.