We introduce a concept of generalized blending and deblending, develop its models and accordingly establish a method of deblended-data reconstruction using these models. The generalized models can handle real situations by including random encoding into the generalized operators both in the space and time domain, and both at the source and receiver side. We consider an iterative optimization scheme using a closed-loop approach with the generalized blending and deblending models, in which the former works for the forward modelling and the latter for the inverse modelling in the closed loop. We applied our method to existing real data acquired in Abu Dhabi. The results show that our method succeeded to fully reconstruct deblended data even from the fully generalized, thus quite complicated blended data. We discuss the complexity of blending properties on the deblending performance. In addition, we discuss the applicability to time-lapse seismic monitoring as it ensures high repeatability of the surveys. Conclusively, we should acquire blended data and reconstruct deblended data without serious problems but with the benefit of blended acquisition.@en

Recently, we established a generalized blending model, which can explain any methods of blended acquisition by including the encoding into the generalized operators. With this highly flexible and tolerant model, we come up with a challenging question: what it is to be, and how to find an optimal blended-acquisition design, which should be the most suitable for deblended-data reconstruction among plenty of concepts of blended acquisition. In this paper, we introduce a method of blended-acquisition encoding: temporally modulated and spatially dispersed source array, namely M-DSA, that jointly uses modulation sequencing in the time dimension and dispersed source array in the space dimension. This allows quite straightforward deblending by filtering and physically separating frequency channels in the frequency domain. We run our blended-acquisition designing based on the deblending performance for several scenarios of blended acquisition. These examples show that: M-DSA attains the best deblending performance; this method has less constraints in the encoding with more operational flexibility, compared to other methods being developed in the industry today. Indeed, this method requires only simple signaturing in the encoding; merely frequency-banded and modulated signatures in the time dimension for each shot in the blended-source array. This could even render any other blending properties unnecessary. Those, such as distance separation among shot locations and time shifts among shot times, might not be required anymore. There might be no limitation on the number of sources, thus no limitation on the blending fold, in order to secure successful deblending. Furthermore, this method allows random sampling; randomly distributed sources in the space dimension in the blended-source array. Consequently, this method makes the blended-acquisition encoding and operations significantly simple and robust, as well as for the deblending processing. We believe that our M-DSA method should be one of the best methods of blended acquisition.@en

Blended acquisition along with efficient spatial sampling is capable of providing high-quality seismic data in a cost-effective and productive manner. While deblending and data reconstruction conventionally accompany this way of data acquisition, the recorded data can be processed directly to estimate subsurface properties. We establish a workflow to design survey parameters that account for the source blending as well as the spatial sampling of sources and detectors. The proposed method involves an iterative scheme to derive the survey design leading to optimum reflectivity and velocity estimation via joint migration inversion. In the workflow, we extend the standard implementation of joint migration inversion to cope with the data acquired in a blended fashion along with irregular detector and source geometries. This makes a direct estimation of reflectivity and velocity models feasible without the need of deblending or data reconstruction. During the iterations, the errors in reflectivity and velocity estimates are used to update the survey parameters by integrating a genetic algorithm and a convolutional neural network. Bio-inspired operators enable the simultaneous update of the blending and sampling operators. To relate the choice of survey parameters to the performance of joint migration inversion, we utilize a convolutional neural network. The applied network architecture discards suboptimal solutions among newly generated ones. Conversely, it carries optimal ones to the subsequent step, which improves the efficiency of the proposed approach. The resultant acquisition scenario yields a notable enhancement in both reflectivity and velocity estimation attributable to the choice of survey parameters.@en

We introduce a blended-acquisition method: temporally signatured and/or modulated and spatially dispersed source array, namely S-/M-DSA. The former S-DSA has much less constraints in the encoding with operational flexibility, allowing non-uniform sampling and non-patterned shooting both in the space and time dimension. The latter M-DSA allows indeed straightforward deblending by filtering and physically separating frequency channels in the frequency domain. We evaluated the deblending performance for several scenarios of blended acquisition. The results showed that: S-DSA attains the best acquisition productivity; M-DSA attains the best deblending performance, compared to other methods. Our S-/M-DSA method makes the blended-acquisition encoding and operations significantly simple and robust; the same is true for the deblending processing.@en

The application of blended acquisition has drawn considerable attention owing to its ability to improve the operational efficiency as well as the data quality and health, safety and environment performance. Furthermore, the acquisition of less data contributes to the business aspect, while the desired data density is still realizable via subsequent data reconstruction. The use of fewer detectors and sources also minimizes operational risks in the field. Therefore, a combined implementation of these technologies potentially enhances the value of a seismic survey further. One way to encourage this is to minimize any imperfection in deblending and data reconstruction during processing. In addition, one may derive survey parameters that enable a further improvement in these processes as introduced in this study. The proposed survey design workflow iteratively performs the following steps to derive the survey parameters responsible for source blending as well as the spatial sampling of detectors and sources. The first step is the application of blending and sampling operators to unblended and well-sampled data. We then apply closed-loop deblending and data reconstruction. The residue for a given design from this step is evaluated and subsequently used by genetic algorithms to simultaneously update the survey parameters related to both blending and spatial sampling. The updated parameters are fed into the next iteration until they satisfy the given termination criteria. We also propose a repeated encoding sequence to form a parameter sequence in genetic algorithms, making the size of problem space manageable. The results of the proposed workflow are outlined using blended dispersed source array data incorporating different scenarios that represent acquisition in marine, transition zone and land environments. Clear differences attributed solely to the parameter design are easily recognizable. Additionally, a comparison among different optimization schemes illustrates the ability of genetic algorithms along with a repeated encoding sequence to find better solutions within a computationally affordable time. The optimized parameters yield a notable enhancement in the deblending and data reconstruction quality and consequently provide optimal acquisition scenarios.@en

We introduce a generalized concept of blending and deblending, establish its models, and accordingly establish a method of deblended-data reconstruction using these models. The generalized models can handle real-life situations by including random encoding into the generalized operators both in the space and time domain, and both at the source and receiver side. We consider an iterative optimization scheme using a closed-loop approach with the generalized-blending and -deblending models, in which the former works for the forward modelling and the latter for the inverse modelling in the closed loop. We established and applied this method to existing real datasets acquired in Abu Dhabi. The results show that our method succeeded to fully reconstruct deblended data even from the fully generalized, thus quite complicated blended data@en