ABSTRAK Karakterisasi reservoir menjadi penting dalam tahapan eksplorasi minyak dan gas bumi. Salah satu hal yang dibutuhkan untuk mencapai keakuratan dalam mengkarakterisasi reservoir adalah penampang seismik yang sesuai dengan penampang aslinya. Struktur lapisan bumi yang kompleks mengakibatkan gelombang terdifraksi, sehingga penampang seismik mengalami pembelokan dari posisi sebenarnya. Penelitian ini menerapkan metode migrasi seismik Kirchhoff dan Stolt (F-K) untuk mengembalikan posisi reflektor pada waktu dan kedalaman yang sebenarnya pada data seismik 2D di Perairan Utara Bali. Data seismik diintegrasikan dengan data sumur APS-1 sebagai kontrol untuk diinversikan dengan teknik inversi berbasis model sehingga dapat mengkarakterisasi reservoir.  Penelitian ini bertujuan membandingkan hasil migrasi seismik yaitu migrasi Stolt dan migrasi Kirchhoff untuk diinversikan menggunakan metode inversi berbasis model sehingga dapat diketahui sejauh mana kualitas data seismik mempengaruhi proses karakterisasi reservoir. Nilai korelasi dari hasil analisis regresi antara log impedansi inversi dengan log impedansi data sumur pada migrasi Kirchhoff sebesar 0,739 dan galat regresi sebesar 873,54, sedangkan pada migrasi Stolt memiliki nilai korelasi sebesar 0,698 dan nilai galat sebesar 1236,17. Hal ini menunjukkan bahwa migrasi Kirchhoff lebih baik dari migrasi Stolt baik secara kualitatif maupun kuantitatif dalam mengkarakterisasi reservoir hidrokarbon. ABSTRACTReservoir characterization is an important method in gas and oil exploration. In order to obtain accuracy for defining reservoir, required seismic image that similar to the actual seismic image. The complexity of earth structure could cause diffracted waves, therefore, seismic image was diffracted from its actual position. This study applies Kirchhoff and Stolt (F-K) seismic migration methods to restore the position of the reflector at the actual time and depth  seismic data in North Bali. Seismic data is integrated with APS-1 well data as controls to be converted with model-based inversion techniques so as to characterize the reservoir. This study aims to compare the results of seismic migration namely Stolt and Kirchhoff migration to be converted using a model-based inversion method so that it can be seen to what extent the quality of seismic data influences the reservoir characterization process. Correlation value from the results of regression analysis between inversion log impedance and well impedance log data in Kirchhoff migration is 0.739 and regression error is 873.54, while the Stolt migration has a correlation value of 0.698 and an error value of 1236.17. This shows that Kirchhoff's migration is better than Stolt migration both qualitatively and quantitatively in characterizing hydrocarbon reservoirs.

No abstract

Seismic method is a good geophysical method in imaging the subsurface conditions using the principle of seismic wave propagation. This method is often used in hydrocarbon exploration. One important step in hydrocarbon exploration is seismic interpretation. In the stages of seismic interpretation, a good basic knowledge of geophysical and geological knowledge is needed regarding the existence and characterization of hydrocarbon reservoirs. One method used in interpreting seismic data is the acoustic impedance inversion method. In this study, 2D seismic inversion was carried out to determine the reservoir characteristics of the MCL-1 well in the Nias basin. This study uses model-based on inversion which aims to obtain the value of acoustic impedance which is useful for the identification of distribution, porosity values and reservoir conditions of the target zone. The results obtained are the target reservoir zone at a depth of 6649-7434 feet or 1705-1810ms for MCL-1 wells with a range of acoustic impedance values of 25556 ((ft/s) * (g/cc)) - 46885 ((ft/s) * (g/cc)) with the type of rock that fills the reservoir is the type of limestone rock. The correlation value for model-based inversion has a relatively small error. This can characterize the hydrocarbon reservoir well.

Some deepwater multiple attenuation processing methods have been developed in the past with partial success. The success of surface multiple attenuation relies on good water bottom reflections for most deepwater marine situations. It brings the bigger ability to build an accurate water bottom multiple prediction model. Major challenges on 2D deepwater seismic data processing especially such a geologically complex structure of Seram Sea, West Papua – Indonesia are to attenuate surface related multiple and to preserve the primary data. Many multiple attenuation methods have been developed to remove surface multiple on these seismic data including most common least-squares, prediction-error filtering and more advanced Radon transform.Predictive Deconvolution and Surface Related Multiple Elimination (SRME) method appears to be a proper solution, especially in complex structure where the above methods fail to distinguish interval velocity difference between primaries and multiples. It does not require any subsurface info as long as source signature and surface reflectivity are provided. SRME method consists of 3 major steps: SRME regularization, multiple modeling and least-square adaptive subtraction. Near offset regularization is needed to fill the gaps on near offset due to unrecorded near traces during the acquisition process. Then, isolating primaries from multiples using forward modeling. Inversion method by subtraction of input data with multiple models to a more attenuated multiple seismic section.Results on real 2D deepwater seismic data show that SRME method as the proper solution should be considered as one of the practical implementation steps in geologically complex structure and to give more accurate seismic imaging for the interpretation.Keywords : multiple attenuation, 2D deepwater seismic, Radon transform, Surface Related Multiple Elimination (SRME). Banyak metode atenuasi pengulangan ganda dikembangkan pada pengolahan data seismik dengan tingkat keberhasilan yang rendah pada masa lalu. Keberhasilan dalam atenuasi pengulangan ganda permukaan salah satunya bergantung pada hasil gelombang pantul pada batas dasar laut dan permukaan pada hampir seluruh survei seismik laut. Hal tersebut menentukan keakuratan dalam membuat model prediksi pengulangan ganda dasar laut dan permukaan air. Tantangan utama dalam pemrosesan data seismik 2D laut dalam khususnya struktur geologi kompleks seperti Laut Seram, Papua Barat – Indonesia adalah pada kegiatan menekan pengulangan ganda permukaan sekaligus mempertahankan data primer. Beberapa metode yang dikembangkan untuk menghilangkan pengulangan ganda permukaan pada data seismik seperti least-square, filter prediksi kesalahan dan transformasi Radon.Dekonvolusi Prediktif dan Metode Surface Related Multiple Elimination (SRME) digunakan sebagai solusi yang baik pada struktur kompleks dimana metode-metode lain gagal untuk memisahkan perbedaan kecepatan interval data primer dan pengulangan ganda. Metode tersebut tidak membutuhkan informasi bawah permukaan selain parameter sumber dan reflektivitas permukaan. Metode SRME terdiri dari 3 tahapan utama : regularisasi SRME, pemodelan pengulangan ganda dan pengurangan adaktif least-square. Regularisasi near offset diperlukan untuk mengisi kekosongan pada near offset yang disebabkan oleh adanya sejumlah tras terdekat yang tidak terekam selama akuisisi. Pemodelan maju digunakan untuk memisahkan data primer dan pengulangan ganda kemudian inversi dengan pengurangan input data dengan model multiple.Hasil pada data seismik 2D laut dalam menunjukkan bahwa metode SRME layak diterapkan sebagai salah satu pengembangan metode atenuasi multiple permukaan serta untuk meningkatkan akurasi data seismik terutama untuk struktur geologi kompleks.Kata kunci : peredaman pengulangan ganda (multiple), seismik 2D laut dalam, transformasi Radon, Surface Related Multiple Attenuation (SRME).

Real marine seismic data are typically embedded with free-surface multiples energy, which are troublesome in imaging an accurate seismic cross-section. In addition, more challenging situation is to bring optimum result with a short-offset streamer due to the coherent nature of multiples. In this study, we present a comparison of three methods for attenuating free-surface multiples energy in short-offset 2D seismic data from Cendrawasih Bay. Multiple attenuation methods include F-K filter, Radon transform, and Surface Related Multiple Elimination (SRME) are processed until the final Pre-Stack Time Migration (PSTM) results. Predictive deconvolution is applied in order to suppress short period multiples prior to free-surface multiple attenuation method. Predictive deconvolution successfully identifies and removes the predictable wavelet of short period multiples. Radon transform shows poor result in short-offset seismic data even if it is combined with F-K filter method. Combination of both F-K filter and SRME are successfully attenuate free-surface multiples and should be considered as proper solution to increase signal to noise ratio.