Incomplete well log data are very commonly encountered problems in petroleum exploration activity. The development of artificial intelligence technology offers a new possible way to predict the required logs using limited information available. Optimizing conventional statistical theory, machine learning is proven to be a reliable tool for any prediction task in many fields of study. Regression is one of the basic methods that has rapid development and evolved many techniques with different approaches and purposes. In this study, parametric and nonparametric regressions {linear regression, Support Vector Machine (SVM), and Gaussian Process Regression (GPR)} are compared to predict the missing log using the available nearby data.

Feature selection was done by performing Principal Component Analysis (PCA) on predictor variables. Different profile of PCA is observed between Cibulakan and Parigi Formations, which is the basis of conducting separate models based on the formation. Among all the selected methods, GPR is consistently making slightly better results. The correlation between the predicted and actual porosity of GPR is observed to be up to 0.19 higher compared to the other methods. Similar observation is also found on the Root Mean Squared Error (RMSE) value comparison. In practice, the GPR method has an inherent advantage compared to other methods, as it provides uncertainty to the prediction based on the standard deviation of each estimation result. The standard deviation of the GPR prediction ranges from 0.006 in high confidence cases and up to 0.077 where uncertainty is high. The models are considered robust and stable according to the RMSE evaluation from cross validation which is consistently giving the value below 0.04. In conclusion, the reliability of regression techniques for predicting the missing well log is exposed in this study, which results demonstrate steady and good accuracy in every formation which are tested on any well logs.

Earthquake insurance is an insurance that covers the risks that occur when buildings are damaged/destroyed due to an earthquake. In order to determine adequate premium for earthquake insurance, a grouping method based on building locations is needed. In Indonesia, the method for grouping the building locations is based on zones where several locations have not been classified to the district/municipality level. In other words, the zone grouping system still has a limited resolution. In addition to detailing the grouping, this paper also aims to make the level of earthquake vulnerability in West Java. The approach used is to make a density map based on earthquake intensity scale (MMI) and magnitude historical data. From these density map, the value of building exposure damage are calculated using the Mean Damage Ratio (MDR) parameter. This value is used as a pricing parameter for determining earthquake insurance premiums in more detail for each location. The results of this study indicate that the levels of damage to buildings are varied and earthquake insurance premiums have a high degree of non-uniformity in each municipality/district in this province.

The East African Rift System (EARS) is widely recognized as a major tectonic system that has excellent geothermal resources. A recent study in southern Africa found that there might still be rifting in the Passarge Basin in central Botswana, suggesting that the East African Rift System (EARS) is extended. This research was done to examine the implication of the extension of the EARS to the geothermal prospect that could be used as the preliminary information for further exploration in Botswana.

In this research, the Curie point depth (CPD) and heat flow map of Botswana were produced from aeromagnetic data. The method used was the spectral analysis (1DPSD) which considered as the ideal approach for a regional CPD study. This method has several parameters dependency, the pycurious algorithm was applied to overcome this issue. A window size 400 km with increments of 50 km was used in the algorithm to produce the CPD deeper than 40 km (based on the crustal thickness estimated in this region). The CPD and heat flow maps were compared to the existing geophysics and geological information to examine the implication of the extension of EARS to the geothermal prospect of Botswana. The prospective zones were delineated according to the heat flow value and the geological condition.

The result indicates that the CPD appears to be consistent with the previously published crustal thickness model for the northern part of Botswana. The shallower CPD zones (<15km) are coincident with the thinner crust zones from the model (<40 km). These zones were distributed along the geologically weak area, from the northern part which is represented by the Ghanzi-Chobe Belt to the south-western part represented by the Nosop Basin. The heat flow map created was able to distinguish the high heat flow (>65 mW/m2) area that was comparable to the heat flow value in the well-developed geothermal field in the western branch of EARS. Two prospective geothermal zones were identified based on the anomalous heat flow, divided by the possible tectonic condition that caused the anomaly. The first zone was interpreted as the result of the continuation of the EARS, the other zone was interpreted as the result of the older tectonic activity in Early Proterozoic era or the current tectonic activity as indicated by the recent earthquake in this area.

Ambient noise tomography is relatively a new method for imaging the shallow structure of the Earth subsurface. We presents the application of this method to produce a Rayleigh wave group velocity maps around the Merapi Volcano, Central Java. Rayleigh waves group velocity maps were reconstructed from the cross-correlation of ambient noise recorded by the DOMERAPI array which consists 43 broadband seismometers. In the processing stage, we first filtered the observation data to separatethe noise from the signal that dominated by the strong volcanic activities. Next, we cross-correlate the filtered data and stack to obtain the Green’s function for all possible station pairs. Then we carefully picked the peak of each Green’s function to estimate the dispersion trend and appliedMultiple Filter Technique to obtain the dispersion curve. Inter-station group velocity curvesare inverted to produceRayleigh wave group velocity maps for periods 1 to 10 s. The resulted Rayleigh group velocity maps show the interesting features around the Merapi Volcano which generally agree with the previous studies. Merapi-Lawu Anomaly (MLA) is emerged as a relatively low anomaly in our group velocity maps.

Seismic data has a broad range of lateral area recorded in time domain. Generally the seismic interpretation is done in time domain. Doing stratigraphic interpretation in time domain is usually sufficient. On the other hand, doing structure interpretation in time domain has a higher risk of pitfall. Time-to-depth conversion is one way to eliminate structure ambiguity in time domain and verify the structure. In this study, authors used the first-arrival travel time tomography method in modeling velocity for time-to-depth conversion. This method used the first-arrival travel time of seismic data. The travel time data, combined with the initial velocity model, became the input for travel time tomography. Tomographic process consisted of two processes: modeling and inversion of travel time. Modeling the travel time itself included travel time calculation and raytracing. The travel time inversion produced velocity model. The velocity model was then used in time-to-depth conversion. Then the result of the conversion was validated with VSP data. Validation of the velocity model showed good results. Hence, this method can be used in building velocity model which does not have data of time versus depth, such as checkshot or VSP data, for time-to-depth conversion.