# Applying Spatial Bootstrap and Bayesian Update in uncertainty assessment at oil reservoir appraisal stages

## TL;DR

This paper introduces a combined spatial bootstrap and Bayesian update method to evaluate uncertainty in reservoir property estimates, accounting for sparse data and conceptual model variability in oil reservoir appraisal.

## Contribution

It proposes an integrated approach using spatial bootstrap and Bayesian updating to improve uncertainty assessment in reservoir modeling.

## Key findings

- Enhanced uncertainty quantification for reservoir estimates
- Ability to compare different conceptual models
- Improved reliability of reservoir property predictions

## Abstract

Geostatistical modeling of the reservoir intrinsic properties starts only with sparse data available. These estimates will depend largely on the number of wells and their location. The drilling costs are so high that they do not allow new wells to be placed for uncertainty assessment. Besides that difficulty, usual geostatistical models do not account for the uncertainty of conceptual models, which should be considered. Spatial bootstrap is applied to assess the estimate reliability when resampling from original field is not an option. Considering different realities (conceptual models) and different scenarios (estimates), spatial bootstrapping applied with Bayesian update allows uncertainty assessment of the initial estimate and of the conceptual model. In this work an approach is suggested to integrate both these techniques, resulting in a method to assess which models are more appropriate for a given scenario.