Skip to main content Skip to secondary navigation

Data set generation and uncertainty analysis

Main content start
AI-Driven Environmental Data Extraction for Energy Sector Assessment (Chen, et al., 2024).

Data set generation and uncertainty analysis are critical components in assessing oil and gas carbon intensity. Researchers from our group have developed sophisticated methods to create comprehensive databases and quantify uncertainties in emissions estimates.

Zhang et al. (2021) created a detailed geographic database of oil and gas fields to estimate global flaring volumes with associated uncertainties. This approach allowed for oilfield-specific estimates, improving the granularity of emissions data.

Brandt (2020) evaluated the accuracy of satellite-derived flaring volume estimates for offshore operations, comparing them with reported data to assess uncertainties in remote sensing techniques.

Vafi and Brandt (2014a, 2014b) conducted extensive work on uncertainty analysis in oil field greenhouse gas emissions. They employed Monte Carlo approaches to address information gaps and quantify uncertainties in emissions estimates. Their research also examined the reproducibility of life-cycle assessment models for crude oil production, highlighting variabilities in results due to data limitations and methodological choices.

Brandt et al. (2015) focused on addressing information gaps in regional-average petroleum GHG intensities through targeted data gathering, demonstrating how strategic data collection can reduce uncertainties in carbon intensity estimates.

Recent work by Chen et al. (2024) introduces AI-driven methods using large language models to extract oil and gas asset information from diverse sources, potentially improving data set generation for carbon intensity assessments.

Related Publications