Top‐Down Versus Bottom‐Up Atmospheric Emission Estimates From Oil and Natural Gas Operations

This study reviews results from literature to investigate top‐down versus bottom‐up emission estimates from oil and natural gas (O&NG) operations. Ten years ago, a landmark study by Brandt et al. (2014, https://doi.org/10.1126/science.1247045 ) found generally higher top‐down emissions determinations than industry estimates. Here, we revisit this topic by examining 10 years of peer‐reviewed literature that has been published since. A total of 73 results for top‐down/bottom‐up ratios from 57 articles were included. Most of the published literature focuses on methane emissions, with only 11 articles reporting other O&NG emissions. In the newer literature, 49 (86%) of the studies reported that inventory (bottom‐up) emissions underestimated results from determinations based on ambient observations (top‐down). This finding is similar to the Brandt et al. (2014) study, which found that 82% of literature reported higher top‐down emissions than inventory‐based estimates, suggesting little improvement in inventory data accuracy over the past decade despite this discrepancy having been well documented earlier. However, fewer extreme ratio values (top‐down/bottom‐up) >10 were reported after 2014. A meta‐analysis building on carefully selected literature that has been published since 2014 resulted in a mean ratio value of 2.50 ± 0.62, implying that measured emissions were on average 250% of inventory values. For North American countries, the mean ratio values ranged from 2.59 to 3.75, exceeding the global average. The lowest ratio values were observed when United Nations Framework Convention on Climate Change (UNFCCC) reports were used as inventory comparisons to top‐down measurements (mean = 1.45); all other inventory types resulted in mean ratios >2.