Reestablishing Logistic Growth Curve Modeling of Energy Production

Logistic growth curve modeling of energy production has literally had its ups, “peaks”, and downs. The use of the logistic equation for modeling energy production began in 1956 when M. King Hubbert first fit a logistic... [ view full abstract ]

Logistic growth curve modeling of energy production has literally had its ups, “peaks”, and downs. The use of the logistic equation for modeling energy production began in 1956 when M. King Hubbert first fit a logistic curve to US crude oil production which accurately timed the 1970 peak of US crude oil production. It was not Hubbert’s intent to predict the exact timing (or magnitude) of the peak, rather he was using the bell-shaped logistic curve to demonstrate the shape a finite resource production tend would take. Hubbert’s other suggestion that nuclear and renewable energy production trends would take the shape of the s-shaped logistic curve over time has also proven valid.

In the 1990s, researchers began applying logistic modeling to global crude oil production, with many models showing conventional production peaking before 2010. Because of Hubbert’s accidental 1970 US “peak oil” prediction, many interpreted these global production results as a sign of an impending global energy crisis. However, the distinction between conventional and unconventional oil production was overlooked, and total global oil production continued to rise from the rapid scale-up of unconventional production – likely resulting in part from the conventional oil modeling results. Though global conventional crude oil production likely peaked before 2010, the perceived failure of logistic modeling has decreased its popularity (as evidenced by the number of academic citations of Hubbert’s 1956 paper peaking in 2013 at 128 and declining to 75 in 2016 – also producing a statistically significant fit with the logistic equation) and made the method somewhat taboo in academic research

Like the new growth cycle in US crude oil extraction from unconventional production – fit well with a multi-cycle logistic model – it is time for logistic modeling of energy production to be reestablished and valued by the academic community. Towards this end, this project used s- and bell-shaped multi-cycle logistic growth curve models to fit each source of US energy production and total US energy consumption for an analysis of the US energy landscape to 2040. The results suggest an urgent need for substantial efforts in sustainably developing new growth cycles in all sources of energy production along with deployment of atmospheric carbon management technologies. This presentation also suggests logistic modeling of a comprehensive spectrum of environmental impacts from each energy production source should be used to determine focus areas of sustainability research and innovation.

Furthermore, this project used s-shaped logistic growth curve models to fit US biofuel production for US biofuel policy analysis. The results suggest new methods for biofuel policy development and implementation are needed for more successful stimulation of sustainable biofuel production. This biofuel production model was then tied to US policy greenhouse gas emission reduction requirements and showed less than half of the required reductions are likely to be achieved by 2022 given current policy and production trends. Finally, logistic models were used to propose long-term, realistic US biofuel production goals to beyond 2050.