Research paperImputing missing data in non-renewable empower time series from night-time lights observations

- Non-renewable empower indicates the flow of non-renewables used by a country.
- Continuous time-series cross section of empower are largely missing from literature.
- The use of non-renewables can be imputed from night-time lights imagery.
- A multiple imputation approach to model missingness in data is used.
- Non-renewable empower 1995-2012 time-series are re-constructed for 57 countries.

Emergy is an environmental accounting tool, with a specific set of indicators, that proved to be highly informative for sustainability assessment of national economies. The empower, defined as emergy per unit time, is a measure of the overall flow of resources used by a system in order to support its functioning. Continuous time-series of empower are not available for most of the world countries, due to the large amount of data needed for its calculation year by year. In this paper, we aim at filling this gap by means of a model that facilitates reconstruction of continuous time series of the non-renewable component of empower for a set of 57 countries of the world from 1995 to 2012. The reconstruction is based on a 3 year global emergy dataset and on the acknowledged relationships between the use of non-renewables, satellite observed artificial lights emitted at night, and Gross Domestic Product. Results show that this method provides accurate estimations of non-renewable empower at the country scale. The estimation model can be extended onward and backward in time and replicated for more countries, also using higher-resolution satellite imageries newly available. Besides representing an important advancement in emergy theory, this information is helpful for monitoring progresses toward Sustainable Development and energy use international goals.