The Unintended Carbon Impacts of Large Scale Electricity Storage

The transition from fossil-fuel power generators to renewable energy facilities requires a significant growth of flexible resources to manage weather-dependent output variations. Key among these resources are storage assets. Although storage assets are carbon neutral in their direct operations, their arbitrage activities can influence the scheduled quantities of other power producers, affecting market carbon emissions indirectly. This raises questions about the impact levels of these indirect emissions and how to limit them effectively. We develop an electricity market optimization model and examine the emission impact of the optimal set of transactions of a storage agent. We derive a tight condition for the worst-case rate of added emissions for a storage transaction to stay below a desired threshold and characterize the minimum sufficient carbon levy to ensure this condition. These results depend on the interplay between the round-trip efficiency of the storage asset, the CO2 emission rates of conventional generators, and their costs. The findings are robust against the uncertainty in demand and renewables. An extensive numerical study using data from the Dutch electricity market is performed supporting our findings and reveals several new insights. This paper offers a comprehensive framework for market players and regulators to evaluate and interpret the unintended carbon impact of storage assets. We demonstrate that, counter-intuitively, a more efficient LSA may increase more the overall emission of the system, and require a higher carbon levy to limit them, than a less efficient one. Also, whilst a carbon levy reduces the worst-case emission rates of a storage transaction, we identify several scenarios where the overall observed emission rates may increase with higher levies.