My thoughts: Here’s the key statement for the finding in this report: “based on today’s Texas grid mix, which is primarily made up of fossil fuels.” If the either the marginal generation on the grid is low or no GHG (e.g., renewables overgeneration which is an increasing problem in California) or the connection to the grid is cut or restricted (e.g., in a microgrid), then this premise doesn’t hold.
This study relies on fossil fueled generation being the marginal energy source. It also focuses solely on operational changes with existing resources. The appropriate frame is looking at the change in generation investment with and without storage, so for example more renewables become cost effective with storage so the overall generation mix changes.
The second problem is that most of the production cost models are yet incapable of capturing reduction in flexible capacity use. That’s why the California Energy Commission has had DNV and LBNL working on modeling those resources. So the emission savings are underestimated.
The third problem is that savings in residual unit commitment (RUC) is underestimated in the models. These are gas units running on standby with no-load, to be available the next day for ramping, load following or reliability. Storage reduces the need for these resources as well. NREL recently released a study on the value of storage that captures this benefit.
If these findings are valid, then the existing Helms pumped storage plant is also increasing GHG emissions. One could go so far as to say that the value of pondage hydropower storage may be so diminished that relicensing conditions that require run of river operations may have little effect on costs and GHG emissions.