Scope and limits

The study states plainly that these figures "do not include upper atmosphere effects of launch emissions, which are assumed to be worse than producing the same emissions on the surface of the Earth, and still under study". Since launch causes 64 to 72 per cent of the total, an unquantified multiplier on the dominant term is a material omission, and the study's own conclusion is hedged as "may have similar lifecycle GHG emissions intensities ... pending further studies". CAUTION ADDED 2026-09-04. NASA's 64-72 per cent launch attribution is contradicted by two independent lifecycle assessments using different methods and different teams: Asakura et al. (2002) put launch and propellant at 17.4 per cent of the total, and Wilson et al. (2020) put the RECTENNA at about 66 per cent of global warming potential. Both model the 1978 Reference System rather than RD1 or RD2, so this is a caution rather than a refutation - but the claim's reasoning above leans on launch being the dominant term, and that dominance is contested. Note also that both of NASA's comparators are its own supplier's data: the 13-43 gCO2eq/kWh range is NREL's LCA Harmonization work, so it cannot be checked against NREL without circularity, and NREL's own text warns its figures "may not serve as an appropriate basis for comparing technologies". SRC-012's p.viii SUMMARY TRANSPOSES COAL AND NATURAL GAS, and this record reports its figures as printed; its Table 25 does not transpose them. NASA OTPS p.viii states "the GHG emissions intensities of coal and natural gas are 486 gCO2eq./kWh and 1001 gCO2eq./kWh, respectively". Its own footnote 11 cites EIA FAQ 74, which gives coal 2.31 lb/kWh and natural gas 0.96 lb/kWh - coal more than DOUBLE gas. Converting at 453.59 g/lb: coal is about 1048 gCO2/kWh and gas about 435. So 1001 is the coal figure and 486 the gas figure, and the report has the two labels the wrong way round. Verified against the EIA source and the arithmetic re-checked 2026-09-05. The transcription here is faithful; the defect is the source's, and the fix is therefore an annotation rather than a rewrite of the quote. SRC-012 Table 25 (printed p.72, PDF 89, "Source data from NREL") gives natural gas 486 and coal 1001 gCO2eq/kWh as life-cycle medians, so the p.viii pair is Table 25's medians with the labels swapped (our reading of the two pages; footnote 11 points to EIA, but the numbers are NREL's). Both are therefore life-cycle figures, like SBSP's 26 and 40; only EIA's own figures (coal 2.31, gas 0.96 lb/kWh) are COMBUSTION CO2, and those are not like-for-like with the life-cycle figures. NASA's comparator set is five renewable technologies plus nuclear fission, and NASA files all six under "terrestrial renewable electricity production technologies" (PDF p.9); nuclear is not renewable, and NASA says it includes nuclear "even though it is not usually grouped with renewables" (PDF p.22, printed 5). NASA's Figure 2 (PDF p.11, printed x) and Table 25 (PDF p.89) give the renewables' medians as wind 13, hydropower 21, concentrating solar power 28, geothermal 37 and utility-scale solar photovoltaics 43 gCO2eq/kWh, and nuclear fission 13. So the 13 to 43 range holds for the five renewables alone, because wind ties nuclear at 13 (CLM-0849 gives the range's provenance).