Comment on Crop Yields Paper

A paper published this week on crop yields under a stratospheric aerosol regime has generated much discussion in the blogosphere. The authors compared two model scenarios, one a world with twice the historical level of CO2 in the atmosphere, the other a 2xCO2 world augmented with stratospheric aerosols, and compared them to each other and to current conditions in terms of food production. The results show that crop yields increased in the geoengineered, 2xCO2 world relative to the non-geoengineered 2xCO2 world. This evidence contradicts the common assertion that agriculture would be compromised in a geoengineered world--according to this study, the opposite would occur since the positive growth effects of CO2 fertilization would not be offset by temperature stresses associated with an unadulterated greenhouse world. Instead, SAI would minimize these stresses so that higher CO2 would lead to higher crop yields and food production.

This has generally been greeted warmly by supporters of geoengineering research. In at least one respect, however, the research design of this and similar studies subtly undercuts their position and strengthens the hand of geoengineering opponents. Specifically, by modeling only two future scenarios, a geoengineered 2xCO2 world and a non-geoengineered 2xCO2 world, and not allowing the inclusion of emissions mitigation, the authors lend credence to the false charge that abatement and geoengineering are mutually exclusive activities, and policymakers must choose either one or the other. This assertion is patently untrue and runs counter to the policy preferences of the vast majority of the geoengineering community, which is clearly committed to emissions cuts first and geoengineering only as a final, auxiliary measure. In framing geoengineering as a stand-alone policy option, these studies inadvertantly provide cover for those who would present this false choice.

One way to avoid this would be for researchers to include an additional mixed mitigation/geoengineering scenario as a standard feature of such modeling exercises. While it is important to know the independent effects of geoengineering for analytical purposes, testing the combined effects of geoengineering and emissions cuts would both disarm irresponsible critics, and tell us more about the conditions likely to obtain under the most responsible policy scenario.