I have many issues with the new geoengineering Nature paper which I'll list below
1) it claims to be original but Kalidindi (2015) finds a similar result with a climate model link.springer.com/article/10.100…
2. it uses very choppy post-eruption data that is over 20 years old - how reliable is this data and how applicable to present/future crop distributions?
3. it uses a very simple linear statistical relationship which does not account for interactions between covariates
4. it uses data from one climate model despite begin a GeoMIP #geoengineering scenario with data from 6 models readily available - climate.envsci.rutgers.edu/GeoMIP/partici… - and with AOD being significantly different between the different models
5. it uses monthly mean data - yet solar #geoengineering may reduce frequency of droughts and heatwaves which are sub-monthly phenomena and may result in crop failure agupubs.onlinelibrary.wiley.com/doi/10.1002/20…
6. it ignores the many studies that find net NPP gains from solar geoengineering AND include interactive C3 / C4 plants which respond to changes in temperature, precipitation, CO2, and solar irradiance e.g. rmets.onlinelibrary.wiley.com/doi/abs/10.100…
7. the paper says that #geoengineering would not prevent crop failure from global warming, but in the period evaluated (2050-2069) crop yields are projected to increase due to increased CO2 (nature.com/articles/s4159…) - which is not accounted for in their simple model
8. the media response has actually managed to embellish an already exaggerated result for instance @guardian: "Reflecting sun's rays would cause crops to fail, scientists warn" - theguardian.com/environment/20…
8 (cont). No @guardian, this paper suggests that reflecting sun's rays would cause little impacts to crops - and I contend with the methodology and the result anyway! (see above)
All in all, I'm very surprised this paper made it past review, but I'd be happy to hear other opinions! @MWClimateSci @AlanRobock @DKeithClimate @KenCaldeira @peteirvine
