Two From Europhysics News: ITER, Global Warming Assurances

The story I ran previously about Japan considering rescinding its bid for ITER reminds me that fusion still has a long ways to go before it can be anything close to viable as an energy source. Proof comes from the latest issue of Europhysics News, in which Jo Lister and Henri Weisen ask the not-so-theoretical question, "What Will We Learn From ITER?" (PDF). According to the duo, the lessons-to-be-learned are many, and include

• the modeling of "anomalous transport" (i.e., how individual ions move about in the plasma)
• discovery of a number of key parameters currently unknown
• how to heat the plasma to the point that the reaction is self-sustaining
• how to handle the generated heat fluxes without damaging crucial equipment
• how to remove helium ash without destroying the reaction

Overall, it's no wonder that the current ITER timeframe doesn't show a workable power plant until 2025.

The second item is called "Global Warming in a Nonlinear Climate - Can We Be Sure?" (PDF). Martin Huber makes the case that while anthropogenic warming is for real, its extent is difficult to know in the absence of large amounts of supercomputing capacity. Starting with the difficulty of making accurate local weather forecasts, he extrapolates to the much larger problem of climate, and says that the real problem associated with such issues is the wide predicted range of outcomes. The range of uncertainty is also disconcerting, as the difference between a 2K to an 11K rise in temperature is enormous. An 11K warming would have nightmarish effects on mankind: "the implications of the sealevel rise implied by the melting of the Greenland and Antarctic ice sheets are enough to focus the mind."

Huber says that we really won't have a good idea how this all will shake out until we have a petaflop machine available by hook or by crook (Pflop = 10¹⁵ floating-point operations per second). Such machines do not exist now, but are thought to be available soon.