On Thu, Jun 20, 2013 at 01:02:54PM +0530, Srini RamaKrishnan wrote: > On Wed, Jun 19, 2013 at 4:00 PM, Eugen Leitl <[email protected]> wrote: > [...snip links...] > > Notice that most of it is very predictable, several days > > in advance. > > You are right up to the point that global climate change bites. And
There's a diagnostic in current predictions, where you vary input slightly, and if strongly divergent outcomes occur you know you're in a strongly nonlinear domain, and adjust your accuracy correctly. > even without a climate apocalypse, I thought the margin of error with > all such predictions was rather hefty - 15-20%? Whatever it is, it is more than good enough in practice. Gas turbines ramp up in 2-3 hours, coal and nuke need 2-3 days. Which means that coal and nuke have a serious handicap, and need to be phased out in favor of more agile plants. > > The problem with gas peak plants is that they run just a few > > weeks per year, and are not economic without subsidies. > > The reason coal hasn't gone the way of the dodo (and the > > nuke) is political. It does increasingly look there will > > be a premature exit from coal. > > Coal will always be needed - if not in Europe, then in China or India. > We are not going to stop digging as long as there's profit. I agree that we will burn everything. We'll probably dump as much CO2 as we did the last 1-2 centuries, and of course it will cause considerable methane outgassing, as well as nonlinearities (north polar ice could be ice-free by summer 2014 already, and water is lower albedo than ice). > >> renewable energy aplenty. This is a problem that won't be solved until > >> we can figure out how to store and normalize the energy or cheaply > > > > MWh scale battery storage is making very good advance, and EV > > battery storage does it at well. You need about an EV scale battery > > for night cycling. > > Nanobatteries are a possibility - ten - twenty years away. We definitely have an order of magnitude improvement potential. In case of nonmobile applications, the power/weight ratio is far less important than total price, and price over lifetime (how many charge/recharge cycles it can take). > > Germany's natural gas grid can currently buffer 3 months. > > We know natural gas lines can tolerate 5-15% of hydrogen without > > refitting, so hydrogen from water electrolysis and synmethane > > (via Sabatier) are likely ways to absorb surplus of renewables > > (which already happens regularly, and will become a permanent > > fixture rather soon). > > Pipelines are not close to renewable sources - the Nord Stream runs I'm talking about water electrolysis, which can be a fridge-sized unit for invididual homes, larger for municipalities. It can be located directly next to a gasholder, or an underground pressurized tank. Sabatier can be also small (they're considering running a Sabatier plant on the ISS). > subsea for example - and the transport infrastructure to integrate > renewables like you say is expensive - this is why they are just ideas No, the transport and storage for 3 months is there. You can bump it up reasonably cheaply, as it's decentral. > with marginal implementations waiting for something big to change. > > Pipelines, even subsea pipelines are vulnerable to political unrest - The whole point of decentral renewable is you produce where your consumption is. If you run the math, Netherlands or Germany would be fully energy self-reliant. There would be no need for import. Other countries would do even better. > and the world is going to be frothing with unrest for the next decade I expect that we'll see serious disruption (famines) beginning in 2020, being very serious on 2030 and the world being an extremely different place in 2050. Due to the sins of omission in the last 40 years the course for the next 40 years is already charted. We can only influence the outcomes by reengingeering the more complex systems of the world to fail gracefully, so that the resulting systems will be more resilient, and that the degradation will be not a series of catastrophes, but a more controlled degradation, eventually arrested. In order for it to happen we need the awareness to be present. Unfortunately, I do not see this awareness at the moment, so we might have chosen the pessimum trajectory. > or two. You can protect an off shore oil platform with a near shore > airbase, but a pipeline requires a network of spies and client states. > That's hard to build and maintain without cold war mentality. In any > case the investments in protecting this sort of thing are huge - and > wipe out potential savings. Speaking about a wipeout, how probable would you see a nuclear conflict arising between failing states? I see huge problems in the Pakistan/India/China corner. The climate shift will probably hit Pakistan much harder than it already is. > > It's interesting that France import electricity during winter > > from nonuclear Germany -- for electric heating -- and in the > > summer -- because during heat spells they have to shut down > > the reactors. > > Yeah, but the peak production capacity is more than France can handle > or distribute effectively, so it's not as if they are lacking the > capacity to produce significant amounts of power - nearly 30% of EU's No, they *are* lacking the capacity to cover domestic production during peak demand (during cold winter spells) and during low production (heat spells in the summer causing shutdowns). The numbers are clear enough. This is not a distribution problem, like renewables do in Germany, forcing them to export on a massive scale, overloading their own grids and their direct neighbors' grids (thus requiring local storage). > power is French. > > France lags behind in wind power and other renewable sources: 0.1% of > production I do think that the French nuclear experiment is failing, and that it will become even more obvious in the coming decade. The recent decision to reduce the nuke power fraction is rational.
