As I delved into the Peak Oil concept, I didnt see any outs, and I still don't see any on the consumption side (which I'll soon be writing about). However, as Ive learned more about wind energy, I think it does give us a hopeful mitigation option, if (as in the case of oil) the 'best-first' planetary areas are quickly scaled, complete with storage and distribution infrastructure. Though there will still likely be liquid fuel shortages, the greater chunk of the transportation system we can replace with electricity as fast as possible, the smaller the supply/demand gap will be.
Clearly people are making money (implying high EROIs) on wind. If the concept of oil depletion (as opposed to oil substitution) became more widely understood, perhaps we could generate momentum to use the global wind 'harvest' as a core part of societies future energy portfolio.
Wind gives me hope. But the turbines and parts are built in factories using oil and transported on vehicles using gasoline - so we need to scale quickly. Since wind has a very long economic (and energetic) life, as energy prices increase, built capacity with long life should increase in value (however you measure it) more than a shorter lived energy technology. I think if more people made decisions optimizing energy instead of dollars, the electrical infrastructure might happen on its own.
Im sure the informed reader will note that the average EROI for wind is higher than for new oil/gas exploration. What a clue. Lets get started.
What kind of hope does this give you, Nate? Wind is a localized resource, it only works well in certain areas and energy can't be transported long distances very efficiently. We also don't have good storage technology so you need some kind of backup power generation (biomass?) even in areas that have good wind potential. And there doesn't seem to be the political motivation to start building a large network of electric trains before its too late. Then there is the little problem of food production when fossil fuels decline.
So I'm not sure what kind of solution you are looking for. How large a population do you think can be supported on wind power?
He hasn't said that wind will be the only source, and I can't imagine that he would.
We've got our work cut out for us, for sure. But I see wind as a very hopeful part of that work. Solar, too, of course. I hope tidal has some breakthroughs, too. It's basically a massive form of 'pumped storage'. All these natural sources that are periodic have the rep of being 'inconsistent', but I think that their patterns will start looking a lot more reliable to us, when the curve is clearly on the downslope.
The train issue will be some serious teeth-pulling in the US, which has developed such a strong idea of 'doing it on my own', anything as collective as rail transport threatens a lot of people's sense of 'privacy', I think.
The short answer, in that Im in Logan Airport, is 'more than I had before'. The longer answer has to do with the electrification of transportation system and a larger baseload for communities from combination wind/solar
How large a population do you think can be supported on wind power?
I think the confluence of water, energy, and environmental events will one day show that we are near the peak in human population. I will make no predictions of how much smaller it will be in 20-30 years, but irrespective of the number, wind will be a larger part of the energy mix for those people than I originally envisioned.
In a sense, society has been using a one-time subsidy in the form of oil - we now need to wisely use whats left to create systems able to regularly harness a repeating subsidy of solar energy - wind will play the largest part of that. I agree with you that storage tech and backup are issues - at this stage of development if Peak Liquid Fuels is within 5 years then wind wont make much of a difference -if its 10 years out, wind could be huge. The high EROIs of wind basically mean that a hungry society has found a bounty of renewable cows, but as yet does not have milkers, milking machines, buckets or butchers.
And for the record, I have been reasonably freaked out by what I see on the horizon for several years, so please allow me some hopeful angles...:)
Anyone have more info on transmission distance, esp HVDC?
"don't have good storage technology"
You don't need storage under roughly 15% market share. OTOH, there are some very good storage methods. Alanfrombigeasy has calculated that wind could provide up to 51% of the grid. Alan, could you share the calc's?
"a large network of electric trains"
Electric vehicles are about 8 times more efficient than your average gasoline vehicle (1,600 watt-hours/mile vs 200 whrs/mile), and actually more efficient than electric trains (though electric trains have other benefits, like supporting urban living).
" the little problem of food production when fossil fuels decline"
Tractors can be electric. Fertilizer is a small % of FF use, and could come from biomass.
"How large a population do you think can be supported on wind power?"
All of it. See the first reference above. OTOH, that would be an expensive way to go. Much better would be a mix of wind, solar, hydro, biomass, wave, etc.
What technology in particular are you talking about for electric vehicles? 8x the energy consumption even in the same-aerodynamics chassis?
What are the 'very good storage methods'?
I seem to remember pumped storage being about $50/kwh in today's dollars for the Racoon Mountain system, and flow batteries costing around 3-4x that in large installations (though they aren't site-limited).
HVDC seems to be an evolutionary improvement, rather than a disruptive technology, over HVAC - around 5% loss per 1000km rather than around 8%. Land use is much lower, but the loss improvements are nothing compared to, say, HTSC lines. HVDC is naturally suited to large-capacity dynamic load balancing (as slow transformers don't need to be involved) and DC power sources like solar.
IMO, even removing a 20% loss to the farthest parts of the country won't suddenly make a particular technology viable - We CAN move power long distances efficiently with current technology. Though being able to pack 3x the conductors into the same right of way in urban areas (without ELF health nuts) might help.
"What technology in particular are you talking about for electric vehicles? 8x the energy consumption even in the same-aerodynamics chassis?"
The question I was answering was: could the grid support the replacement of all light duty (cars, SUV's, pickups) gasoline vehicles with EV's? I used efficient EV's (Tesla) and HEV's (Prius) and compared them to the current fleet average. The comparison helps answer the intuitive question: "isn't that a lot of energy for the grid to supply?" The answer is that it's not really as much energy as you might expect. OTOH, if you compared within the same class of aerodynamics chassis the ratio might be 4-6:1.
The Tesla uses 215 wh/mile, outlet to wheel, and it's optimized for speed, not efficiency.
"What are the 'very good storage methods'?"
I'm mostly thinking of the same things: off-setting hydro, pumped storage, flow batteries, EV planned charging and Vehicle to Grid. "very good" might have been a little strong - "good enough" is probably better, though Alan feels very strongly about the effectiveness of hydro & pumped storage, and I think PHEV & EV's will be very, very useful.
If I understand you, you feel that if wind is otherwise viable that transmission won't be a barrier to it's use. Is that right?
I feel that in the many orders of magnitude of technology improvement necessary to shift to a sustainable energy future, a 25% energy loss to transmit electricity from as far as Seattle to Boston is a pittance. The many years of wind capacity growth of ~25% only needs an extra 1 year if you were producing it all in Seattle and bringing it to Boston. Which you're not.
That sending your solar produced in Texas to Los Angeles probably has less of of an energy footprint than storing it in pumped storage in Texas for later use in Dallas is helpful.
A good portion of industrial use can be tempered to low-usage times. Smelters don't have to operate at 4:30pm when everyone at once turns on their AC. That and EV planned seem like they'd have a lot bigger effect than vehicle to grid, which is hopelessly decentralized + inconveniant IMO.
Yeah, V2G would be pretty complicated to implement in a largescale way. Using it for household backup might be easier.
OTOH, these days cars are pretty much computers that happen to have wheels, and communication & control through intelligent meters might not be difficult to do in the long run. Things will change a great deal in the next couple of decades, I think.
I feel that in the many orders of magnitude of technology improvement necessary to shift to a sustainable energy future...
I think you've overestimated the problem here. It appears that less than 1 order of magnitude in conversion from biomass to energy will do to replace all petroleum motor fuels. There are energy-positive structures being built; with continued improvement in their cost structure (a large part of which will be economies of scale) and increasing price of fossil fuels, and they'll be cheaper than conventional structures too.
Smelters don't have to operate at 4:30pm when everyone at once turns on their AC.
Actually, many industrial processes require continuous control. Thermal cycling of the insulation in a smelter is bad; blast furnaces are often rebuilt after each shutdown.
That and EV planned seem like they'd have a lot bigger effect than vehicle to grid, which is hopelessly decentralized
Many commentators consider decentralization a virtue.
Many commentators are going to be pissed when they find their new EVs half charged because it wasn't very windy today(though PHEVs have a bit of an advantage here). Vehicle to grid requires perfectly sinchronized 60hz invertors at every house with near zero drift. It requires intelligent load balancing across a network of vehicles so prone to break down that you have a repair shop within a few miles of your house. I'm all for solar decentralization, perhaps inverted at the neighorhood level. But load balancing based on a vehicle that's driven off the grid, needs to be reliably kept at a high charge percentage, and which relies on battery tech with limited charge/recharge cycles, doesn't seem like a good way to use resources. Even solar-to-grid is rather difficult - preventing islanding and keeping in phase with good power factor and such are hard. It's simply much easier to drive the grid waveform from a single or few highly managed sources. Using car batteries for distributed storage requires very smart management that isn't possible with our current grid.
Shifting charging demand over to certain times is much, much easier. It's trivial and self-regulating to setup a wifi or wimax network and send out an expected power price over time chart, then have a locally smart charger fill that up with the cheapest juice.
-----------------------
I wasn't really talking about short-term demand, though I guess I'm a bit out of my league here. Would changing the standard electricity-intensive heavy industry worker over to a night shift be possible as a means of deflecting demand from peak periods?
-------------------------
Sorry I was rather vague in the first statement - I should have stuck to wind capacity. A ~10% or so average hit in transmission costs from the nearest rural area populated with wind turbines is nothing, compared to the many thousands of percent that would be needed to shift most of our energy production over to wind turbines.
It means that yes, if we wanted to we could built an underwater nuclear complex in the center of the south pacific thousands of miles from the nearest human, and shift over all the energy to our homes at a cost of increasing the complex size by a paltry 50% or so over what we could build right here.
he many thousands of percent that would be needed to shift most of our energy production over to wind turbines.
Your cost estimates are WAY off. With no economic value attached to GW today, the zero GHG grid that I proposed would likely raise rates 50% to 75% (which would happen anyway).
A steady rise in carbon taxes would push us towards that sort of grid anyway (nukes need pumped storage as well to neet anything more than peak load). Depending on costs associated with nukes (remember costs of the last dozen finished in the US, and the new Finnish one seems in trouble early) and just how steep the decline is in WTs and other renewable costs (WT electricity will be cheaper in 2012 than today, Not so for nuke) the mix will be somewhere between 23% nuke and ~/2.3rds nuke on strictly economics alone.
More than 2/3rds nuke begins to run into significant problems. France is able to get up to 90% nuke becasue they sell power all night long to ALL of their neighbors. Swiss utilites buy night power from several French nukes and save their water for selling back to the French, Germans, Italians at peak (at 3 to 5 times the price). Perhaps we can do the same with Canada.
Also nuke is VERY risky to build a society on because of common design flaws. Any design can have a hidden flaw, which, when discovered, requires shutting down ALL reactors of that type for months tp years. It has happened several times already and will happen again. No one reactor type should IMHO supply more than 4% of national power. Unexpectedly losing 4% of your generation is a blow, but it can be worked around with luck. More than 4% ? Nope.
EP, I'm surprised at your emphasis here. A LOT of industrial power is shifted to the night to take advantage of lower rates. Heck, I have a steel mill a mile from my home that shifts into overdrive at night...
Look I hate to be a party pooper because I can see we have some real wind enthusiasts here but GET REAL people. How many years and how many dollars would it cost to construct all these wind turbines we would need to even make a dent in our power consumption. And then we have what 200 million gasoline and diesel powered cars and trucks here in the US and you want to convert them all over to electric? How many years and how much money would that take?
You are talking about a major societal transformation here for wind power to make any impact on mitigating the fossil fuel crisis. And peak oil is within five years?
" How many years and how many dollars would it cost to construct all these wind turbines we would need to even make a dent in our power consumption."
We're there now. Wind is supplying 43% of planned new generation in 2007 in the US. It can easily ramp up to supply all new demand growth (2% per year) within 5 years. Wind can handle demand growth replacement of existing plants that are planned for replacement, and substitution for depleting nat gas if we made a modest societal commitment to using it to the exclusion of coal. Actually replacing existing power plants before their planned end-of-life, and replacing existing coal usage are more difficult questions: those would be expensive, and require a major societal commitment that we're not yet close to.
"How many years and how much money would that take? "
Keep in mind that we don't have to replace all 210M vehicles: newer vehicles get much higher useage (something Hirsch didn't take into account), and there are only 100M households. Probably 5 years US vehicle production (85M vehicles) could replace 60% of miles driven.
There are two different questions: is there enough power for the grid, and is there enough portable power for transportation. I think unquestionably the grid will be ok with only relatively modest investments in infrastructure. Transportation? That could be painful. There will certainly be enough for key needs such as transporting wind turbines, but visiting mom in Florida, or commuting to distant low wage jobs, may get expensive.
Don't forget that we are talking about exponential growth for renewable energy. Today it is in the <1% range (excluding hydro), a decade from now it will be several %, two decades from now it will be tens of percent and four decades from now it will be close to 100%. You can check adoption curves for other disruptive technologies (bronze, iron, steel, railways, cars, computers) and you will inevitably find the same laws at work. Just because it takes decades to get something started does not mean it will take centuries for it to take the lead. Quite the contrary. Soon we will have to worry about keeping environmental effects of renewables under control. See the issues with wind energy.
As indirectly noted by InfinitePossibilities above, wind is growing in the US very fast. It's roughly doubling every two years, and as I have noted elsewhere, it accounts for 43% of planned generation in the US for 2007 (after adjustment for capacity factor) - see page 8 http://www.nei.org/documents/Energy%20Markets%20Report.pdf
To me that says wind has "arrived". What do you think?
There is no "law" that guarantees wind's acension. Bronze, iron, steel, railways, cars, computer, etc. replaced the status quo technologies because they were superior in multiple ways. Wind has some advantages (good EROI in good locations, low CO2/MW) and some disadvantages (intermittent, far from load) centers.
Wind at 100% of power generation in 40 years? No way. Remember, in the US wind competes against baseload coal, gas, hydro and nuclear. System operators look at the *relative* value and cost when they dispatch power. Coal is abundant and cheap. Gas is less abundant but also pretty cheap. Wind is also cheap in terms of operating cost, but it is a not under operator control and is variable based on weather conditions. From the system opeators this reduces the value of wind energy, it reduces its contribution to reserve margins that are dictated by regulations, and it reduces the
value to wind plant owners in surplus generation that occurs when wind power saturates the flexible dispatch portion of grid operations.
These are not insurmountable prolems, but they are formidable barriers.
The EIA wind experts told me that if the 1.9 cent production tax credit goes away, wind energy goes into a deep stasis in the US. What does that say about its viability?
It strikes me that to calculate an 'honest' EROI for wind, one must take into account the energy necessary to build and maintain a storage system. Obviously this would vary depending on a lot of factors such as the base load flexibility in a given grid. But just to count the raw power output of a turbine with the energy needed to build the turbine doesn't yield a very useful figure in terms of substituting one form of energy for another.
I suppose it really would make more sense to figure the EROI of a 'black box' which delivers the amounts of energy we want when it is needed. Within the 'black box' would be a mix of wind, solar, nuclear, etc. It may be useful for starters to have the figures for individual sources, but real world applications need more complex analyses.
Cutler, you have been in this business for a while. Do you know of anyone who has done, even on an abstract level, this type of 'composite' EROI analysis? I suspect it might surprise us that the mix could be far better than the individual sources. Whole greater than the sum and all that.
Good point about the storage--I have not seen an EROI for wind that accounts for this. However, the point may be moot. In the US at least, no one is even considering building storage systems for wind--way to expensive (an hence lower EROI). Wind power will be dumped on to the grid--hence the reliability issues I mentioned.
Based on the costs to build a wind farm from Pacca and Horvaths (summary of article below, sorry it doesn't format properly), consider a windmill composed of steel and concrete. A windmill farm in the Escalante desert, built to produce 5.55 TWh of power, would require 13.8 million pounds of aluminum, 2.8 trillion pounds of concrete, 639 billion pounds of steel, etc. The wind farm would occupy over 189 square miles. Pacca & Horvath don't give the capacity factor for these windmills, but an often used number is 30% (i.e. wind blows hard enough 30% of the time), so a 5.55 TWh wind farm might serve around 175,000 to 350,000 people, depending on the wind speed and how close people were to the windmills, since power is lost via transmission over long distances.
In 1992 such a wind farm would cost 200 million dollars, which doesn't include labor and maintenance costs, and would serve less than one percent of the United States population. It would cost over $200,000,000,000 to build enough windmills to generate electrical power for everyone (though of course, you couldn't, since not all areas have enough wind). With energy prices many times higher now than in 1992, the cost would be far more expensive.
---------------------
Summary of Sergio Pacca and Darpa Horvath 2002 Greenhouse Gas Emissions from Building and Operating Electric Power Plants in the Upper Colorado River Basin
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 14 pp. 3194-3200
There is a large area of research devoted to figuring out how much material, energy, and cost is required to build various types of power plants. To estimate the overall greenhouse gas (GHG) emissions over the life cycle of a plant, Pacca and Horvath used Life Cycle Assessment (LCA), a method that calculates materials extraction, manufacturing and production, operations, and the disposal of the materials at the end of the life of the power plant.
As you can imagine, this isn't easy. There are two main LCA models -- Pacca and Horvath chose the EIOLCA approach, which uses a large commodity matrix that tries to identify the entire chain of suppliers of the raw materials, and then this matrix is multiplied by another one containing emissions and energy use per dollar.
Because dollars fluctuate in value, a better method would be to calculate the energy used at every step of the chain, but still, these dollar amounts give a rough idea of the embedded energy.
The study compares the Glen Canyon dam with four other types of power plants, all figures are scaled to each plant producing 5.55 TWh of energy per year.
This kind of study could help decide which direction a future energy Manhattan project should. This study rules out a Photovoltaic power plant, which is not possible now -- it requires 4118 MW of power, but the total world production of PV modules up to 1997 was only 125 MW, less than 3% of what's required for just this one plant. The PV plant also displaces an enormous ecosystem, about 20 square miles.
This study does not cover nuclear power plants. Another study states "nuclear fission energy requires small inputs of natural resources compared to most other fossil and non-fossil energy technologies. When we consider net electricity generation (e.g., net electricity after subtracting consumption by internal plant loads and by uranium enrichment plants), the life-cycle resource inputs for non-fossil power sources are dominated by construction materials, most notably steel and concrete. The construction of existing 1970-vintage U.S. nuclear power plants required 40 metric tons (MT) of steel and 190 cubic meters (m3) of concrete per average megawatt of electricity (MW(e)) generating capacity. For comparison, a typical wind energy system operating with 6.5 meters-per-second average wind speed requires construction inputs of 460 MT of steel and 870 m3 of concrete per average MW(e). Coal uses 98 MT of steel and 160 m3 of concrete per average MW(e); & natural-gas combined cycle plants use 3.3 MT steel and 27 m3 concrete" (1).
Below are two tables summarizing the data.
GWE: Global Warming Effect is the Greenhouse Gas (GHG) emissions in MegaTons of CO2 equivalent, which is calculated by adding CO2 + CH4 +N2O together
MT = MegaTon = 1,000,000 Metric tons. 1 MT = 2,204.62262 pounds
Here's just wind since it didn't format properly below:
Wind
Construction Farm
Input Total MT
--------------- ---------
aluminum.............6,275
cement
concrete.........1,266,172
copper.............. 1,569
electricity MWh..1,691,678
excavation m3
glass................4,930
oil....................448
plastics............20,169
sand.................9,412
steel..............289,987
(1) Peterson, P. F. Will the United States Need a Second Geologic Repository? The Bridge 2003, 33 (3), 26-32.
TABLE 1: COMPARISON OF INPUTS
Hydro PV Wind Coal Nat Gas
Construction Plant Plant Farm Plant Plant
Input Total MT Total MT Total MT Total MT TOTAL MT
--------------- --------- --------- -------- -------- --------
aluminum 67 177,788 6,275 624 230
cement 2,222,356
concrete 9.906.809 1,266,172 178,320 71,270
copper 90 480,029 1,569
electricity MWh 7,556,010 1,691,678
excavation m3 4,711,405
glass 1,066,731 4,930
oil 448
plastics 20,169
sand 9,412
steel 32,183 4,600,276 289,987 62,200 51,130
Operational Inputs
------------------
coal combustion 2,336,000
coal extraction 2,336,000
transportation by railroad 2,336,000
natural gas combustion 1,560,300,000 m3
natural gas transportation 1,560,300,000 m3
natural gas extraction 1,560,300,000 m3
TABLE 2: COST, GWE (Global Warming Effect), and Area required
Total Cost Area
(1992 $) GWE required
----------- ------- --------------
Coal Power Plant 149,772,446 90,000,000 n/a
Wind Farm 206,881,416 800,000 489,580,000 m2
Natural Gas Plant 374,033,481 50,000,000 n/a
Hydroelectric Dam 503,240,216 500,000 651,141,400 m2
Photovoltaic Plant 3,578,457,990 10,000,000 51,386,400 m2
NOTE: the cost in 1992 dollars doesn't include labor, installation, or maintenance costs.
Photovoltaic Plant 100-W panels of dimensions 1.316 x 0.66 m with array units of 3 x 10 panels, each having its own concrete foundation, for a surface area of 3.9 x 6.6 m, sited at 30° latitude, at a 30-deg tilt (approximately 1.2 m of additional width is needed to account for shading by the array due to the sun's angle). There is 0.9 m between each of these array units for personnel access. Each adjacent unit covers a land area of 37.44 m2 and has a capacity rating of 3 kW. Some 1,372,500 of these 3 kW units are required.
Wind Farm location: Southern Utah, at 7,000 feet. average windspeed 6.5 m/s turbine: 600 kW in 4480 turbines
Hydropower: As the U.S. Bureau of Reclamation has suggested, "upgrading hydroelectric generator and turbine units at existing power plants is one of the most immediate, cost-effective, and environmentally acceptable means for developing additional electrical power".
I had trouble with the units in what you posted, but 2.8 trillion pounds of contcrete is
2.8X10EE12 pounds of concrete
or 1.27 X 10EE12 kgs of concrete
or 1.27 X 10EE9 tonnes of concrete
or 283,482 tonnes of concrete per turbine?
That is basically as much concrete, per turbine, as you would use to build a substantial skyscraper.
That number looks really wrong. Similar for all the other raw materials numbers quoted.
Just on Load Factor, for any power plant it means the per cent. of the rated capacity you will achieve.
So for a wind turbine, 30% means 30% of the time it will blow at 100% of rated capacity, or 100% of the time it will produce at 30% of rated capacity.
Nukes typically run in the low 80s (distorted a bit by the fact that every few years they have a complete maintenance shutdown). All other power stations run below that level (because nuclear and hydro produce most of the baseload).
My own calculations from the problem set above
5.500 TWhr = 0.1% ish of US power consumption
5500 GWhr requires 6278MW of capacity at 100% Load Factor (divide by 8760 hrs pa)
So therefore at 0.3 LF 20,926 1 MW turbines (actually 1.2-1.4MW/ turbine is more like it).
Cost would be about $20bn.
Cost to do that in nuclear would be about $16bn (assuming 3rd Gen technology ie 4X1650MW units at $4bn each) + whatever price you care to put on waste disposal and long term decommissioning. Gas or coal would be less than $10bn but you would then have fuel cost.
So if you did that 800 times you would cover the entire US power consumption. For $1.6 trillion. Which is about 15% of US GDP now, or about equal to what the US spends on fixed commercial assets every year (capital spending by companies).
So over 20 years, 5% of US capital spending to cover the entire US energy consumption.
The estimated total cost of the war in Iraq is between 1 and $2 trillion (that was actually a 2004 estimate, so I am assuming the current costs of $15bn a month or so are offset by no rise in future costs).
Now there are a few other factors: depreciation (but that affects the turbines much more than the structures), growth in power demand (however GDP would also grow), the fact that you wouldn't use wind for all that capacity (because of grid issues).
If concrete weighs about 100 pounds/cubic foot * 27
= 2700 lbs/cubic yard
200 cubic yards times 2700 pounds = 540,000 lbs
540,000 / 2,000 = 270 tons per windmill foundation
270 tons times 4480 windmills = 1,209,600 tons for this windmill farm, which is within 5% of what Pacca and Horvath use (1,266,172)
Here is what their paper had to say about windmills:
A wind farm producing 5.55 TWh of electricity per year was assumed to be in southern Utah, at an elevation of 2134 m (7000 ft), close to the Escalante Desert where the average wind speed is 6.5 m/s (35). A turbine of 600 kW (36) was used as the unit for the farm's total of 4480 turbines that would occupy an area of 489 580 000 m2 (37). The total cost of materials and construction of the facility would amount to $206,881,000 (in 1992 dollars) without labor/installation and maintenance costs. Given a range of prices between $250 and $1200 per ha, the required land would add an additional $12,000,000-59,000,000 to the cost. Given the large area, land between the turbines could be used for other activities such as agriculture. No NEP loss was anticipated. The contribution of construction materials and energy to theGWE of the wind farm after 20 yr of operation (800 000 MT of CO2 equiv) is shown in Table 3.
It was assumed that after 20 yr of operation all turbines had to be replaced (but not the concrete foundations) and that the required construction energy was 30% of the original electricity and 100% of petroleum used. The electricity output of the facility remained constant. The refurbishment resulted in 900,000 MT of CO2 emissions, two-thirds of the original emissions from manufacturing and constructing the plant (1,300,000 MT of CO2).
I thought about analyzing this, and realized it was pointless. The costs of wind farms have fallen so much since 1992 that any data from that time is really, really out of date and unrealistic.
Really? The cost of the Bay Bridge has tripled from 2 billion to 6 billion dollars because of the high cost of concrete and steel. Which costs exactly have come down?
In the last 14 years there has been a revolution in wind turbine design. Massively larger (often following the cube/square law for greater efficiency i.e. increase physical dimensions, square materials required, cube output) and better designs in all areas, blades, generators, gearboxes (Vestas has some problems there recently) and even towers are better.
Any 1992 wind turbine data is of historical interest only. Simply not relevant to today or, even more, tomorrow.
I hardly know where to start. First, wind turbine size has risen sharply (power output is the square of size, and cost is linear, so cost drops proportionately to size). As noted by Valuethinker, nobody uses 600KW turbines now - they range from 1 to 3 MW. Also, manufacturing cost has dropped dramatically in the last 14 years largely due to operational experience and improved methods, despite the jump in material costs in the last 2 years.
Second, either something is seriously wrong with this study, or wind was already a lot cheaper in 1992 than any other source of electricity: 5.55 TWhr per year, at $.10/kwhr, is worth $555 million. If this windfarm costs $200 million, then that's a 4 month payback and a return on investment of about 300% per year. In other words, this windfarm would have generated electricity at a cost of about a half penny per kwhr.
Don't be confused by the difference between my answer and Alan's. I was talking about the relationship between blade length and swept area, and he was talking about increased wind speed (due to higher turbines) and wind power.
What that tells you is that we allow CO2 emissions for free!
It's called a negative externality in economics.
Essentially we allow polluters to pollute without restriction, the most dangerous industrial pollutant of all-- the one that could trigger the end of human life on this planet (or, more likely, make our current civilisation unsustainable).
If you charge $100/tonne for Carbon emissions ($28 per tonne of CO2) the economics of coal look very different. European permit prices under the emissions trading scheme have reached those kinds of levels.
There is actually no economic case for allowing coal fired power, without carbon sequestration, given the potential damage of those CO2 emissions.
The fossil fuel industry, world wide, is a major recipient of government subsidies, implicit or explicit. From the destruction of natural habitats for which there are tax allowances for any restorative work (or the work is just not done) through to the high human cost of an industry with a very high mortality rate. (I won't mention lung cancer from particulates emission).
The nuclear industry is itself the recipient of massive government subsidies. The Price Anderson Act provides insurance which would not be available in private markets, limiting the liability in the case of an accident. The R&D was paid for by governments. The future waste storage liability is undertaken by governments when we have a solution.
No nuclear utility operates in a pure 'merchant power' context. British Energy tried, selling into the pool, and when the pool price crashed, it went broke and the government had to stump up £3.5bn to refinance it (to prevent a renationalisation).
Nuclear utilities across the world are either state controlled or have arrangements with the regulators that allow their cost of production to be loaded onto the consumer (effectively a guaranteed floor price).
The Bush Energy Act and the proposed British nuclear restart both provide for explicit price subsidies for new nuclear facilities.
The UK decommissioning liability for existing nuclear plants is £70bn present value.
Clearly people are making money (implying high EROIs) on wind. If the concept of oil depletion (as opposed to oil substitution) became more widely understood, perhaps we could generate momentum to use the global wind 'harvest' as a core part of societies future energy portfolio.
Wind gives me hope. But the turbines and parts are built in factories using oil and transported on vehicles using gasoline - so we need to scale quickly. Since wind has a very long economic (and energetic) life, as energy prices increase, built capacity with long life should increase in value (however you measure it) more than a shorter lived energy technology. I think if more people made decisions optimizing energy instead of dollars, the electrical infrastructure might happen on its own.
Im sure the informed reader will note that the average EROI for wind is higher than for new oil/gas exploration. What a clue. Lets get started.
So I'm not sure what kind of solution you are looking for. How large a population do you think can be supported on wind power?
We've got our work cut out for us, for sure. But I see wind as a very hopeful part of that work. Solar, too, of course. I hope tidal has some breakthroughs, too. It's basically a massive form of 'pumped storage'. All these natural sources that are periodic have the rep of being 'inconsistent', but I think that their patterns will start looking a lot more reliable to us, when the curve is clearly on the downslope.
The train issue will be some serious teeth-pulling in the US, which has developed such a strong idea of 'doing it on my own', anything as collective as rail transport threatens a lot of people's sense of 'privacy', I think.
Bob
The short answer, in that Im in Logan Airport, is 'more than I had before'. The longer answer has to do with the electrification of transportation system and a larger baseload for communities from combination wind/solar
I think the confluence of water, energy, and environmental events will one day show that we are near the peak in human population. I will make no predictions of how much smaller it will be in 20-30 years, but irrespective of the number, wind will be a larger part of the energy mix for those people than I originally envisioned.
In a sense, society has been using a one-time subsidy in the form of oil - we now need to wisely use whats left to create systems able to regularly harness a repeating subsidy of solar energy - wind will play the largest part of that. I agree with you that storage tech and backup are issues - at this stage of development if Peak Liquid Fuels is within 5 years then wind wont make much of a difference -if its 10 years out, wind could be huge. The high EROIs of wind basically mean that a hungry society has found a bounty of renewable cows, but as yet does not have milkers, milking machines, buckets or butchers.
And for the record, I have been reasonably freaked out by what I see on the horizon for several years, so please allow me some hopeful angles...:)
It's available in most places, and in the US it's available in all parts of the US. See http://news-service.stanford.edu/news/2005/may25/wind-052505.html
"energy can't be transported long distances very efficiently"
It can be transported pretty far: a quick search found references to 700 mile long transmission lines to California, and 800 km long lines in this discussion: http://www.worldbank.org/html/fpd/em/transmission/technology_abb.pdf
Anyone have more info on transmission distance, esp HVDC?
"don't have good storage technology"
You don't need storage under roughly 15% market share. OTOH, there are some very good storage methods. Alanfrombigeasy has calculated that wind could provide up to 51% of the grid. Alan, could you share the calc's?
"a large network of electric trains"
Electric vehicles are about 8 times more efficient than your average gasoline vehicle (1,600 watt-hours/mile vs 200 whrs/mile), and actually more efficient than electric trains (though electric trains have other benefits, like supporting urban living).
" the little problem of food production when fossil fuels decline"
Tractors can be electric. Fertilizer is a small % of FF use, and could come from biomass.
"How large a population do you think can be supported on wind power?"
All of it. See the first reference above. OTOH, that would be an expensive way to go. Much better would be a mix of wind, solar, hydro, biomass, wave, etc.
What are the 'very good storage methods'?
I seem to remember pumped storage being about $50/kwh in today's dollars for the Racoon Mountain system, and flow batteries costing around 3-4x that in large installations (though they aren't site-limited).
HVDC seems to be an evolutionary improvement, rather than a disruptive technology, over HVAC - around 5% loss per 1000km rather than around 8%. Land use is much lower, but the loss improvements are nothing compared to, say, HTSC lines. HVDC is naturally suited to large-capacity dynamic load balancing (as slow transformers don't need to be involved) and DC power sources like solar.
IMO, even removing a 20% loss to the farthest parts of the country won't suddenly make a particular technology viable - We CAN move power long distances efficiently with current technology. Though being able to pack 3x the conductors into the same right of way in urban areas (without ELF health nuts) might help.
The question I was answering was: could the grid support the replacement of all light duty (cars, SUV's, pickups) gasoline vehicles with EV's? I used efficient EV's (Tesla) and HEV's (Prius) and compared them to the current fleet average. The comparison helps answer the intuitive question: "isn't that a lot of energy for the grid to supply?" The answer is that it's not really as much energy as you might expect. OTOH, if you compared within the same class of aerodynamics chassis the ratio might be 4-6:1.
The Tesla uses 215 wh/mile, outlet to wheel, and it's optimized for speed, not efficiency.
"What are the 'very good storage methods'?"
I'm mostly thinking of the same things: off-setting hydro, pumped storage, flow batteries, EV planned charging and Vehicle to Grid. "very good" might have been a little strong - "good enough" is probably better, though Alan feels very strongly about the effectiveness of hydro & pumped storage, and I think PHEV & EV's will be very, very useful.
If I understand you, you feel that if wind is otherwise viable that transmission won't be a barrier to it's use. Is that right?
That sending your solar produced in Texas to Los Angeles probably has less of of an energy footprint than storing it in pumped storage in Texas for later use in Dallas is helpful.
A good portion of industrial use can be tempered to low-usage times. Smelters don't have to operate at 4:30pm when everyone at once turns on their AC. That and EV planned seem like they'd have a lot bigger effect than vehicle to grid, which is hopelessly decentralized + inconveniant IMO.
OTOH, these days cars are pretty much computers that happen to have wheels, and communication & control through intelligent meters might not be difficult to do in the long run. Things will change a great deal in the next couple of decades, I think.
I meant household demand management & time shifting. Though I have seen somebody use a Prius as a household UPS....
Actually, many industrial processes require continuous control. Thermal cycling of the insulation in a smelter is bad; blast furnaces are often rebuilt after each shutdown.
Many commentators consider decentralization a virtue.
Shifting charging demand over to certain times is much, much easier. It's trivial and self-regulating to setup a wifi or wimax network and send out an expected power price over time chart, then have a locally smart charger fill that up with the cheapest juice.
-----------------------
I wasn't really talking about short-term demand, though I guess I'm a bit out of my league here. Would changing the standard electricity-intensive heavy industry worker over to a night shift be possible as a means of deflecting demand from peak periods?
-------------------------
Sorry I was rather vague in the first statement - I should have stuck to wind capacity. A ~10% or so average hit in transmission costs from the nearest rural area populated with wind turbines is nothing, compared to the many thousands of percent that would be needed to shift most of our energy production over to wind turbines.
It means that yes, if we wanted to we could built an underwater nuclear complex in the center of the south pacific thousands of miles from the nearest human, and shift over all the energy to our homes at a cost of increasing the complex size by a paltry 50% or so over what we could build right here.
hi
why is the synchronization at 60 Hz with system wide phase coherence difficult ?
i'm worked on R&D teams where phase locked loops were up in the 200 MHz ballpark.
there's got to be a way to synchronize phase at 60 Hz.
wwswimming
at
yahoo.com
Your cost estimates are WAY off. With no economic value attached to GW today, the zero GHG grid that I proposed would likely raise rates 50% to 75% (which would happen anyway).
A steady rise in carbon taxes would push us towards that sort of grid anyway (nukes need pumped storage as well to neet anything more than peak load). Depending on costs associated with nukes (remember costs of the last dozen finished in the US, and the new Finnish one seems in trouble early) and just how steep the decline is in WTs and other renewable costs (WT electricity will be cheaper in 2012 than today, Not so for nuke) the mix will be somewhere between 23% nuke and ~/2.3rds nuke on strictly economics alone.
More than 2/3rds nuke begins to run into significant problems. France is able to get up to 90% nuke becasue they sell power all night long to ALL of their neighbors. Swiss utilites buy night power from several French nukes and save their water for selling back to the French, Germans, Italians at peak (at 3 to 5 times the price). Perhaps we can do the same with Canada.
Also nuke is VERY risky to build a society on because of common design flaws. Any design can have a hidden flaw, which, when discovered, requires shutting down ALL reactors of that type for months tp years. It has happened several times already and will happen again. No one reactor type should IMHO supply more than 4% of national power. Unexpectedly losing 4% of your generation is a blow, but it can be worked around with luck. More than 4% ? Nope.
EP, I'm surprised at your emphasis here. A LOT of industrial power is shifted to the night to take advantage of lower rates. Heck, I have a steel mill a mile from my home that shifts into overdrive at night...
You are talking about a major societal transformation here for wind power to make any impact on mitigating the fossil fuel crisis. And peak oil is within five years?
It's too late for this.
We're there now. Wind is supplying 43% of planned new generation in 2007 in the US. It can easily ramp up to supply all new demand growth (2% per year) within 5 years. Wind can handle demand growth replacement of existing plants that are planned for replacement, and substitution for depleting nat gas if we made a modest societal commitment to using it to the exclusion of coal. Actually replacing existing power plants before their planned end-of-life, and replacing existing coal usage are more difficult questions: those would be expensive, and require a major societal commitment that we're not yet close to.
"How many years and how much money would that take? "
Keep in mind that we don't have to replace all 210M vehicles: newer vehicles get much higher useage (something Hirsch didn't take into account), and there are only 100M households. Probably 5 years US vehicle production (85M vehicles) could replace 60% of miles driven.
There are two different questions: is there enough power for the grid, and is there enough portable power for transportation. I think unquestionably the grid will be ok with only relatively modest investments in infrastructure. Transportation? That could be painful. There will certainly be enough for key needs such as transporting wind turbines, but visiting mom in Florida, or commuting to distant low wage jobs, may get expensive.
To me that says wind has "arrived". What do you think?
Wind at 100% of power generation in 40 years? No way. Remember, in the US wind competes against baseload coal, gas, hydro and nuclear. System operators look at the *relative* value and cost when they dispatch power. Coal is abundant and cheap. Gas is less abundant but also pretty cheap. Wind is also cheap in terms of operating cost, but it is a not under operator control and is variable based on weather conditions. From the system opeators this reduces the value of wind energy, it reduces its contribution to reserve margins that are dictated by regulations, and it reduces the
value to wind plant owners in surplus generation that occurs when wind power saturates the flexible dispatch portion of grid operations.
These are not insurmountable prolems, but they are formidable barriers.
The EIA wind experts told me that if the 1.9 cent production tax credit goes away, wind energy goes into a deep stasis in the US. What does that say about its viability?
I suppose it really would make more sense to figure the EROI of a 'black box' which delivers the amounts of energy we want when it is needed. Within the 'black box' would be a mix of wind, solar, nuclear, etc. It may be useful for starters to have the figures for individual sources, but real world applications need more complex analyses.
Cutler, you have been in this business for a while. Do you know of anyone who has done, even on an abstract level, this type of 'composite' EROI analysis? I suspect it might surprise us that the mix could be far better than the individual sources. Whole greater than the sum and all that.
In 1992 such a wind farm would cost 200 million dollars, which doesn't include labor and maintenance costs, and would serve less than one percent of the United States population. It would cost over $200,000,000,000 to build enough windmills to generate electrical power for everyone (though of course, you couldn't, since not all areas have enough wind). With energy prices many times higher now than in 1992, the cost would be far more expensive.
---------------------
Summary of Sergio Pacca and Darpa Horvath 2002 Greenhouse Gas Emissions from Building and Operating Electric Power Plants in the Upper Colorado River Basin
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 14 pp. 3194-3200
There is a large area of research devoted to figuring out how much material, energy, and cost is required to build various types of power plants. To estimate the overall greenhouse gas (GHG) emissions over the life cycle of a plant, Pacca and Horvath used Life Cycle Assessment (LCA), a method that calculates materials extraction, manufacturing and production, operations, and the disposal of the materials at the end of the life of the power plant.
As you can imagine, this isn't easy. There are two main LCA models -- Pacca and Horvath chose the EIOLCA approach, which uses a large commodity matrix that tries to identify the entire chain of suppliers of the raw materials, and then this matrix is multiplied by another one containing emissions and energy use per dollar.
Because dollars fluctuate in value, a better method would be to calculate the energy used at every step of the chain, but still, these dollar amounts give a rough idea of the embedded energy.
The study compares the Glen Canyon dam with four other types of power plants, all figures are scaled to each plant producing 5.55 TWh of energy per year.
This kind of study could help decide which direction a future energy Manhattan project should. This study rules out a Photovoltaic power plant, which is not possible now -- it requires 4118 MW of power, but the total world production of PV modules up to 1997 was only 125 MW, less than 3% of what's required for just this one plant. The PV plant also displaces an enormous ecosystem, about 20 square miles.
This study does not cover nuclear power plants. Another study states "nuclear fission energy requires small inputs of natural resources compared to most other fossil and non-fossil energy technologies. When we consider net electricity generation (e.g., net electricity after subtracting consumption by internal plant loads and by uranium enrichment plants), the life-cycle resource inputs for non-fossil power sources are dominated by construction materials, most notably steel and concrete. The construction of existing 1970-vintage U.S. nuclear power plants required 40 metric tons (MT) of steel and 190 cubic meters (m3) of concrete per average megawatt of electricity (MW(e)) generating capacity. For comparison, a typical wind energy system operating with 6.5 meters-per-second average wind speed requires construction inputs of 460 MT of steel and 870 m3 of concrete per average MW(e). Coal uses 98 MT of steel and 160 m3 of concrete per average MW(e); & natural-gas combined cycle plants use 3.3 MT steel and 27 m3 concrete" (1).
Below are two tables summarizing the data.
GWE: Global Warming Effect is the Greenhouse Gas (GHG) emissions in MegaTons of CO2 equivalent, which is calculated by adding CO2 + CH4 +N2O together
MT = MegaTon = 1,000,000 Metric tons. 1 MT = 2,204.62262 pounds
Here's just wind since it didn't format properly below:
Wind
Construction Farm
Input Total MT
--------------- ---------
aluminum.............6,275
cement
concrete.........1,266,172
copper.............. 1,569
electricity MWh..1,691,678
excavation m3
glass................4,930
oil....................448
plastics............20,169
sand.................9,412
steel..............289,987
(1) Peterson, P. F. Will the United States Need a Second Geologic Repository? The Bridge 2003, 33 (3), 26-32.
TABLE 1: COMPARISON OF INPUTS
Hydro PV Wind Coal Nat Gas
Construction Plant Plant Farm Plant Plant
Input Total MT Total MT Total MT Total MT TOTAL MT
--------------- --------- --------- -------- -------- --------
aluminum 67 177,788 6,275 624 230
cement 2,222,356
concrete 9.906.809 1,266,172 178,320 71,270
copper 90 480,029 1,569
electricity MWh 7,556,010 1,691,678
excavation m3 4,711,405
glass 1,066,731 4,930
oil 448
plastics 20,169
sand 9,412
steel 32,183 4,600,276 289,987 62,200 51,130
Operational Inputs
------------------
coal combustion 2,336,000
coal extraction 2,336,000
transportation by railroad 2,336,000
natural gas combustion 1,560,300,000 m3
natural gas transportation 1,560,300,000 m3
natural gas extraction 1,560,300,000 m3
TABLE 2: COST, GWE (Global Warming Effect), and Area required
Total Cost Area
(1992 $) GWE required
----------- ------- --------------
Coal Power Plant 149,772,446 90,000,000 n/a
Wind Farm 206,881,416 800,000 489,580,000 m2
Natural Gas Plant 374,033,481 50,000,000 n/a
Hydroelectric Dam 503,240,216 500,000 651,141,400 m2
Photovoltaic Plant 3,578,457,990 10,000,000 51,386,400 m2
NOTE: the cost in 1992 dollars doesn't include labor, installation, or maintenance costs.
Photovoltaic Plant 100-W panels of dimensions 1.316 x 0.66 m with array units of 3 x 10 panels, each having its own concrete foundation, for a surface area of 3.9 x 6.6 m, sited at 30° latitude, at a 30-deg tilt (approximately 1.2 m of additional width is needed to account for shading by the array due to the sun's angle). There is 0.9 m between each of these array units for personnel access. Each adjacent unit covers a land area of 37.44 m2 and has a capacity rating of 3 kW. Some 1,372,500 of these 3 kW units are required.
Wind Farm location: Southern Utah, at 7,000 feet. average windspeed 6.5 m/s turbine: 600 kW in 4480 turbines
Hydropower: As the U.S. Bureau of Reclamation has suggested, "upgrading hydroelectric generator and turbine units at existing power plants is one of the most immediate, cost-effective, and environmentally acceptable means for developing additional electrical power".
Modern turbines are c. 1MW (1000kw).
4480 turbines of 660KW.
I had trouble with the units in what you posted, but 2.8 trillion pounds of contcrete is
2.8X10EE12 pounds of concrete
or 1.27 X 10EE12 kgs of concrete
or 1.27 X 10EE9 tonnes of concrete
or 283,482 tonnes of concrete per turbine?
That is basically as much concrete, per turbine, as you would use to build a substantial skyscraper.
That number looks really wrong. Similar for all the other raw materials numbers quoted.
Just on Load Factor, for any power plant it means the per cent. of the rated capacity you will achieve.
So for a wind turbine, 30% means 30% of the time it will blow at 100% of rated capacity, or 100% of the time it will produce at 30% of rated capacity.
Nukes typically run in the low 80s (distorted a bit by the fact that every few years they have a complete maintenance shutdown). All other power stations run below that level (because nuclear and hydro produce most of the baseload).
My own calculations from the problem set above
5.500 TWhr = 0.1% ish of US power consumption
5500 GWhr requires 6278MW of capacity at 100% Load Factor (divide by 8760 hrs pa)
So therefore at 0.3 LF 20,926 1 MW turbines (actually 1.2-1.4MW/ turbine is more like it).
Cost would be about $20bn.
Cost to do that in nuclear would be about $16bn (assuming 3rd Gen technology ie 4X1650MW units at $4bn each) + whatever price you care to put on waste disposal and long term decommissioning. Gas or coal would be less than $10bn but you would then have fuel cost.
So if you did that 800 times you would cover the entire US power consumption. For $1.6 trillion. Which is about 15% of US GDP now, or about equal to what the US spends on fixed commercial assets every year (capital spending by companies).
So over 20 years, 5% of US capital spending to cover the entire US energy consumption.
The estimated total cost of the war in Iraq is between 1 and $2 trillion (that was actually a 2004 estimate, so I am assuming the current costs of $15bn a month or so are offset by no rise in future costs).
Now there are a few other factors: depreciation (but that affects the turbines much more than the structures), growth in power demand (however GDP would also grow), the fact that you wouldn't use wind for all that capacity (because of grid issues).
But it's a measure of what one can achieve.
http://www.progressiveengineer.com/PEWebBackissues2002/PEWeb%2028%20Jul%2002-2/Wind.htm
If concrete weighs about 100 pounds/cubic foot * 27
= 2700 lbs/cubic yard
200 cubic yards times 2700 pounds = 540,000 lbs
540,000 / 2,000 = 270 tons per windmill foundation
270 tons times 4480 windmills = 1,209,600 tons for this windmill farm, which is within 5% of what Pacca and Horvath use (1,266,172)
Here is what their paper had to say about windmills:
A wind farm producing 5.55 TWh of electricity per year was assumed to be in southern Utah, at an elevation of 2134 m (7000 ft), close to the Escalante Desert where the average wind speed is 6.5 m/s (35). A turbine of 600 kW (36) was used as the unit for the farm's total of 4480 turbines that would occupy an area of 489 580 000 m2 (37). The total cost of materials and construction of the facility would amount to $206,881,000 (in 1992 dollars) without labor/installation and maintenance costs. Given a range of prices between $250 and $1200 per ha, the required land would add an additional $12,000,000-59,000,000 to the cost. Given the large area, land between the turbines could be used for other activities such as agriculture. No NEP loss was anticipated. The contribution of construction materials and energy to theGWE of the wind farm after 20 yr of operation (800 000 MT of CO2 equiv) is shown in Table 3.
It was assumed that after 20 yr of operation all turbines had to be replaced (but not the concrete foundations) and that the required construction energy was 30% of the original electricity and 100% of petroleum used. The electricity output of the facility remained constant. The refurbishment resulted in 900,000 MT of CO2 emissions, two-thirds of the original emissions from manufacturing and constructing the plant (1,300,000 MT of CO2).
Any 1992 wind turbine data is of historical interest only. Simply not relevant to today or, even more, tomorrow.
Alan
Second, either something is seriously wrong with this study, or wind was already a lot cheaper in 1992 than any other source of electricity: 5.55 TWhr per year, at $.10/kwhr, is worth $555 million. If this windfarm costs $200 million, then that's a 4 month payback and a return on investment of about 300% per year. In other words, this windfarm would have generated electricity at a cost of about a half penny per kwhr.
It's called a negative externality in economics.
Essentially we allow polluters to pollute without restriction, the most dangerous industrial pollutant of all-- the one that could trigger the end of human life on this planet (or, more likely, make our current civilisation unsustainable).
If you charge $100/tonne for Carbon emissions ($28 per tonne of CO2) the economics of coal look very different. European permit prices under the emissions trading scheme have reached those kinds of levels.
There is actually no economic case for allowing coal fired power, without carbon sequestration, given the potential damage of those CO2 emissions.
The fossil fuel industry, world wide, is a major recipient of government subsidies, implicit or explicit. From the destruction of natural habitats for which there are tax allowances for any restorative work (or the work is just not done) through to the high human cost of an industry with a very high mortality rate. (I won't mention lung cancer from particulates emission).
The nuclear industry is itself the recipient of massive government subsidies. The Price Anderson Act provides insurance which would not be available in private markets, limiting the liability in the case of an accident. The R&D was paid for by governments. The future waste storage liability is undertaken by governments when we have a solution.
No nuclear utility operates in a pure 'merchant power' context. British Energy tried, selling into the pool, and when the pool price crashed, it went broke and the government had to stump up £3.5bn to refinance it (to prevent a renationalisation).
Nuclear utilities across the world are either state controlled or have arrangements with the regulators that allow their cost of production to be loaded onto the consumer (effectively a guaranteed floor price).
The Bush Energy Act and the proposed British nuclear restart both provide for explicit price subsidies for new nuclear facilities.
The UK decommissioning liability for existing nuclear plants is £70bn present value.