Like liveblogging, but really obscure and not live.
DUDE! I went to the most awesome seminar on Monday. It was on climate change and energy in California, mostly hydropower. I've never thought much about hydropower before because it isn't irrigation, so who cares?! But going over to other people's fields turns out to be neato. Everything is new! Even though those people tended to think about things in skewed and odd ways, they said interesting things.
First thing I noticed is that those hydropower people do not care that water is water. As far as they are concerned, their dams are holding back money. It is money shaped like water that could turn a turbine, but they do not care about its other water-like properties. All they want their dams to do is hold their money until August, when electricity is worth lots to all you air-conditioning folks. The idea that water stored behind your dam is simply stored head reminds me a little of the guy who pointed out to me that wheat is stored sunshine and water.
The first guy to shake me was the guy who gave a presentation on extreme heat events in California. They care about that stuff because heavy air-conditioning days are apparently the peak load. Who knew? It could get a lot hotter, y'all. Under the A2 Scenario* (worst case), there may be about ten times as many T9 heat events (days the temperature exceeds the 90th percentile hot day for your locale (here in Sac, that's 105° F)). Basically, the T9 events will last all summer.
I'm looking at that and thinking, "hm. Well, I like the heat. I don't like everybody whining about the heat, but by then I'll be even better at ignoring it." Then he put up a graph of daily temperature, which looks like you'd think. Low in the morning, high in the late afternoon, low in the evening and night. I'm looking at this and I was actually sure I was reading it right, but I wanted to say the words out loud. So I asked. "Does that chart say that it will be 105° by 7:30 in the morning and will be at least that hot until 10:30 at night?" Yes. That is what it said. Dude. I'm fine through the late afternoon peak, but even I don't like to roll out of bed to 105°F. Then I was thinking, "that's the kind of thing that kills old people" and then I looked at the chart and it said 2050. I'LL BE OLD! CHANGE YOUR WAYS!
Then, they were talking about the ways the new hydrology will change hydropower generation. There'll be two effects. First is that when water falls as rain, it runs off earlier and second is that there'll be less overall. Unlike farmers or flood people, they don't care that it is earlier. They can catch that in their dams whether it shows up in March as rain or April as snowmelt. But they do care that there's less. From the quick look I got, ten percent less water means seven or eight percent less hydropower.
Here's the thing I didn't get, and I'm hoping one of you can tell me. They were talking about strategies for getting more power out of less precipitation. They talked about changing reservoir operations** and raising reservoir heights (to capture more liquid money). Increasing power generation showed up in the presentation, but no one talked about it. Why not? Internet, please tell me. Why not stick more turbines at the base of your dam? I can imagine a bunch of problems, but I don't know which one is the binding constraint. Is it the cost of punching another hole in your dam and attaching it to a new power plant? Is the process of FERC re-licensing prohibitive? Is that when people put in expensive infrastructure like powerplants, they did a really good job sizing them and there aren't a lot more gains to be had? This would surprise me, because I know of power plants in Sequoia that were packed in on mules a long time ago. They're totally gorgeous, with the real brass fittings on the turbines. But can it really be true that you couldn't do much better now? Have there been big efficiency gains in water turbines in the last century? Are the mountains still so hard to reach that it is not economically worth it yet? See, internets? I'm so ignorant that I can't even guess which is the big problem.
*The cognescenti seem to know a suite of model scenarios by name. I'm starting to recognize a couple of them. A2 is Bad News, and what we're doing now. There's the one where we drastically cut back emissions and find Jesus and things only get worse by about ten percent. I think that is B-something. Those are the brackets. I guess they run these standardized scenarios through lots of different models to compare the model outcomes. Sadly, they weren't given intuitive names way back when people were first doing these models, like "SUVs Everywhere" and "I bring my own shopping bag." Maybe if they'd known these were the names everyone would be stuck with for the next fifteen years....
That reminds me that I spent a summer duplicating the very first water model done in California, which might also be the first big water model ever. It was a 1956 study of how much water would be in the Sacramento River during the summer if Reclamation hadn't built Shasta Dam. They originally did it on huge sheets of paper, with hundreds of boxes. At the front of each row, it said what the operation was and what boxes above you multiplied or subtracted or whatever. Then you go down a row and do the new operation. People filled in the boxes by hand to run the model. There was a new version a year later called version C-57 (or something like that). The "C" stood for Computer and freedom from huge scrolls of paper.
**This is code for knowing when your water is going to arrive. If you know you'll only get February storms, you hold on to every drop until you use it in August. If you know you'll get an April storm, you sell your February hydropower and store your April water until August. The big risk is that your reservoir will be full when a flood arrives and you can't catch it. That screws people downstream and also wastes money you could have captured.
First thing I noticed is that those hydropower people do not care that water is water. As far as they are concerned, their dams are holding back money. It is money shaped like water that could turn a turbine, but they do not care about its other water-like properties. All they want their dams to do is hold their money until August, when electricity is worth lots to all you air-conditioning folks. The idea that water stored behind your dam is simply stored head reminds me a little of the guy who pointed out to me that wheat is stored sunshine and water.
The first guy to shake me was the guy who gave a presentation on extreme heat events in California. They care about that stuff because heavy air-conditioning days are apparently the peak load. Who knew? It could get a lot hotter, y'all. Under the A2 Scenario* (worst case), there may be about ten times as many T9 heat events (days the temperature exceeds the 90th percentile hot day for your locale (here in Sac, that's 105° F)). Basically, the T9 events will last all summer.
I'm looking at that and thinking, "hm. Well, I like the heat. I don't like everybody whining about the heat, but by then I'll be even better at ignoring it." Then he put up a graph of daily temperature, which looks like you'd think. Low in the morning, high in the late afternoon, low in the evening and night. I'm looking at this and I was actually sure I was reading it right, but I wanted to say the words out loud. So I asked. "Does that chart say that it will be 105° by 7:30 in the morning and will be at least that hot until 10:30 at night?" Yes. That is what it said. Dude. I'm fine through the late afternoon peak, but even I don't like to roll out of bed to 105°F. Then I was thinking, "that's the kind of thing that kills old people" and then I looked at the chart and it said 2050. I'LL BE OLD! CHANGE YOUR WAYS!
Then, they were talking about the ways the new hydrology will change hydropower generation. There'll be two effects. First is that when water falls as rain, it runs off earlier and second is that there'll be less overall. Unlike farmers or flood people, they don't care that it is earlier. They can catch that in their dams whether it shows up in March as rain or April as snowmelt. But they do care that there's less. From the quick look I got, ten percent less water means seven or eight percent less hydropower.
Here's the thing I didn't get, and I'm hoping one of you can tell me. They were talking about strategies for getting more power out of less precipitation. They talked about changing reservoir operations** and raising reservoir heights (to capture more liquid money). Increasing power generation showed up in the presentation, but no one talked about it. Why not? Internet, please tell me. Why not stick more turbines at the base of your dam? I can imagine a bunch of problems, but I don't know which one is the binding constraint. Is it the cost of punching another hole in your dam and attaching it to a new power plant? Is the process of FERC re-licensing prohibitive? Is that when people put in expensive infrastructure like powerplants, they did a really good job sizing them and there aren't a lot more gains to be had? This would surprise me, because I know of power plants in Sequoia that were packed in on mules a long time ago. They're totally gorgeous, with the real brass fittings on the turbines. But can it really be true that you couldn't do much better now? Have there been big efficiency gains in water turbines in the last century? Are the mountains still so hard to reach that it is not economically worth it yet? See, internets? I'm so ignorant that I can't even guess which is the big problem.
*The cognescenti seem to know a suite of model scenarios by name. I'm starting to recognize a couple of them. A2 is Bad News, and what we're doing now. There's the one where we drastically cut back emissions and find Jesus and things only get worse by about ten percent. I think that is B-something. Those are the brackets. I guess they run these standardized scenarios through lots of different models to compare the model outcomes. Sadly, they weren't given intuitive names way back when people were first doing these models, like "SUVs Everywhere" and "I bring my own shopping bag." Maybe if they'd known these were the names everyone would be stuck with for the next fifteen years....
That reminds me that I spent a summer duplicating the very first water model done in California, which might also be the first big water model ever. It was a 1956 study of how much water would be in the Sacramento River during the summer if Reclamation hadn't built Shasta Dam. They originally did it on huge sheets of paper, with hundreds of boxes. At the front of each row, it said what the operation was and what boxes above you multiplied or subtracted or whatever. Then you go down a row and do the new operation. People filled in the boxes by hand to run the model. There was a new version a year later called version C-57 (or something like that). The "C" stood for Computer and freedom from huge scrolls of paper.
**This is code for knowing when your water is going to arrive. If you know you'll only get February storms, you hold on to every drop until you use it in August. If you know you'll get an April storm, you sell your February hydropower and store your April water until August. The big risk is that your reservoir will be full when a flood arrives and you can't catch it. That screws people downstream and also wastes money you could have captured.
posted by Megan at 4:24 PM
5 Comments:
"Why not stick more turbines at the base of your dam? "
I'm not an engineer but I don't think that'll help. The amount of energy stored in a reservoir should (and here I'm making things up) have some relationship to the force stored - i.e. the mass and the velocity. You should get the same energy out of a given reservoir no matter how many turbines you attach, otherwise we'd just add an infinite number of turbines to a river to increase our yield.
Yes, of course. The energy of water stored behind a dam is fixed, no matter how many turbines you have.
But maybe turbines aren't perfect? Some are better at extracting energy than others?
And penstocks are only so big. Would an additional one help? You could suck your reservoir dry in one gulp in August? I just don't know the capacity constraints.
...that's the kind of thing that kills old people...
An order of magnitude more old people die during extreme low temperature waves than during extreme high temperature waves. This is one channel for which the net effect of global warming is positive: if we want to help old people, we should turn up the global thermostat.
I don't say this to dispute that there are channels for which the net effect of global warming is negative -- there certainly are. But slight-of-hand antics like calling out only negative effects are what give a bad name to the people who want to use global warming to guilt others into doing what they wanted them to do even before they had ever heard of global warming.
dsw, lots of old people die during heat waves. Extremes of temperature (in either direction) are the problem, and that's what we'll be getting a lot more of as the climate changes.
During the July 1995 heat wave in Chicago, there were 696 excess deaths and 514 heat-related deaths; 72% of those who died were over age 65.
Speaking of air conditioning and peak loads, one of the major factors in this high fatality rate was widespread power outages that left thousands of people without air conditioning. Malcolm Gladwell has an interesting article highlighting the role of air-conditioner SEER standards (that's seasonal energy-efficiency ratio), which Clinton raised to 13 and Bush then lowered to 12, in the Chicago deaths. He suggests that since the difference between the peak load forecast and the actual peak load during the heat wave was relatively small, efficiency improvements like having SEER-13 rather than SEER-12 air conditioners could have kept Chicago's utility from getting overloaded and experiencing power outages. And that could have saved a lot of old people's lives.
-Liz
They originally did it on huge sheets of paper, with hundreds of boxes.
I've been going through tables of integrals this week (I'm in dire need of tricks involving Bessel functions). It's amazing that people could do so much without computers. The funny thing is, I also tried using Maple, and was really happy with it until it made a mistake... sigh.
Post a Comment
<< Home