WEBVTT 00:00:00.000 --> 00:00:02.430 align:middle line:90% [SQUEAKING] 00:00:02.430 --> 00:00:04.374 align:middle line:90% [RUSTLING] 00:00:04.374 --> 00:00:05.832 align:middle line:90% [CLICKING] 00:00:05.832 --> 00:00:10.197 align:middle line:90% 00:00:10.197 --> 00:00:11.530 align:middle line:90% DAVID HSU: My name is David Hsu. 00:00:11.530 --> 00:00:13.990 align:middle line:84% Welcome to another installment of my lecture series 00:00:13.990 --> 00:00:16.540 align:middle line:84% on urban energy systems and policy. 00:00:16.540 --> 00:00:18.970 align:middle line:84% This is a class taught by the Department of Urban Studies 00:00:18.970 --> 00:00:20.993 align:middle line:90% and Planning at MIT. 00:00:20.993 --> 00:00:22.660 align:middle line:84% And what we're going to talk about today 00:00:22.660 --> 00:00:26.140 align:middle line:84% is the next installment of talking about renewables 00:00:26.140 --> 00:00:27.680 align:middle line:90% and renewable energy. 00:00:27.680 --> 00:00:29.320 align:middle line:84% We're going to talk about solar today. 00:00:29.320 --> 00:00:31.960 align:middle line:84% And this, again, is a fundamental building block 00:00:31.960 --> 00:00:35.350 align:middle line:84% for how we are going to decarbonize cities. 00:00:35.350 --> 00:00:37.370 align:middle line:84% The readings for today are on the syllabus. 00:00:37.370 --> 00:00:38.950 align:middle line:84% If you have the PDF hard copies, you 00:00:38.950 --> 00:00:41.860 align:middle line:84% can click through to the readings in the pink links 00:00:41.860 --> 00:00:44.560 align:middle line:90% here on this page. 00:00:44.560 --> 00:00:47.470 align:middle line:84% I'll just recap what I said in the last lecture. 00:00:47.470 --> 00:00:49.660 align:middle line:90% You have a renewable toolbox. 00:00:49.660 --> 00:00:52.000 align:middle line:84% The exciting thing is that our building blocks, 00:00:52.000 --> 00:00:55.930 align:middle line:84% like solar, wind, storage, geothermal, electrolysis, 00:00:55.930 --> 00:00:58.120 align:middle line:84% are all becoming cheaper quickly. 00:00:58.120 --> 00:00:59.830 align:middle line:84% They're all cleaner technologies. 00:00:59.830 --> 00:01:02.260 align:middle line:84% But they are fundamentally different technologies 00:01:02.260 --> 00:01:04.000 align:middle line:84% than the fossil fuel system that we've 00:01:04.000 --> 00:01:06.830 align:middle line:90% built our civilization on. 00:01:06.830 --> 00:01:10.630 align:middle line:84% So we need to build about 47 times more solar, about 28 00:01:10.630 --> 00:01:13.900 align:middle line:84% times more wind than we've built in the last 40 years, 00:01:13.900 --> 00:01:16.930 align:middle line:84% to basically provide enough energy to decarbonize 00:01:16.930 --> 00:01:18.610 align:middle line:90% our energy system. 00:01:18.610 --> 00:01:21.190 align:middle line:84% That gives you a very lengthy to do list, 00:01:21.190 --> 00:01:23.890 align:middle line:84% but an exciting to do list for your careers. 00:01:23.890 --> 00:01:26.590 align:middle line:84% Your to do list includes financing, generation, 00:01:26.590 --> 00:01:29.300 align:middle line:84% transmission, distributing power, balancing power, 00:01:29.300 --> 00:01:30.820 align:middle line:84% making sure that power is reliable, 00:01:30.820 --> 00:01:32.950 align:middle line:84% making sure that the energy systems are resilient, 00:01:32.950 --> 00:01:36.100 align:middle line:84% and siting these energy resources. 00:01:36.100 --> 00:01:38.360 align:middle line:84% Last time we talked about wind basics. 00:01:38.360 --> 00:01:41.290 align:middle line:84% We talked about capacity factors as a key concept 00:01:41.290 --> 00:01:44.095 align:middle line:90% to understand intermittency. 00:01:44.095 --> 00:01:45.970 align:middle line:84% Today we're going to talk about solar basics. 00:01:45.970 --> 00:01:47.678 align:middle line:84% And we're going to use solar technologies 00:01:47.678 --> 00:01:50.830 align:middle line:84% as an example of adoption costs, or learning curves, 00:01:50.830 --> 00:01:52.330 align:middle line:90% and siting issues. 00:01:52.330 --> 00:01:55.390 align:middle line:84% And the next lecture, we'll talk about storage and geothermal. 00:01:55.390 --> 00:01:57.850 align:middle line:84% Those are technologies that are developing new niches 00:01:57.850 --> 00:02:01.810 align:middle line:84% in the energy system, and may be critical building blocks going 00:02:01.810 --> 00:02:03.730 align:middle line:84% forward, but haven't quite developed 00:02:03.730 --> 00:02:08.289 align:middle line:84% into the same kind of competitive technology 00:02:08.289 --> 00:02:10.789 align:middle line:90% that solar and wind have become. 00:02:10.789 --> 00:02:13.160 align:middle line:84% So just to review basically, here's 00:02:13.160 --> 00:02:14.930 align:middle line:90% a diagram from MacKay's book. 00:02:14.930 --> 00:02:17.960 align:middle line:84% This is a solar, basically converts incident 00:02:17.960 --> 00:02:20.210 align:middle line:90% light on the Earth into energy. 00:02:20.210 --> 00:02:23.270 align:middle line:84% Of course, you have to capture the portion 00:02:23.270 --> 00:02:26.310 align:middle line:84% of the electromagnetic energy in the ray. 00:02:26.310 --> 00:02:29.120 align:middle line:84% And so this diagram simply shows that at the equator, 00:02:29.120 --> 00:02:33.320 align:middle line:84% you capture the full amount of the energy landing in Nairobi 00:02:33.320 --> 00:02:36.200 align:middle line:84% with a horizontally-mounted ground panel, 00:02:36.200 --> 00:02:38.750 align:middle line:84% a panel simply lying flat on the ground. 00:02:38.750 --> 00:02:43.340 align:middle line:84% Whereas at 52 degrees latitude, Cambridge, United Kingdom, 00:02:43.340 --> 00:02:45.500 align:middle line:84% or Cambridge, Massachusetts, are the same, 00:02:45.500 --> 00:02:49.310 align:middle line:84% you actually only capture about 60% of the energy 00:02:49.310 --> 00:02:52.970 align:middle line:84% because a 52 degree angle doesn't capture the full ray 00:02:52.970 --> 00:02:56.030 align:middle line:84% energy of the electromagnetic energy, 00:02:56.030 --> 00:02:58.400 align:middle line:84% or the sunlight falling on the Earth. 00:02:58.400 --> 00:03:01.400 align:middle line:84% Helpfully, however, as this last line points out, 00:03:01.400 --> 00:03:04.400 align:middle line:84% 52 degrees is about the pitch of a roof angle. 00:03:04.400 --> 00:03:08.280 align:middle line:84% So if you lay solar panels flat on a roof in Cambridge, 00:03:08.280 --> 00:03:10.520 align:middle line:84% Massachusetts, and those are pointed 00:03:10.520 --> 00:03:13.610 align:middle line:84% in the direction of the sun, predominantly South, coming 00:03:13.610 --> 00:03:15.410 align:middle line:84% from the South in the northern hemisphere, 00:03:15.410 --> 00:03:18.650 align:middle line:84% then you can capture the full amount of the solar energy 00:03:18.650 --> 00:03:21.140 align:middle line:90% falling on the roof. 00:03:21.140 --> 00:03:24.660 align:middle line:84% This is why we have different maps for solar resources. 00:03:24.660 --> 00:03:27.953 align:middle line:84% This is a map of the global horizontal irradiation. 00:03:27.953 --> 00:03:29.870 align:middle line:84% This is the amount of energy you would capture 00:03:29.870 --> 00:03:33.380 align:middle line:84% if you laid a solar panel simply down on the ground flat 00:03:33.380 --> 00:03:35.300 align:middle line:90% facing up at the sky. 00:03:35.300 --> 00:03:37.640 align:middle line:84% As you can see, around the equator, 00:03:37.640 --> 00:03:39.650 align:middle line:84% there are relatively high amounts 00:03:39.650 --> 00:03:41.780 align:middle line:90% of global horizontal radiation. 00:03:41.780 --> 00:03:44.930 align:middle line:84% And, of course, slightly off the equator, 00:03:44.930 --> 00:03:48.680 align:middle line:84% you also get a lot of global horizontal radiation 00:03:48.680 --> 00:03:50.000 align:middle line:90% probably due to climate. 00:03:50.000 --> 00:03:54.290 align:middle line:84% This map probably incorporates not only climate, but also 00:03:54.290 --> 00:03:58.520 align:middle line:84% the level of radiation that's measured in these places. 00:03:58.520 --> 00:04:00.650 align:middle line:84% For the direct, normal irradiation, 00:04:00.650 --> 00:04:03.020 align:middle line:84% this is how much energy you'd capture 00:04:03.020 --> 00:04:06.710 align:middle line:84% if you laid a solar panel at the precise angle 00:04:06.710 --> 00:04:10.370 align:middle line:84% to capture the maximum amount of irradiation. 00:04:10.370 --> 00:04:13.020 align:middle line:84% And you can see that it's a much more balanced view. 00:04:13.020 --> 00:04:15.950 align:middle line:84% This probably includes climate, probably includes 00:04:15.950 --> 00:04:18.620 align:middle line:84% weather conditions, but actually is based 00:04:18.620 --> 00:04:19.730 align:middle line:90% on empirical measurements. 00:04:19.730 --> 00:04:22.730 align:middle line:84% And you can see that there, solar is actually 00:04:22.730 --> 00:04:25.950 align:middle line:84% a resource that's spread fairly well around the world, 00:04:25.950 --> 00:04:28.850 align:middle line:84% especially closer to the equator. 00:04:28.850 --> 00:04:30.560 align:middle line:84% The northern, southern hemispheres also 00:04:30.560 --> 00:04:34.240 align:middle line:90% have abundant solar potential. 00:04:34.240 --> 00:04:37.410 align:middle line:84% So if you look at the total photovoltaic power potential, 00:04:37.410 --> 00:04:40.200 align:middle line:84% again, a good message from this map 00:04:40.200 --> 00:04:44.880 align:middle line:84% is that photovoltaic power is evenly spread across the world. 00:04:44.880 --> 00:04:47.550 align:middle line:84% It can be accessed in many countries. 00:04:47.550 --> 00:04:50.340 align:middle line:84% Quite importantly, it can be accessed in parts of countries, 00:04:50.340 --> 00:04:55.690 align:middle line:84% depending on climate, depending on incident solar radiation. 00:04:55.690 --> 00:04:59.200 align:middle line:84% And, of course, we can also use solar tracking devices 00:04:59.200 --> 00:05:01.720 align:middle line:84% because the sun tends to come low 00:05:01.720 --> 00:05:04.570 align:middle line:84% from the south in the northern hemisphere, 00:05:04.570 --> 00:05:08.500 align:middle line:84% you might have a house that's not 00:05:08.500 --> 00:05:09.810 align:middle line:90% pointed in the right direction. 00:05:09.810 --> 00:05:12.340 align:middle line:90% It may not be pointed East West. 00:05:12.340 --> 00:05:16.390 align:middle line:84% You may want a rooftop facing to the South where the sun is. 00:05:16.390 --> 00:05:19.000 align:middle line:84% As the sun comes over at different portions 00:05:19.000 --> 00:05:22.210 align:middle line:84% of the year-- it's higher in the sky in the summer and lower 00:05:22.210 --> 00:05:23.660 align:middle line:90% in the sky in the winter-- 00:05:23.660 --> 00:05:25.390 align:middle line:84% you may want a solar collector that 00:05:25.390 --> 00:05:27.250 align:middle line:84% actually has different angles, angles 00:05:27.250 --> 00:05:29.860 align:middle line:84% that can tilt to capture this seasonal variation 00:05:29.860 --> 00:05:34.150 align:middle line:84% between summer and winter, and may even tilt as the sun moves 00:05:34.150 --> 00:05:36.050 align:middle line:90% throughout the sky. 00:05:36.050 --> 00:05:39.790 align:middle line:84% So that's a two axis solar tracking panel. 00:05:39.790 --> 00:05:44.080 align:middle line:84% Another technology that could be used to capture solar energy 00:05:44.080 --> 00:05:46.750 align:middle line:84% is called concentrated solar power. 00:05:46.750 --> 00:05:50.530 align:middle line:84% This is a CSP plant, or a solar thermal tower. 00:05:50.530 --> 00:05:54.220 align:middle line:84% This is Crescent Dunes in the American West. 00:05:54.220 --> 00:05:56.800 align:middle line:84% This is actually 10,000 heliostats. 00:05:56.800 --> 00:05:59.420 align:middle line:84% These are those tracking mirrors I showed you. 00:05:59.420 --> 00:06:01.690 align:middle line:84% But instead of tracking the sun directly, 00:06:01.690 --> 00:06:05.020 align:middle line:84% what these mirrors are doing are actually tracking the sun, 00:06:05.020 --> 00:06:07.420 align:middle line:84% reflecting the sun to this one point, 00:06:07.420 --> 00:06:10.390 align:middle line:84% and super heating up a fluid here 00:06:10.390 --> 00:06:12.440 align:middle line:84% that's used to drive a steam turbine. 00:06:12.440 --> 00:06:15.730 align:middle line:84% So these 10,000 mirrors are moving in concert, 00:06:15.730 --> 00:06:18.610 align:middle line:84% are all moving differently, all concentrating the sun 00:06:18.610 --> 00:06:21.310 align:middle line:84% on this one point to superheat the site 00:06:21.310 --> 00:06:25.000 align:middle line:84% and create superheated liquid or steam to drive a turbine 00:06:25.000 --> 00:06:26.530 align:middle line:90% to create energy. 00:06:26.530 --> 00:06:28.780 align:middle line:84% Of course, what we're going to talk about mostly today 00:06:28.780 --> 00:06:31.270 align:middle line:84% is much humbler, smaller technology. 00:06:31.270 --> 00:06:33.730 align:middle line:84% It's the solar photovoltaic panel. 00:06:33.730 --> 00:06:37.120 align:middle line:84% A utility-scale solar array is composed 00:06:37.120 --> 00:06:39.910 align:middle line:84% of many of these modules, a rooftop 00:06:39.910 --> 00:06:43.480 align:middle line:84% on a residential installation, or in a, let's say, 00:06:43.480 --> 00:06:46.930 align:middle line:84% camping cabin would probably be based on one standard panel. 00:06:46.930 --> 00:06:48.940 align:middle line:84% This is a solar panel installation 00:06:48.940 --> 00:06:50.890 align:middle line:90% on top of a big box store. 00:06:50.890 --> 00:06:54.140 align:middle line:84% But solar panels are fundamentally modular. 00:06:54.140 --> 00:06:55.928 align:middle line:84% They can be deployed in different ways. 00:06:55.928 --> 00:06:58.220 align:middle line:84% And we're going to spend most of our time talking today 00:06:58.220 --> 00:07:02.170 align:middle line:84% about why solar photovoltaics, that modular technology, 00:07:02.170 --> 00:07:06.310 align:middle line:84% has gotten cheap much faster than other technologies. 00:07:06.310 --> 00:07:10.010 align:middle line:84% And so if you look at the fundamental cell efficiency, 00:07:10.010 --> 00:07:15.280 align:middle line:84% this is how efficiently a solar cell converts incident energy 00:07:15.280 --> 00:07:16.760 align:middle line:90% to electricity. 00:07:16.760 --> 00:07:19.360 align:middle line:84% This is what's called the champion graph that NREL does. 00:07:19.360 --> 00:07:21.440 align:middle line:84% Every year, they update this graph. 00:07:21.440 --> 00:07:23.050 align:middle line:84% And you can see steadily, there are 00:07:23.050 --> 00:07:25.960 align:middle line:84% these different trajectories of different technologies. 00:07:25.960 --> 00:07:28.480 align:middle line:84% The technologies are all in upper left. 00:07:28.480 --> 00:07:30.400 align:middle line:84% They range from multi junction cells 00:07:30.400 --> 00:07:33.670 align:middle line:84% to crystalline silicon cells, to emerging 00:07:33.670 --> 00:07:35.230 align:middle line:90% photovoltaic technologies. 00:07:35.230 --> 00:07:38.230 align:middle line:84% You can see these technologies have all developed over time 00:07:38.230 --> 00:07:40.230 align:middle line:90% and gotten better. 00:07:40.230 --> 00:07:42.300 align:middle line:84% The best research cell efficiency 00:07:42.300 --> 00:07:47.710 align:middle line:84% is 47% conversion of incident sunlight into electricity. 00:07:47.710 --> 00:07:50.370 align:middle line:84% But your average solar panel that's 00:07:50.370 --> 00:07:55.735 align:middle line:84% deployed on most solar arrays ranges from 20% to 25%. 00:07:55.735 --> 00:07:57.360 align:middle line:84% What MacKay says in his book, that he'd 00:07:57.360 --> 00:08:00.930 align:middle line:84% be shocked if there's better research cell, or better 00:08:00.930 --> 00:08:01.860 align:middle line:90% efficiencies. 00:08:01.860 --> 00:08:04.560 align:middle line:84% Efficiencies have gone up over time. 00:08:04.560 --> 00:08:07.050 align:middle line:84% But that's not the key story I want to focus on today. 00:08:07.050 --> 00:08:10.200 align:middle line:84% It's the fact that solar panels have gotten much cheaper 00:08:10.200 --> 00:08:12.600 align:middle line:84% much more quickly, even though the research 00:08:12.600 --> 00:08:15.420 align:middle line:84% efficiency, or the efficiency of most solar panels 00:08:15.420 --> 00:08:20.560 align:middle line:84% sold has stayed relatively constant, around 20%. 00:08:20.560 --> 00:08:24.600 align:middle line:84% So if you look at-- this graph from the BP Statistical 00:08:24.600 --> 00:08:26.370 align:middle line:90% Review of World Energy. 00:08:26.370 --> 00:08:30.030 align:middle line:84% It shows the share of global electricity generation by fuel. 00:08:30.030 --> 00:08:34.650 align:middle line:84% And you can see that oil and coal and hydroelectricity 00:08:34.650 --> 00:08:36.720 align:middle line:90% are all relatively flat. 00:08:36.720 --> 00:08:38.679 align:middle line:90% Natural gas has trended up. 00:08:38.679 --> 00:08:40.919 align:middle line:84% But the one that's quite remarkably trending up 00:08:40.919 --> 00:08:41.970 align:middle line:90% is renewables. 00:08:41.970 --> 00:08:43.890 align:middle line:84% And renewables is trending up both in terms 00:08:43.890 --> 00:08:45.257 align:middle line:90% of wind and solar. 00:08:45.257 --> 00:08:47.340 align:middle line:84% If you look on the right-hand graph and renewables 00:08:47.340 --> 00:08:49.560 align:middle line:84% as a share of power generation by region, 00:08:49.560 --> 00:08:51.210 align:middle line:84% in every single region of the world, 00:08:51.210 --> 00:08:55.950 align:middle line:84% renewables seem to be on this very steep, increasing curve. 00:08:55.950 --> 00:08:58.080 align:middle line:84% That curve is reminiscent of this 00:08:58.080 --> 00:09:00.330 align:middle line:84% putting of what's called a cumulative distribution 00:09:00.330 --> 00:09:01.110 align:middle line:90% function. 00:09:01.110 --> 00:09:06.070 align:middle line:84% And this is basically simply a sum of a bell curve. 00:09:06.070 --> 00:09:08.860 align:middle line:84% And you get this S curve by summing up the bell curve. 00:09:08.860 --> 00:09:11.280 align:middle line:84% So using different parameters for the mean 00:09:11.280 --> 00:09:14.250 align:middle line:84% or the standard deviation, you get different steepness 00:09:14.250 --> 00:09:16.800 align:middle line:84% of S curves between this mustard-colored curve 00:09:16.800 --> 00:09:19.200 align:middle line:90% and this green curve. 00:09:19.200 --> 00:09:22.010 align:middle line:84% And so just to point out how that is the sum of these bell 00:09:22.010 --> 00:09:24.530 align:middle line:84% curves, those S-shaped curves come 00:09:24.530 --> 00:09:29.030 align:middle line:84% from these green, and mustard, and blue, and red bell curves. 00:09:29.030 --> 00:09:32.000 align:middle line:84% The reason why this is important is because this 00:09:32.000 --> 00:09:35.390 align:middle line:84% is an idea about how innovations diffuse, 00:09:35.390 --> 00:09:38.390 align:middle line:84% that you may have seen this famous curve 00:09:38.390 --> 00:09:39.950 align:middle line:84% called the diffusion of innovations 00:09:39.950 --> 00:09:42.530 align:middle line:84% curve that was coined by Everett Rogers. 00:09:42.530 --> 00:09:44.570 align:middle line:84% And you may have heard this phrase, "innovators 00:09:44.570 --> 00:09:45.980 align:middle line:90% and early adopters." 00:09:45.980 --> 00:09:50.360 align:middle line:84% And he adopts this idea of a bell curve of different people 00:09:50.360 --> 00:09:53.300 align:middle line:84% or different organizations willing to adopt technologies 00:09:53.300 --> 00:09:54.420 align:middle line:90% in different ways. 00:09:54.420 --> 00:09:57.140 align:middle line:84% There's innovators and early adopters who are first. 00:09:57.140 --> 00:09:59.750 align:middle line:84% There's an early majority, late majority, and there's laggers. 00:09:59.750 --> 00:10:03.380 align:middle line:84% And if you take this bell curve of behaviors and add it up, 00:10:03.380 --> 00:10:06.470 align:middle line:84% you get this distribution, this S-shaped distribution 00:10:06.470 --> 00:10:07.610 align:middle line:90% of adoption. 00:10:07.610 --> 00:10:10.250 align:middle line:84% And that's really what we want to talk about today 00:10:10.250 --> 00:10:12.860 align:middle line:84% because if you look at this S-shaped curve, 00:10:12.860 --> 00:10:15.800 align:middle line:84% this is a curve from a New York Times 00:10:15.800 --> 00:10:18.410 align:middle line:90% article by Nicholas Felton. 00:10:18.410 --> 00:10:20.090 align:middle line:84% And you can see that our adoption 00:10:20.090 --> 00:10:22.490 align:middle line:84% of different technologies over the last century 00:10:22.490 --> 00:10:25.520 align:middle line:84% have essentially followed this S-shaped curve. 00:10:25.520 --> 00:10:29.210 align:middle line:84% We have electricity which has a kind of bump here in the 1920s, 00:10:29.210 --> 00:10:30.950 align:middle line:90% but looks roughly S-shaped. 00:10:30.950 --> 00:10:33.710 align:middle line:84% You see rapid adoption of refrigerators. 00:10:33.710 --> 00:10:36.050 align:middle line:84% You see rapid adoption of color TVs. 00:10:36.050 --> 00:10:38.480 align:middle line:84% And the point of this column by Nicholas Felton 00:10:38.480 --> 00:10:40.670 align:middle line:84% a number of years ago was that these curves get 00:10:40.670 --> 00:10:42.860 align:middle line:90% increasingly steep over time. 00:10:42.860 --> 00:10:46.010 align:middle line:84% Of course, we can't assume that this is a causal process, 00:10:46.010 --> 00:10:49.328 align:middle line:84% or this is a inevitable or ineluctable process. 00:10:49.328 --> 00:10:50.870 align:middle line:84% Obviously, some of these technologies 00:10:50.870 --> 00:10:53.630 align:middle line:84% like clothes, dryers, microwaves, color TVs, cell 00:10:53.630 --> 00:10:56.180 align:middle line:84% phones, computers, VCRs, could only 00:10:56.180 --> 00:10:59.000 align:middle line:84% exist if electricity existed beforehand. 00:10:59.000 --> 00:11:02.090 align:middle line:84% You see the track of the telephone 00:11:02.090 --> 00:11:06.090 align:middle line:84% has a much more mixed, not S-shaped curve. 00:11:06.090 --> 00:11:08.720 align:middle line:84% So we shouldn't assume that a S-shaped curve is 00:11:08.720 --> 00:11:11.060 align:middle line:84% inevitable or ineluctable, we obviously 00:11:11.060 --> 00:11:14.060 align:middle line:84% have to do things to adopt these technologies. 00:11:14.060 --> 00:11:16.190 align:middle line:84% The progress of our technologies may not 00:11:16.190 --> 00:11:19.820 align:middle line:84% be completely predictable or linear. 00:11:19.820 --> 00:11:22.480 align:middle line:84% But the point I want to make about solar power 00:11:22.480 --> 00:11:26.230 align:middle line:84% is that solar power has declined by price by about 90% 00:11:26.230 --> 00:11:27.490 align:middle line:90% in the last 10 years. 00:11:27.490 --> 00:11:30.010 align:middle line:84% This is, again, that graph from Our World in Data. 00:11:30.010 --> 00:11:33.520 align:middle line:84% And the key point of this being part of your renewable toolbox 00:11:33.520 --> 00:11:35.620 align:middle line:84% is that you want to work with technologies 00:11:35.620 --> 00:11:38.740 align:middle line:84% that are getting cheaper and also happen to be cleaner. 00:11:38.740 --> 00:11:41.470 align:middle line:84% So if you look at all the energy or electricity 00:11:41.470 --> 00:11:43.600 align:middle line:84% generating technologies on this graph, 00:11:43.600 --> 00:11:46.030 align:middle line:84% again, nuclear's gotten more expensive. 00:11:46.030 --> 00:11:47.710 align:middle line:90% Coal stayed relatively flat. 00:11:47.710 --> 00:11:51.640 align:middle line:84% Gas has gotten cheaper, but not as quickly as solar power has 00:11:51.640 --> 00:11:52.480 align:middle line:90% gotten cheaper. 00:11:52.480 --> 00:11:55.990 align:middle line:84% Solar thermal towers, the concentrating solar technology 00:11:55.990 --> 00:11:58.600 align:middle line:84% I showed you a few slides ago, that kind 00:11:58.600 --> 00:12:01.330 align:middle line:84% of awesome installation in the desert, 00:12:01.330 --> 00:12:03.280 align:middle line:84% though it's awesome in how it looks, 00:12:03.280 --> 00:12:06.820 align:middle line:84% it has not necessarily gone cheap as quick 00:12:06.820 --> 00:12:09.750 align:middle line:90% as the humble solar panel. 00:12:09.750 --> 00:12:12.120 align:middle line:84% And, also, of course, onshore wind 00:12:12.120 --> 00:12:15.280 align:middle line:90% has declined rapidly over time. 00:12:15.280 --> 00:12:17.710 align:middle line:84% And, again, if you look at these curves, 00:12:17.710 --> 00:12:21.160 align:middle line:84% you can see that lithium ion batteries since 1992 00:12:21.160 --> 00:12:23.810 align:middle line:84% are also steeply decreasing in price. 00:12:23.810 --> 00:12:25.420 align:middle line:84% And I think I said in the last lecture 00:12:25.420 --> 00:12:27.760 align:middle line:84% that this is called the learning rate. 00:12:27.760 --> 00:12:29.740 align:middle line:84% For every doubling in cumulative capacity, 00:12:29.740 --> 00:12:33.700 align:middle line:84% price has declined an average of 18.9%. 00:12:33.700 --> 00:12:36.070 align:middle line:84% Solar technologies, electrolyzers, 00:12:36.070 --> 00:12:40.370 align:middle line:84% are also following a very rapid curve. 00:12:40.370 --> 00:12:43.880 align:middle line:84% And, again, the key point is that, yes, 00:12:43.880 --> 00:12:47.000 align:middle line:90% coal is a dirty fuel. 00:12:47.000 --> 00:12:51.980 align:middle line:84% But nuclear is also a low carbon or clean resource. 00:12:51.980 --> 00:12:54.650 align:middle line:84% I would argue that nuclear will probably 00:12:54.650 --> 00:12:59.030 align:middle line:84% be a critical or important firm resource for the future. 00:12:59.030 --> 00:13:02.060 align:middle line:84% But the problem is, our existing ways of building nuclear power 00:13:02.060 --> 00:13:04.580 align:middle line:84% plants have only gotten more expensive over time. 00:13:04.580 --> 00:13:07.250 align:middle line:84% They've not demonstrated any learning rate, 00:13:07.250 --> 00:13:12.110 align:middle line:84% whereas solar photovoltaics have a learning rate of 36%. 00:13:12.110 --> 00:13:14.750 align:middle line:84% Offshore wind has a learning rate of 10%. 00:13:14.750 --> 00:13:17.600 align:middle line:84% Onshore wind has a learning rate of 23%. 00:13:17.600 --> 00:13:20.210 align:middle line:84% And lithium ion batteries, which are 00:13:20.210 --> 00:13:23.210 align:middle line:84% crucial to electrifying vehicles and all 00:13:23.210 --> 00:13:27.350 align:middle line:84% of our electronic devices, have a learning rate of 19%. 00:13:27.350 --> 00:13:31.190 align:middle line:84% You can perhaps, if not count on it, 00:13:31.190 --> 00:13:33.410 align:middle line:84% assume that some of these learning rates 00:13:33.410 --> 00:13:35.930 align:middle line:84% will continue in the near future that will only 00:13:35.930 --> 00:13:40.510 align:middle line:84% make these technologies more competitive in the near future. 00:13:40.510 --> 00:13:44.290 align:middle line:84% And so a key question is, why have the solar prices declined? 00:13:44.290 --> 00:13:46.920 align:middle line:84% This is a topic that my colleague Jessika 00:13:46.920 --> 00:13:49.320 align:middle line:90% Trancik at MIT has studied. 00:13:49.320 --> 00:13:54.090 align:middle line:84% And she studied in a 2016 paper all the various factors 00:13:54.090 --> 00:13:57.360 align:middle line:84% leading to these declines in solar prices, 00:13:57.360 --> 00:13:58.440 align:middle line:90% or this learning rate. 00:13:58.440 --> 00:13:59.940 align:middle line:84% And it's also important to point out 00:13:59.940 --> 00:14:02.550 align:middle line:84% that, as the Nicholas Felton graph 00:14:02.550 --> 00:14:04.830 align:middle line:84% I showed you a few slides ago, it's 00:14:04.830 --> 00:14:07.950 align:middle line:84% not necessarily ineluctable, or inevitable, 00:14:07.950 --> 00:14:11.380 align:middle line:84% or fated for all these things to happen. 00:14:11.380 --> 00:14:14.460 align:middle line:84% But it is a little strange, and it is a little weirdly 00:14:14.460 --> 00:14:17.850 align:middle line:84% wonderful how they follow a fairly consistent pattern, 00:14:17.850 --> 00:14:20.910 align:middle line:84% even if different factors are leading to these price 00:14:20.910 --> 00:14:22.140 align:middle line:90% declines. 00:14:22.140 --> 00:14:24.270 align:middle line:84% For example, in Jessika Trancik's paper 00:14:24.270 --> 00:14:27.810 align:middle line:84% she decomposes the sources of price decline 00:14:27.810 --> 00:14:34.330 align:middle line:84% from 1980 to 2001, from 2001 to 2012, and the overall effect. 00:14:34.330 --> 00:14:37.800 align:middle line:84% And so she finds that the overall effect, the biggest 00:14:37.800 --> 00:14:42.750 align:middle line:84% contribution to solar prices declining is the efficiency. 00:14:42.750 --> 00:14:46.620 align:middle line:84% Nonsilicon material costs have also improved. 00:14:46.620 --> 00:14:49.740 align:middle line:84% Silicon prices themselves have improved. 00:14:49.740 --> 00:14:51.630 align:middle line:84% The usage of silicon has improved. 00:14:51.630 --> 00:14:53.040 align:middle line:90% The wafer areas have improved. 00:14:53.040 --> 00:14:54.880 align:middle line:90% Plant sizes have gotten bigger. 00:14:54.880 --> 00:14:56.170 align:middle line:90% The yields have gotten better. 00:14:56.170 --> 00:14:59.160 align:middle line:84% And there's been slight impact on price. 00:14:59.160 --> 00:15:01.620 align:middle line:84% But you can see from 1980 to 2001, 00:15:01.620 --> 00:15:05.460 align:middle line:84% and 2001, 2012, a lot of these different factors 00:15:05.460 --> 00:15:07.150 align:middle line:90% play in at different times. 00:15:07.150 --> 00:15:09.360 align:middle line:84% In other words, for some reason, we 00:15:09.360 --> 00:15:11.700 align:middle line:84% have these fairly steady declines in solar, 00:15:11.700 --> 00:15:13.470 align:middle line:84% though they're not completely linear when 00:15:13.470 --> 00:15:15.480 align:middle line:90% graphed on a logarithmic graph. 00:15:15.480 --> 00:15:18.720 align:middle line:84% But it's also weirdly wonderful how all these different factors 00:15:18.720 --> 00:15:19.800 align:middle line:90% play together. 00:15:19.800 --> 00:15:21.570 align:middle line:84% And when one part of the industry 00:15:21.570 --> 00:15:24.300 align:middle line:84% improved in one part of the technology, 00:15:24.300 --> 00:15:27.510 align:middle line:84% other parts of industry were then able to, let's say, 00:15:27.510 --> 00:15:28.660 align:middle line:90% improve others. 00:15:28.660 --> 00:15:31.530 align:middle line:84% For example, Germany was a pioneer in solar 00:15:31.530 --> 00:15:34.770 align:middle line:84% until China pioneered large-scale manufacturing 00:15:34.770 --> 00:15:37.140 align:middle line:90% of solar panels. 00:15:37.140 --> 00:15:40.420 align:middle line:84% If we break down some of these changes 00:15:40.420 --> 00:15:44.070 align:middle line:84% and attribute them to larger categories, 00:15:44.070 --> 00:15:46.110 align:middle line:84% you can see that there's an overall effect 00:15:46.110 --> 00:15:49.320 align:middle line:84% of public and private research and development funding. 00:15:49.320 --> 00:15:51.330 align:middle line:84% There are some effects of learning by doing. 00:15:51.330 --> 00:15:54.180 align:middle line:84% The more you do something, the better you learn how to do it. 00:15:54.180 --> 00:15:56.370 align:middle line:84% Economies of scale-- the more of something 00:15:56.370 --> 00:15:58.530 align:middle line:84% you build, it actually gets cheaper 00:15:58.530 --> 00:16:01.520 align:middle line:84% if you build it in large volumes, and other factors. 00:16:01.520 --> 00:16:05.460 align:middle line:84% But, again, between 1980 and 2001, and 2001, 2012, 00:16:05.460 --> 00:16:07.830 align:middle line:84% you actually see the really different contributions 00:16:07.830 --> 00:16:11.200 align:middle line:90% from these different factors. 00:16:11.200 --> 00:16:13.110 align:middle line:84% And so another question is, will solar prices 00:16:13.110 --> 00:16:14.805 align:middle line:90% continue to decline? 00:16:14.805 --> 00:16:16.830 align:middle line:84% And this is that learning rate graph 00:16:16.830 --> 00:16:21.360 align:middle line:84% showing solar actual asset costs by year. 00:16:21.360 --> 00:16:25.410 align:middle line:84% This is from a recent paper by Way, Ives, Mealy and Farmer. 00:16:25.410 --> 00:16:28.095 align:middle line:84% And what's really entertaining about this graph to me 00:16:28.095 --> 00:16:29.970 align:middle line:84% is that you can see the different projections 00:16:29.970 --> 00:16:31.003 align:middle line:90% for solar costs. 00:16:31.003 --> 00:16:32.670 align:middle line:84% And the important thing about this graph 00:16:32.670 --> 00:16:36.360 align:middle line:84% is that, in every single case, the empirically-observed 00:16:36.360 --> 00:16:39.810 align:middle line:84% capacity cost has been lower than the projected costs. 00:16:39.810 --> 00:16:43.380 align:middle line:84% In fact, for some reason, a lot of integrated assessment models 00:16:43.380 --> 00:16:45.300 align:middle line:84% keep on projecting that solar is not 00:16:45.300 --> 00:16:46.630 align:middle line:90% going to get cheaper faster. 00:16:46.630 --> 00:16:49.210 align:middle line:90% They have a shallower curve. 00:16:49.210 --> 00:16:51.810 align:middle line:84% But in reality, solar has gotten much cheaper 00:16:51.810 --> 00:16:54.105 align:middle line:84% much more quickly than anyone ever projected 00:16:54.105 --> 00:16:56.650 align:middle line:90% in the first place. 00:16:56.650 --> 00:17:00.700 align:middle line:84% The authors of this paper argue and forecast 00:17:00.700 --> 00:17:03.010 align:middle line:84% that solar prices will continue to get cheaper 00:17:03.010 --> 00:17:05.530 align:middle line:84% because they look at other empirical cost 00:17:05.530 --> 00:17:07.450 align:middle line:90% declines over time. 00:17:07.450 --> 00:17:11.589 align:middle line:84% If you look at wind, wind has been on a similar trajectory 00:17:11.589 --> 00:17:12.579 align:middle line:90% of decline. 00:17:12.579 --> 00:17:15.160 align:middle line:84% And in all of these cases, many models 00:17:15.160 --> 00:17:17.530 align:middle line:84% keep on predicting that wind is going to cost 00:17:17.530 --> 00:17:19.119 align:middle line:90% more than it actually has. 00:17:19.119 --> 00:17:22.000 align:middle line:84% So this leads to the question, if solar and wind continue 00:17:22.000 --> 00:17:24.609 align:middle line:84% to get cheaper more quickly than we expect, 00:17:24.609 --> 00:17:27.190 align:middle line:84% what might happen in the energy transition? 00:17:27.190 --> 00:17:28.630 align:middle line:84% The authors of this paper actually 00:17:28.630 --> 00:17:31.900 align:middle line:84% argue that there could be a fast transition between clean energy 00:17:31.900 --> 00:17:35.380 align:middle line:84% and fossil fuels, a slow transition, or no transition 00:17:35.380 --> 00:17:39.550 align:middle line:84% at all, where fossil fuels remain in our energy system. 00:17:39.550 --> 00:17:41.600 align:middle line:84% And clean energy simply adds to it. 00:17:41.600 --> 00:17:43.510 align:middle line:84% But they actually argue that based 00:17:43.510 --> 00:17:46.190 align:middle line:84% on the probability of technical and technological process, 00:17:46.190 --> 00:17:48.850 align:middle line:84% which could be low, medium, or high, 00:17:48.850 --> 00:17:50.890 align:middle line:84% we'd actually save quite a bit of energy 00:17:50.890 --> 00:17:53.830 align:middle line:84% if we had a fast transition to renewables. 00:17:53.830 --> 00:17:57.790 align:middle line:84% That would actually result in less costs in this scenario 00:17:57.790 --> 00:18:00.370 align:middle line:84% than it would if we continue to use fossil fuels. 00:18:00.370 --> 00:18:02.110 align:middle line:84% And that's assuming that solar and wind 00:18:02.110 --> 00:18:04.090 align:middle line:90% prices continue to get cheaper. 00:18:04.090 --> 00:18:06.610 align:middle line:84% If we have a slower, medium transition, 00:18:06.610 --> 00:18:10.030 align:middle line:84% then we will still save money over a no transition scenario. 00:18:10.030 --> 00:18:11.680 align:middle line:90% It just won't be as much. 00:18:11.680 --> 00:18:14.840 align:middle line:90% And it won't be as fast. 00:18:14.840 --> 00:18:16.870 align:middle line:84% If you look at some of the price breakdowns 00:18:16.870 --> 00:18:18.730 align:middle line:84% of solar installation costs, this 00:18:18.730 --> 00:18:23.800 align:middle line:84% is, again, the cost to actually install and get solar running, 00:18:23.800 --> 00:18:26.620 align:middle line:84% you can see that NREL did a study in 2015 00:18:26.620 --> 00:18:29.710 align:middle line:84% looking at benchmark prices and price breakdowns. 00:18:29.710 --> 00:18:31.600 align:middle line:84% And you can see in a lot of these cases, 00:18:31.600 --> 00:18:34.840 align:middle line:84% that residential, small installations are much more 00:18:34.840 --> 00:18:37.240 align:middle line:84% expensive than commercial large installations, 00:18:37.240 --> 00:18:39.490 align:middle line:90% or utility-scale installations. 00:18:39.490 --> 00:18:42.580 align:middle line:84% But when I point out to you is that the actual hardware, 00:18:42.580 --> 00:18:45.040 align:middle line:84% the solar module, the inverter, the rack, and the balance 00:18:45.040 --> 00:18:47.560 align:middle line:84% of system, are actually not much different 00:18:47.560 --> 00:18:49.210 align:middle line:84% between residential and commercial 00:18:49.210 --> 00:18:51.220 align:middle line:90% and utility-scale installations. 00:18:51.220 --> 00:18:53.900 align:middle line:84% The panels are basically the same panel. 00:18:53.900 --> 00:18:57.730 align:middle line:84% It might cost different amounts to mount them and put them 00:18:57.730 --> 00:18:58.510 align:middle line:90% in a rack. 00:18:58.510 --> 00:19:00.130 align:middle line:84% But the costs are not very different 00:19:00.130 --> 00:19:03.490 align:middle line:84% between even a single solar panel on a residential house 00:19:03.490 --> 00:19:05.530 align:middle line:90% versus utility-scale solar. 00:19:05.530 --> 00:19:08.480 align:middle line:84% The big difference is between the soft costs. 00:19:08.480 --> 00:19:12.070 align:middle line:84% The labor is slightly higher to install on a residential house, 00:19:12.070 --> 00:19:14.230 align:middle line:84% probably because we haven't achieved economies 00:19:14.230 --> 00:19:16.250 align:middle line:90% of scale or large volumes. 00:19:16.250 --> 00:19:19.150 align:middle line:84% But the soft costs are a big difference. 00:19:19.150 --> 00:19:20.350 align:middle line:90% Why residential? 00:19:20.350 --> 00:19:22.330 align:middle line:84% Rooftop solar is much more expensive 00:19:22.330 --> 00:19:24.400 align:middle line:90% than utility-scale solar. 00:19:24.400 --> 00:19:27.460 align:middle line:84% And so if I had to urge urban planners in cities 00:19:27.460 --> 00:19:31.370 align:middle line:84% to work on one problem, I would argue that if we were to-- 00:19:31.370 --> 00:19:34.660 align:middle line:84% if we were able to decrease the soft cost of installation, 00:19:34.660 --> 00:19:38.170 align:middle line:84% then rooftop solar could be competitive with utility-scale 00:19:38.170 --> 00:19:40.860 align:middle line:90% solar installations. 00:19:40.860 --> 00:19:43.710 align:middle line:84% And so NREL also did a study in 2018, 00:19:43.710 --> 00:19:48.630 align:middle line:84% showing a road map to achieving cheaper solar electricity. 00:19:48.630 --> 00:19:51.815 align:middle line:84% This shows the 2017 real levelized cost 00:19:51.815 --> 00:19:53.190 align:middle line:84% of energy, something we've talked 00:19:53.190 --> 00:19:55.030 align:middle line:90% about in this class before. 00:19:55.030 --> 00:19:59.280 align:middle line:84% This shows what the goal is, $0.05 per kilowatt hour 00:19:59.280 --> 00:20:01.680 align:middle line:90% for a residential PV system. 00:20:01.680 --> 00:20:05.730 align:middle line:84% And it shows that the roadmap for cheaper rooftop 00:20:05.730 --> 00:20:07.950 align:middle line:84% or residential solar goes entirely 00:20:07.950 --> 00:20:09.340 align:middle line:90% through soft cost reduction. 00:20:09.340 --> 00:20:11.530 align:middle line:84% This is a waterfall diagram again. 00:20:11.530 --> 00:20:14.430 align:middle line:84% You can see all the things you can do to get to the goal. 00:20:14.430 --> 00:20:16.800 align:middle line:84% But, obviously, the bulk of the savings 00:20:16.800 --> 00:20:18.780 align:middle line:90% comes from reducing soft cost. 00:20:18.780 --> 00:20:20.220 align:middle line:90% That's reducing permitting. 00:20:20.220 --> 00:20:21.540 align:middle line:90% That's reducing design costs. 00:20:21.540 --> 00:20:23.970 align:middle line:84% That's reducing design complexity. 00:20:23.970 --> 00:20:26.730 align:middle line:84% It's making permitting easier at city hall. 00:20:26.730 --> 00:20:29.460 align:middle line:84% If urban planners and policy makers 00:20:29.460 --> 00:20:31.080 align:middle line:84% were to focus their attention on this, 00:20:31.080 --> 00:20:34.350 align:middle line:84% we could have equally cheap rooftop solar. 00:20:34.350 --> 00:20:36.780 align:middle line:84% And that would probably not have the same land-use 00:20:36.780 --> 00:20:38.775 align:middle line:90% impacts as utility-scale solar. 00:20:38.775 --> 00:20:41.420 align:middle line:90% 00:20:41.420 --> 00:20:43.280 align:middle line:84% Now, let's go to the MacKay graphs 00:20:43.280 --> 00:20:45.440 align:middle line:84% for a second, MacKay papers for a second, 00:20:45.440 --> 00:20:47.600 align:middle line:84% because I think they usually show points 00:20:47.600 --> 00:20:50.210 align:middle line:84% about the energy density per area required 00:20:50.210 --> 00:20:51.620 align:middle line:90% for renewable technologies. 00:20:51.620 --> 00:20:54.740 align:middle line:84% This brings us into talking about siting. 00:20:54.740 --> 00:20:58.550 align:middle line:84% This is a graph of a vertical axis and energy consumption 00:20:58.550 --> 00:21:01.370 align:middle line:84% per person the population density of people 00:21:01.370 --> 00:21:02.930 align:middle line:90% per square kilometer. 00:21:02.930 --> 00:21:04.940 align:middle line:84% And you can see, the world has generally 00:21:04.940 --> 00:21:07.940 align:middle line:84% in the last two centuries, both consumed more energy 00:21:07.940 --> 00:21:09.710 align:middle line:90% and gotten more dense. 00:21:09.710 --> 00:21:11.930 align:middle line:84% The United States had higher energy consumption 00:21:11.930 --> 00:21:15.680 align:middle line:84% at the beginning and lower population densities 00:21:15.680 --> 00:21:16.855 align:middle line:90% than the world. 00:21:16.855 --> 00:21:18.230 align:middle line:84% Developing countries tend to have 00:21:18.230 --> 00:21:21.890 align:middle line:84% lower energy consumption and higher population densities. 00:21:21.890 --> 00:21:24.950 align:middle line:84% England, over the last four centuries, 00:21:24.950 --> 00:21:27.680 align:middle line:84% has gotten to very high energy consumption per person, 00:21:27.680 --> 00:21:30.050 align:middle line:84% though not as high as the US, but much higher 00:21:30.050 --> 00:21:33.580 align:middle line:84% population densities than the US. 00:21:33.580 --> 00:21:37.930 align:middle line:84% And what MacKay does is graph simply where we are currently. 00:21:37.930 --> 00:21:39.340 align:middle line:90% This is the United States. 00:21:39.340 --> 00:21:43.000 align:middle line:84% And he uses this graph to show that if you multiply 00:21:43.000 --> 00:21:46.060 align:middle line:84% these factors together, you can get these kind of isoclines, 00:21:46.060 --> 00:21:48.310 align:middle line:90% or these gradient lines. 00:21:48.310 --> 00:21:50.920 align:middle line:84% And you can see that the energy consumption of the US 00:21:50.920 --> 00:21:53.800 align:middle line:84% and the size of the circle is the population of the country 00:21:53.800 --> 00:21:56.470 align:middle line:84% is basically approaching what can be provided by energy 00:21:56.470 --> 00:21:58.840 align:middle line:84% crops, which is to say, given this energy 00:21:58.840 --> 00:22:00.370 align:middle line:84% consumption per person for the US, 00:22:00.370 --> 00:22:02.470 align:middle line:84% and given this population density, 00:22:02.470 --> 00:22:06.130 align:middle line:84% we could supply all of our energy needs with energy crops, 00:22:06.130 --> 00:22:10.030 align:middle line:84% assuming we use the entire United States for energy crops. 00:22:10.030 --> 00:22:12.760 align:middle line:84% If we add other possible technologies, 00:22:12.760 --> 00:22:16.630 align:middle line:84% you can see the limit for wind power is further out. 00:22:16.630 --> 00:22:19.270 align:middle line:84% Of course, that would require us to use much of the US 00:22:19.270 --> 00:22:20.380 align:middle line:90% for wind power. 00:22:20.380 --> 00:22:23.390 align:middle line:84% And the US has enough energy density. 00:22:23.390 --> 00:22:26.110 align:middle line:84% So we would have to use the entire country for wind power. 00:22:26.110 --> 00:22:28.100 align:middle line:84% But you can see countries like Germany, 00:22:28.100 --> 00:22:30.070 align:middle line:84% the United Kingdom, and Japan, are already 00:22:30.070 --> 00:22:32.350 align:middle line:84% at the limit for what wind power can provide. 00:22:32.350 --> 00:22:34.510 align:middle line:84% South Korea does not have enough land 00:22:34.510 --> 00:22:38.335 align:middle line:84% to provide all of its energy through wind power. 00:22:38.335 --> 00:22:39.960 align:middle line:84% If you look at these other technologies 00:22:39.960 --> 00:22:43.780 align:middle line:84% that MacKay discusses, solar PV parks 00:22:43.780 --> 00:22:46.300 align:middle line:84% in sunny locations or northern Europe 00:22:46.300 --> 00:22:48.010 align:middle line:90% are concentrating solar power. 00:22:48.010 --> 00:22:50.140 align:middle line:84% You can see that they have lower energy-- 00:22:50.140 --> 00:22:52.750 align:middle line:84% the higher energy densities or lower land take. 00:22:52.750 --> 00:22:56.360 align:middle line:84% So the frontier gets pushed further out. 00:22:56.360 --> 00:22:58.990 align:middle line:84% So for one of the students from Singapore, 00:22:58.990 --> 00:23:01.750 align:middle line:84% he asked how much, what technologies could 00:23:01.750 --> 00:23:04.240 align:middle line:90% we use to supply Singapore? 00:23:04.240 --> 00:23:08.050 align:middle line:84% You can see that it's possible to use the UK 00:23:08.050 --> 00:23:10.090 align:middle line:90% sun before conversion-- 00:23:10.090 --> 00:23:12.640 align:middle line:84% sorry, and the desert sun before conversion. 00:23:12.640 --> 00:23:15.880 align:middle line:84% Those are things that the population and energy 00:23:15.880 --> 00:23:18.040 align:middle line:84% consumption of Singapore would fit into. 00:23:18.040 --> 00:23:20.860 align:middle line:84% Obviously, it's not entirely possible to convert 00:23:20.860 --> 00:23:23.260 align:middle line:84% all of Singapore into energy production. 00:23:23.260 --> 00:23:26.050 align:middle line:84% But this gives you a back of the envelope estimate 00:23:26.050 --> 00:23:28.270 align:middle line:84% of how much energy, or how much land you 00:23:28.270 --> 00:23:31.350 align:middle line:90% need to provide that energy. 00:23:31.350 --> 00:23:33.020 align:middle line:84% And so this brings me to the paper 00:23:33.020 --> 00:23:36.770 align:middle line:84% by Carley et al 2020, which focuses on energy 00:23:36.770 --> 00:23:39.210 align:middle line:90% infrastructure and NIMBY-ism. 00:23:39.210 --> 00:23:41.610 align:middle line:84% And let's just start with what is a systematic literature 00:23:41.610 --> 00:23:42.450 align:middle line:90% review. 00:23:42.450 --> 00:23:44.040 align:middle line:84% A systematic literature review is 00:23:44.040 --> 00:23:48.570 align:middle line:84% a paper where you set particular search terms and parameters. 00:23:48.570 --> 00:23:50.813 align:middle line:84% You extract the relevant academic literature. 00:23:50.813 --> 00:23:52.980 align:middle line:84% The reason why you do this kind of literature search 00:23:52.980 --> 00:23:55.200 align:middle line:84% is that there's just too much literature to read. 00:23:55.200 --> 00:23:57.540 align:middle line:84% So we occasionally write papers to summarize 00:23:57.540 --> 00:23:59.190 align:middle line:90% what's been written. 00:23:59.190 --> 00:24:01.440 align:middle line:84% And then you code and analyze and summarize 00:24:01.440 --> 00:24:02.820 align:middle line:90% all the papers you find. 00:24:02.820 --> 00:24:05.520 align:middle line:84% And you categorize the vital data they use, 00:24:05.520 --> 00:24:08.870 align:middle line:84% what kind of methods and findings they come to. 00:24:08.870 --> 00:24:11.190 align:middle line:84% And so the findings of this particular systematic 00:24:11.190 --> 00:24:14.460 align:middle line:84% literature review is that on page 1 they say, 00:24:14.460 --> 00:24:16.830 align:middle line:84% knowledge, trust, and positive perceptions 00:24:16.830 --> 00:24:19.230 align:middle line:84% of the benefits of projects are positively 00:24:19.230 --> 00:24:21.450 align:middle line:84% correlated with support for projects with variation 00:24:21.450 --> 00:24:22.950 align:middle line:90% across energy types. 00:24:22.950 --> 00:24:26.040 align:middle line:84% They actually say that we don't focus on the positive aspects 00:24:26.040 --> 00:24:27.630 align:middle line:90% of energy projects enough. 00:24:27.630 --> 00:24:30.360 align:middle line:84% And our study of energy infrastructure and often 00:24:30.360 --> 00:24:33.360 align:middle line:84% NIMBY-ism were often focused on some of the drawbacks 00:24:33.360 --> 00:24:34.320 align:middle line:90% of these policies. 00:24:34.320 --> 00:24:37.260 align:middle line:84% And people ask, or asked in survey literature, 00:24:37.260 --> 00:24:40.000 align:middle line:90% much more about the drawbacks. 00:24:40.000 --> 00:24:42.600 align:middle line:84% And so, and also they find in this paper 00:24:42.600 --> 00:24:45.600 align:middle line:84% that in our assessment, it's not clear that NIMBY-- 00:24:45.600 --> 00:24:48.360 align:middle line:84% which stands for "Not In My Backyard"-- 00:24:48.360 --> 00:24:51.330 align:middle line:84% traditionally defined is often being evaluated. 00:24:51.330 --> 00:24:53.910 align:middle line:84% At best, we have a set of inconsistent findings regarding 00:24:53.910 --> 00:24:56.580 align:middle line:84% the weight that people give to the proximity of energy 00:24:56.580 --> 00:24:58.470 align:middle line:90% infrastructure. 00:24:58.470 --> 00:25:01.490 align:middle line:84% And, finally, they say on page 13 the high rates of support 00:25:01.490 --> 00:25:03.920 align:middle line:84% or found literature may seem to contradict 00:25:03.920 --> 00:25:07.170 align:middle line:84% what is often portrayed as significant opposition. 00:25:07.170 --> 00:25:10.310 align:middle line:84% And so this leads us to the question, how much opposition 00:25:10.310 --> 00:25:12.890 align:middle line:84% is there really to energy infrastructure? 00:25:12.890 --> 00:25:15.260 align:middle line:84% Do we actually have significant opposition 00:25:15.260 --> 00:25:16.430 align:middle line:90% to energy infrastructure? 00:25:16.430 --> 00:25:18.770 align:middle line:84% We have a million anecdotes about this happening 00:25:18.770 --> 00:25:20.000 align:middle line:90% in many places. 00:25:20.000 --> 00:25:22.940 align:middle line:84% But this paper is trying to get at the existing survey 00:25:22.940 --> 00:25:25.370 align:middle line:84% literature, summarizing a lot of papers, 00:25:25.370 --> 00:25:28.370 align:middle line:84% trying to understand, what is the actual opposition? 00:25:28.370 --> 00:25:31.160 align:middle line:84% Does it actually fit into our definition 00:25:31.160 --> 00:25:34.720 align:middle line:84% of "not in my backyard" sentiments? 00:25:34.720 --> 00:25:36.160 align:middle line:90% And so I like this graph. 00:25:36.160 --> 00:25:37.910 align:middle line:84% Actually it's in this paper quite a bit. 00:25:37.910 --> 00:25:40.090 align:middle line:84% It's a way of summarizing the papers. 00:25:40.090 --> 00:25:42.640 align:middle line:84% They look at these independent variables, 00:25:42.640 --> 00:25:45.040 align:middle line:84% whether or not they're considered 00:25:45.040 --> 00:25:47.440 align:middle line:84% in the literature, terms of knowledge, 00:25:47.440 --> 00:25:50.620 align:middle line:84% gender, trust, benefits of energy, 00:25:50.620 --> 00:25:52.780 align:middle line:84% and drawbacks of energy infrastructure. 00:25:52.780 --> 00:25:55.630 align:middle line:84% And they look at the total portfolio of energy types. 00:25:55.630 --> 00:25:57.100 align:middle line:84% Total portfolio-- and then that's 00:25:57.100 --> 00:26:01.240 align:middle line:84% composed of wind, solar, nuclear power plants, fossil fuels, 00:26:01.240 --> 00:26:03.220 align:middle line:84% and transmission and distribution lines. 00:26:03.220 --> 00:26:06.010 align:middle line:84% And the red, yellow, and blue columns 00:26:06.010 --> 00:26:10.300 align:middle line:84% represent percent of papers that include this factor, 00:26:10.300 --> 00:26:12.820 align:middle line:84% the percent of papers that find it to be statistically 00:26:12.820 --> 00:26:15.040 align:middle line:84% significant, and the percent of papers that find 00:26:15.040 --> 00:26:17.048 align:middle line:90% it to have a positive effect. 00:26:17.048 --> 00:26:19.090 align:middle line:84% So what's important to note from this graph, what 00:26:19.090 --> 00:26:21.940 align:middle line:84% I want to point out, is that wind, solar, 00:26:21.940 --> 00:26:24.040 align:middle line:84% nuclear, fossil fuels, and transmission lines 00:26:24.040 --> 00:26:29.060 align:middle line:84% have very different profiles in terms of who supports it. 00:26:29.060 --> 00:26:31.510 align:middle line:84% Not many researchers have considered 00:26:31.510 --> 00:26:35.210 align:middle line:84% asking people about their actual knowledge of wind projects. 00:26:35.210 --> 00:26:37.480 align:middle line:84% But if you include a gender variable, 00:26:37.480 --> 00:26:41.230 align:middle line:84% you'll often find that one gender, I'm guessing female, 00:26:41.230 --> 00:26:44.723 align:middle line:84% is actually quite supportive of wind projects. 00:26:44.723 --> 00:26:47.140 align:middle line:84% Not all the papers find this as statistically significant. 00:26:47.140 --> 00:26:50.380 align:middle line:84% But almost all the papers find this support of gender 00:26:50.380 --> 00:26:54.410 align:middle line:84% to be a very strong factor in supporting wind power. 00:26:54.410 --> 00:26:56.050 align:middle line:84% If you look at a different technology, 00:26:56.050 --> 00:27:00.490 align:middle line:84% let's say, nuclear technology, only about half of the papers 00:27:00.490 --> 00:27:03.760 align:middle line:84% asked people their actual knowledge of nuclear power. 00:27:03.760 --> 00:27:05.590 align:middle line:84% Only about 50% of papers find it to be 00:27:05.590 --> 00:27:06.860 align:middle line:90% statistically significant. 00:27:06.860 --> 00:27:10.390 align:middle line:84% And, actually, they find that knowledge of nuclear power 00:27:10.390 --> 00:27:15.520 align:middle line:84% is quite positively correlated with support for nuclear power. 00:27:15.520 --> 00:27:18.730 align:middle line:84% At the same, time if you include gender, almost 100% 00:27:18.730 --> 00:27:20.380 align:middle line:90% of the papers include gender. 00:27:20.380 --> 00:27:22.790 align:middle line:84% 2/3 of them find to be statistically significant. 00:27:22.790 --> 00:27:26.380 align:middle line:84% And they don't find a strong positive effect of gender 00:27:26.380 --> 00:27:28.660 align:middle line:90% sentiment about nuclear power. 00:27:28.660 --> 00:27:30.580 align:middle line:84% What I'm trying to point out this graph 00:27:30.580 --> 00:27:34.210 align:middle line:84% is that all of these factors, knowledge, gender, trust, 00:27:34.210 --> 00:27:38.060 align:middle line:84% benefits, and drawbacks, have different profiles. 00:27:38.060 --> 00:27:39.940 align:middle line:84% Some of these technologies like solar 00:27:39.940 --> 00:27:43.000 align:middle line:84% have extremely positive correlations 00:27:43.000 --> 00:27:45.970 align:middle line:84% with the benefits of solar power, 00:27:45.970 --> 00:27:48.250 align:middle line:84% whereas the drawbacks are quite often asked about 00:27:48.250 --> 00:27:50.380 align:middle line:84% and have very little difference in how 00:27:50.380 --> 00:27:52.210 align:middle line:90% people feel about solar power. 00:27:52.210 --> 00:27:54.710 align:middle line:84% What I encourage you to do is just look at this graph. 00:27:54.710 --> 00:27:56.960 align:middle line:84% Look at these different factors on the top. 00:27:56.960 --> 00:27:58.360 align:middle line:90% Look at the energy technologies. 00:27:58.360 --> 00:28:01.630 align:middle line:84% And think about how you might appeal to different people 00:28:01.630 --> 00:28:03.490 align:middle line:84% to support different technologies. 00:28:03.490 --> 00:28:08.500 align:middle line:84% Obviously, the profile for fossil fuels 00:28:08.500 --> 00:28:11.050 align:middle line:84% is quite different than the profile for transmission 00:28:11.050 --> 00:28:12.410 align:middle line:90% and distribution lines. 00:28:12.410 --> 00:28:15.310 align:middle line:84% And so we should think about this research, 00:28:15.310 --> 00:28:16.930 align:middle line:84% and think about how to use it to get 00:28:16.930 --> 00:28:19.420 align:middle line:84% people to support these different technologies that we 00:28:19.420 --> 00:28:21.120 align:middle line:90% do need. 00:28:21.120 --> 00:28:23.060 align:middle line:84% So another paper I want to talk about 00:28:23.060 --> 00:28:26.270 align:middle line:84% is written by my colleague Larry Susskind at MIT. 00:28:26.270 --> 00:28:29.090 align:middle line:84% It's just written this year about sources of opposition 00:28:29.090 --> 00:28:30.650 align:middle line:90% to renewable energy. 00:28:30.650 --> 00:28:33.880 align:middle line:84% And to summarize this paper, a team of students-- 00:28:33.880 --> 00:28:35.630 align:middle line:84% many students I know from our department-- 00:28:35.630 --> 00:28:37.640 align:middle line:84% helped build this database of opposition 00:28:37.640 --> 00:28:40.180 align:middle line:90% to renewable energy projects. 00:28:40.180 --> 00:28:43.870 align:middle line:84% And they did this by reviewing open access media reports 00:28:43.870 --> 00:28:46.810 align:middle line:84% and published scholarship to identify instances 00:28:46.810 --> 00:28:48.910 align:middle line:84% in which conflict or opposition paused, 00:28:48.910 --> 00:28:51.700 align:middle line:84% delayed, or canceled a utility-scale renewable energy 00:28:51.700 --> 00:28:52.690 align:middle line:90% project. 00:28:52.690 --> 00:28:54.460 align:middle line:84% There were media reports on projects 00:28:54.460 --> 00:28:57.250 align:middle line:84% that had been defeated, including some concerns 00:28:57.250 --> 00:29:00.400 align:middle line:84% about these projects that did not lead to stoppage or delay. 00:29:00.400 --> 00:29:02.750 align:middle line:84% They did not include these in their study. 00:29:02.750 --> 00:29:04.570 align:middle line:84% So essentially what they did is they 00:29:04.570 --> 00:29:06.130 align:middle line:84% went and looked at media reports, 00:29:06.130 --> 00:29:08.410 align:middle line:84% and looked for projects in which there actually 00:29:08.410 --> 00:29:11.900 align:middle line:90% had been delays or opposition. 00:29:11.900 --> 00:29:13.790 align:middle line:84% What they find in this database, they 00:29:13.790 --> 00:29:18.980 align:middle line:84% find 53 projects equivalent to 9,586 megawatts 00:29:18.980 --> 00:29:20.370 align:middle line:90% that were affected. 00:29:20.370 --> 00:29:23.090 align:middle line:84% They find that 34% of the projects were delayed. 00:29:23.090 --> 00:29:25.580 align:middle line:84% 50% of the projects were canceled permanently. 00:29:25.580 --> 00:29:28.940 align:middle line:84% And 26% of the projects were resumed 00:29:28.940 --> 00:29:32.030 align:middle line:84% after being stopped for several months or years. 00:29:32.030 --> 00:29:33.860 align:middle line:84% In terms of the energy transition, 00:29:33.860 --> 00:29:35.960 align:middle line:84% this is a very concerning finding 00:29:35.960 --> 00:29:38.960 align:middle line:84% because we know we need to build solar and wind much more 00:29:38.960 --> 00:29:42.020 align:middle line:84% rapidly than in the past to achieve our decarbonization 00:29:42.020 --> 00:29:44.150 align:middle line:90% goals. 00:29:44.150 --> 00:29:46.070 align:middle line:84% And they have seven distinct hypotheses 00:29:46.070 --> 00:29:48.650 align:middle line:84% regarding the sources of opposition and barriers 00:29:48.650 --> 00:29:50.840 align:middle line:90% to renewable energy development. 00:29:50.840 --> 00:29:54.170 align:middle line:84% First, communities could be expressing concerns 00:29:54.170 --> 00:29:55.770 align:middle line:84% about possible environmental impacts, 00:29:55.770 --> 00:29:58.410 align:middle line:90% including impacts on wildlife. 00:29:58.410 --> 00:30:00.930 align:middle line:84% There could be challenges to project financing and revenue 00:30:00.930 --> 00:30:02.550 align:middle line:90% generation. 00:30:02.550 --> 00:30:03.930 align:middle line:84% There could be public perceptions 00:30:03.930 --> 00:30:05.970 align:middle line:84% of unfair participation processes 00:30:05.970 --> 00:30:09.910 align:middle line:84% or inadequate inclusion in light of regulatory requirements. 00:30:09.910 --> 00:30:11.460 align:middle line:84% There could be a failure to respect 00:30:11.460 --> 00:30:14.640 align:middle line:84% tribal rights, including the right to consultation. 00:30:14.640 --> 00:30:17.010 align:middle line:84% There could be health and safety concerns. 00:30:17.010 --> 00:30:19.380 align:middle line:90% The intergovernmental disputes. 00:30:19.380 --> 00:30:22.020 align:middle line:84% And there could be potential impacts on land and property 00:30:22.020 --> 00:30:23.540 align:middle line:90% value. 00:30:23.540 --> 00:30:26.090 align:middle line:84% In the paper, one of the notable findings, I think, 00:30:26.090 --> 00:30:29.060 align:middle line:84% is that they find that most opposition to projects 00:30:29.060 --> 00:30:32.930 align:middle line:84% is not composed of a single source of opposition. 00:30:32.930 --> 00:30:35.330 align:middle line:84% As they characterize it, only 20% 00:30:35.330 --> 00:30:40.850 align:middle line:84% of the sources of opposition have one rationale 00:30:40.850 --> 00:30:42.860 align:middle line:90% or one basis for opposition. 00:30:42.860 --> 00:30:47.510 align:middle line:84% Most projects represent multiple concerns or multiple sources 00:30:47.510 --> 00:30:50.090 align:middle line:84% of opposition, or a collision of different groups 00:30:50.090 --> 00:30:52.830 align:middle line:84% that may be opposed to the project. 00:30:52.830 --> 00:30:55.010 align:middle line:84% Now, this is a very concerning finding 00:30:55.010 --> 00:30:56.750 align:middle line:84% in terms of what we are going to be 00:30:56.750 --> 00:30:59.720 align:middle line:84% able to site in terms of our future renewable energy 00:30:59.720 --> 00:31:00.930 align:middle line:90% projects. 00:31:00.930 --> 00:31:02.990 align:middle line:84% But one thing I actually noticed about this paper 00:31:02.990 --> 00:31:06.710 align:middle line:84% is that I actually thought back to other studies I've seen. 00:31:06.710 --> 00:31:09.800 align:middle line:84% This is a graph from the US Energy Information 00:31:09.800 --> 00:31:10.730 align:middle line:90% Administration. 00:31:10.730 --> 00:31:13.400 align:middle line:84% And this actually shows how much generation capacity we've 00:31:13.400 --> 00:31:16.430 align:middle line:90% added over the last 22 years. 00:31:16.430 --> 00:31:20.450 align:middle line:84% In 1990, you can see that our energy, our renewable energy 00:31:20.450 --> 00:31:24.620 align:middle line:84% generation is actually dominated by hydroconventional and 00:31:24.620 --> 00:31:26.240 align:middle line:90% hydropumped storage. 00:31:26.240 --> 00:31:30.170 align:middle line:84% Even by 2005, we had very little wind and very little solar. 00:31:30.170 --> 00:31:33.650 align:middle line:84% And by 2021, 16 years later, we basically 00:31:33.650 --> 00:31:35.330 align:middle line:84% had this massive explosion of wind, 00:31:35.330 --> 00:31:39.020 align:middle line:84% and this very rapid explosion of solar that is probably 00:31:39.020 --> 00:31:41.810 align:middle line:90% only increasing this year also. 00:31:41.810 --> 00:31:43.770 align:middle line:84% But what I realized when I calculated 00:31:43.770 --> 00:31:47.240 align:middle line:84% this is that you had this question 00:31:47.240 --> 00:31:50.450 align:middle line:84% about whether the findings of this paper are generalizable. 00:31:50.450 --> 00:31:53.540 align:middle line:84% The Susskind et al paper finds 53 projects 00:31:53.540 --> 00:31:56.060 align:middle line:90% over the years 2008 to 2021. 00:31:56.060 --> 00:32:00.720 align:middle line:84% And that basically is 9,586 megawatts, 00:32:00.720 --> 00:32:02.870 align:middle line:90% which is 9.6 gigawatts. 00:32:02.870 --> 00:32:04.730 align:middle line:84% But from that graph I just showed you, 00:32:04.730 --> 00:32:06.560 align:middle line:84% the Energy Information Administration 00:32:06.560 --> 00:32:10.340 align:middle line:84% finds roughly over the same period of time 2005 to 2021, 00:32:10.340 --> 00:32:13.340 align:middle line:90% that we added 185 gigawatts. 00:32:13.340 --> 00:32:15.620 align:middle line:84% So if you calculate the percentage, 00:32:15.620 --> 00:32:17.060 align:middle line:84% essentially the Susskind paper is 00:32:17.060 --> 00:32:19.580 align:middle line:90% looking at 5% of the projects. 00:32:19.580 --> 00:32:22.310 align:middle line:84% And so there's three perhaps stupid questions 00:32:22.310 --> 00:32:25.670 align:middle line:84% that we need to ask about this question of permitting 00:32:25.670 --> 00:32:28.460 align:middle line:84% and sources of opposition to renewable energy. 00:32:28.460 --> 00:32:30.800 align:middle line:84% First, how big is the problem of permitting? 00:32:30.800 --> 00:32:34.070 align:middle line:84% If these opposition sources are expressing themselves 00:32:34.070 --> 00:32:36.110 align:middle line:84% through denying permits for projects, 00:32:36.110 --> 00:32:40.160 align:middle line:84% we can find 5% of the projects have been blocked or opposed, 00:32:40.160 --> 00:32:43.970 align:middle line:84% does that mean that our other 185 gigawatts of projects that 00:32:43.970 --> 00:32:47.360 align:middle line:84% were built, could we have had another 185 gigawatts 00:32:47.360 --> 00:32:50.780 align:middle line:84% of projects if 50% of the projects were canceled? 00:32:50.780 --> 00:32:54.740 align:middle line:84% Or do we find in this case the projects that 00:32:54.740 --> 00:32:58.070 align:middle line:84% were canceled or opposed, and the vast majority of projects 00:32:58.070 --> 00:33:03.170 align:middle line:84% are actually built without as much controversy? 00:33:03.170 --> 00:33:06.003 align:middle line:84% And so this brings me to the question of permitting reform, 00:33:06.003 --> 00:33:08.420 align:middle line:84% which we started the class, which is discussed quite a bit 00:33:08.420 --> 00:33:09.980 align:middle line:90% in Congress, and we in class. 00:33:09.980 --> 00:33:12.240 align:middle line:90% How pervasive is this problem? 00:33:12.240 --> 00:33:14.190 align:middle line:84% And, finally, are the future problems 00:33:14.190 --> 00:33:16.190 align:middle line:84% we're going to have around sources of opposition 00:33:16.190 --> 00:33:19.580 align:middle line:84% or permitting similar to the past problems that have 00:33:19.580 --> 00:33:21.290 align:middle line:90% been found in this database? 00:33:21.290 --> 00:33:22.790 align:middle line:84% Those are all open questions for us 00:33:22.790 --> 00:33:24.620 align:middle line:90% to consider in class discussion. 00:33:24.620 --> 00:33:29.050 align:middle line:84% I look forward to seeing you in class tomorrow. 00:33:29.050 --> 00:33:46.000 align:middle line:90%