Renewable Electricity Futures

Stanford University Wow thank you all very much is the microsoft phone working perfect okay it’s always a pleasure to come to California see colleagues and see so many enthusiastic students faculty research staff and others really interested in energy and sustainability and it’s a it warms my heart because I live eat and breathe it every day and it’s always a pleasure to be here as well even though the fogs did slow me down a little bit this morning the topic that I’d like to talk to you about today is to cover in in highlight the renewable electricity futures study that we led out of the National Renewable Energy Laboratory conducted with with many colleagues from around the country and reviewers actually from around the world we released it earlier this year if you may know it’s up online I’ll show you where that link is and I don’t know if anyone’s had time to actually read through the 800 plus pages but I’ll give you a quick synopsis of what we did and the highlights from those results and I think the important part is to recognize that it’s an investigation of the technical feasibility of renewable electricity futures for the contiguous United States it’s not a policy study it’s not a projection and things like that we were really asked to pull together a world-class group of technical analysts but also electricity sector analysts to evaluate the technical feasibility of whether or not these futures were possible and and what does that mean from a technical standpoint ie reliability meeting demand looking at the variability that’s introduced by more and more renewable electricity generation and the motivation came from as many of you likely know but if not an enormous growth in the renewable energy used domestically as well as globally this graph just shows a quick synopsis of the growth between 2000 and 2010 you know this year we’re going to surpass 160 gigawatts of installed renewables in the United States which continues the double-digit growth rate of renewables over the past numbers of numbers of decades and it’s really quite enormous and that’s not nah it’s just at the tip of the iceberg relative to the fact that there are about a hundred and fifteen countries I’ll put it that way that have some kind of renewable electricity law or goal in place and that the renewables really have become I’ll change it from what used to be thought of as alternative to just being thought of as renewable energy so it’s not no longer the alternative aspect but really when you think about it if you look at the annual growth rates in the in the United States or elsewhere it’s about fifty percent of new power additions each year so it’s really it’s really matured substantially so the questions we really had were a technical feasibility analysis they’re really three sub questions here that you can read through very quickly but just to really set up you concisely talk about them it’s can we meet load with the various supply options that we’re going to look at what’s involved in integrating all of that new supply into the grid and that’s a very very simple question but incredibly complicated to answer and and I’ll get into some of that and some of the variants that we looked at and can we I’ll call it exercise some really leading edge models with unprecedented both temporal and geographic coverage for the United States which really haven’t been exercised like this before then really explore what are this what are the constraints what are the synergies what are the challenges recognizing again that we’re not going to do an in-depth policy study the report was published earlier this year four volumes a lot of depth and breadth

a wonderful bedtime reading or weekend reading if you don’t have enough between now and finals but but feel free to look at it it’s all up online there were a hundred and ten contributors more than 30 institutions and probably another two or three hundred reviewers that we went through a pretty rigorous process a little bit like the IPCC fourth for those of you that are familiar with that where every comment was was written down and responded to and went through a very rigorous review process to get that out the door let me highlight here what the study is and what it’s not and I’m not going to read through this but I want a a couple of key takeaways I already mentioned it it’s not a policy study that would be different a lot of people would say oh isn’t that a national renewable portfolio standard well yes in terms of a quantity but no in terms of the detail of what a policy like that might look like a policy like that might look like all sorts of questions around trading state rights versus federal rights renewable electricity credits and banking and borrowing and the cost of those things etc we didn’t look at any of that so that’s why it’s not what I would call a policy study it’s a technical evaluation study as well and then other things to recognize it’s not a full reliability study it’s the best we could do at the national level to basically prove the feasibility that this should be looked at in greater detail and that’s what we’re continuing to do but it’s not sub hourly for power engineers it’s not frequency response study it’s not a you know a contingency response study and things like that it’s not down in the weeds of that but what we did do which has never been done before for the nation is to look at at an hourly basis meeting load with very different renewable supply scenarios so that’s what we went out to do there are a lot of things that we didn’t do that we hope to do going forward and I’ll lay out some of that in the future as we go forward the modeling framework just to spend a minute on this uses two principal models one is called the regional energy deployment system model reads developed for now over a decade at NREL this is a linear optimization model in two-year increments goes out to 2050 in between each time step actually goes out and does an enormous amount of statistical correlation work and reliability correlation analysis and then steps back into the linear optimization and moves forward that is actually structured for the complete united states plus imports exports to mexico and canada so we do capture those dynamics as well it actually has the temporal resolution of 17 time slices which seems kind of course per se but that’s for time slices in four seasons and a super peak and then what we do with that is actually complement that with 8760 hours where we look at that in an industry standard model called called grid view so that’s the hourly calculation that we look at and again for the whole country normally that’s done for regions because it’s so computationally intensive reed’s just to give you a sense of it it’s million million and a half variables and a few hundred thousand constraints if you’re lucky it’ll solve in 24 hours if not you have to play with it for a little bit longer and in fact one of one of your Stanford grads is that the lead technical technical modeler these days in that he’ll be out here later this week Patrick Sullivan so inputs to that came from technical teams on supply and on the technologies themselves and I’ll go through that a little bit and then we looked at the outputs of it you know what are the implications for greenhouse gas emissions actually water land use and we look at some of the economics as well we also didn’t just look at a simple set of scenarios this is probably where we spent almost the bulk of the time on on the steering committee was designing a series of scenarios or a whole suite of scenarios that could really help inform many of the questions that that each of us want to get answered in terms of a technical feasibility and here you see the suite of some 16 or 17 scenarios the core of it being focused here on what we call the exploratory scenario so what we looked at was national electricity generation so not capacitive a

generation from thirty percent to ninety percent and then we we defined a set of of sensitivity scenarios around the eighty percent scenario and here’s some additional language here relative to technology innovation so none incremental or evolutionary and then we also looked at a number of different constraints and I’m going to actually Quint spend quite some time looking at these constraints in areas and talking about those because they’re actually very insightful so these are the sensitivities that I think we’re all quite interested in when we when we look through these kind of analyses and then we had some alternatives as well as a high demand scenario and I’ll highlight a little bit of those as well but you can tell it’s it’s very rich in terms of a sweet there’s a lot of work that can be extracted and a lot of insights that can be stract it from among those those scenarios themselves and the basic set of assumptions again driven by in in this sense for the demand side of the equation so we had a whole energy efficiency and demand team run and led by Donna ha stick out of Pacific Northwest National Laboratories and you can see that that team actually referred back to the National Academies work on a potential for energy efficiency did some thinking deeply about the penetration of electric vehicles or plug-in electric vehicles and came up with essentially a fairly what what people might call aggressive baseline scenario I mean we so this going out to 2050 essentially is decreasing electricity demand out to 2050 then with a slight increase from the transportation sector making it essentially flat well that’s fairly different than i’ll call it traditional growth scenarios which have energy consumption essentially proportional to growth in GDP and you would see a half to one to one-and-a-half percent growth which is our high demand scenario so there’s a lot of assumptions built in to the demand side of the equation that we put into our reference scenario or a baseline scenario sorry that did have to do with that and then we want to look at variants around that okay so that that’s a that’s a very solid place and if key thing to remember as we go forward resources these are the maps many of them that nrel produces in fact we we update these relatively regularly so this is a kind of course scale maps of the national level of bio power wind solar geothermal direct solar so concentrating solar power CSP and hydro as well so there are details of this at very fine rez spatial resolution on the NRL website and even more detail in the Western interconnect in the eastern air connect that we’ve used for other studies as well but this is just a quick synopsis of that I think one of the key takeaways to observe on this is the enormous size of the solar resource which many of you I think are familiar with but to give that some context that’s 80,000 gigawatts right I think I actually have a mistake in my slides going forward with the wrong number but but that’s pretty enormous and the other ones and this will come out because actually did a resource constraint scenario where in fact what we did was look at well what if what if this is wrong or what if you can’t get to it or what if they’re citing problems or or development problems and things like that and that’ll be another insight so that sets the background of what we what we did and let me just focus then on highlights of the results and go through some of the scenarios so the first kind of key takeaway which was an output of the scenarios was an output of the analysis because it wasn’t predetermined was can the system actually meet hourly demand at eighty percent total renewable generation in 2050 and the answer we found was yes you could in fact keep and meet today’s technical requirements ie ancillary reserve requirements contingency reserve requirements etc with eighty percent renewable energy on aggregate across the country renewable electricity okay and so you see here what the resource map not the resource map the the actual power plant map and the Trant and the transmission map would

look like for the eighty percent incremental technology unconstrained transmission scenario looks like here and you do note a quite a difference so you see the build-out of a lot of solar in the west a fair amount of wind in the middle a fair amount of biomass throughout the West web sites throughout the East and then a significant amount of new transmission now I’ll contrast that with some scenarios where we in fact constrained the transmission and you’ll see some differences here and then I’ll actually show you some very nice visuals which will show you the build-out of this but also the hourly dispatch of it as it goes through a year which is actually fairly insightful if you’re if you’re visually inclined so that’s one of the most fundamental takeaways that came out of the study was this looks like it’s technically feasible we should look at it in some more detail and what else can we learn so this is a fairly complicated set of graphs and let me just see if I can highlight what they’re what they’re saying this is the installed capacity and gigawatts as a function of scenario for the non transmission constrained exploratory cases if you remember the scenario map so this is from thirty percent to ninety percent and but the demand is the same and so one of the key things that could recognize here is that when you add more renewables to the system you actually need more capacity why is that because as an electricity modeler you give less capacity value to the variable renewables when they increase their penetration going forward so the fiftieth marginal percent of wind gets less value to the system than the twenty-fifth marginal percent of wind okay and we do that because of the variability the predictability of the wind etc that’s all been looked at in a lot of detail there are a lot of references I can help you with on that one the oak sorry the other and so hopefully you can read these in the back these are stacked charts going from nuclear coal co fired coal gas co fired biomass dedicated biomass geothermal hydro CSP utility PV rooftop PV onshore wind offshore wind and storage the other thing that I’d failed to mention verbally in the modeling work is that we use that second NREL model called solar deployment system all solar ds2 actually model where the distributed solar would be and we did that exogenously we put it into this one and then solve the rest of the system the other piece here then is this is the percent of total supply and then maybe more importantly is this red line which is the percent of variable generation as a function of again the scenario and one of the important things that I always emphasize with this graph and also just in the discussion of the study is that when we talk about 80% scenario scenarios it’s eighty percent total renewables not eighty percent variable renewables so you can see here in the eighty percent case it’s about 42 if I get the number if I remember the number correctly variable renewable in that case although it’s eighty percent total why because we have hydro in there we have a base load biomass in there we actually have concentrating solar power CSP with storage so that it’s dispatch ball in there and even in ninety percent it’s only about 45 maybe forty six percent variable renewables so that’s also very important just to remember in the back your head we’re not analyzing scenarios with a hundred percent variable it’s it’s less than fifty percent and then here is the energy of the baseline scenario where you see again nuclear coal gas and then the renewables here and again the flat demand going forward relative to the eighty percent unconstrained transmission scenario where nuclear phases out coal phases out quite a bit gasps slims down and then you build up renewables on the side and this was done pre a fair number of changes here in the 2010 2012 time crime where you see a lot of switch between gas and coal it really wouldn’t have a lot of impact out in the 2050 time frame it would reset the start point quite a bit and I’ll talk a little bit about you know whether or not fossil fuel prices impact the results of this study and the answer just so you can think about it is not really and the simple way of thinking about that is that it’s really quite a hoop sorry a small part of the overall generation okay another key result and this is a

synthesis across many many different scenarios is that there are many different combinations of the renewable resources and the technology is available in the united states that would supply potentially 80 of renewable electricity power and that’s important to realize because each state has its unique set of assets each region has a unique set of assets and there may be preferential treatment toward a more distributed future or centralized future or protecting future you know certain land areas or water areas and that would change the mix and so the question would be is there only one solution or is there a whole variety of potential solutions and the answer is there’s a whole variety of potential solutions so here you see the the mapping of so for example from solar PV of somewhere slightly less than 100 gigawatts two maybe three hundred gigawatts is available in potential solutions those solutions vary across those 17 scenarios constraint transmission could train resources constrained flexibility whether or not the technology improves enormously or we actually stop improving now same thing with wind it also varies considerably depending upon demand and I’ll talk about that a little bit future going out and so here you see the installed capacity and here you just see the percent of total electricity there’s some notable pieces for example solar CSP there’s a scenario where we would calculate that zero CSP is actually added to the grid now it’s not very feasible given what’s already in the pipeline but economically that that’s kind of put there but you could also see that you could build more than 100 gigawatts and that’s dependent upon the technology innovation going forward and the competition with other renewable resources as well as in particular for CSP the availability of transmission to move it from the desert southwest to other demand centers all right one of the other key messages for us was really captured in the in the the geospatial bution of the energy supply and I’m going to show you a second map along this line with constrained transmission but this is one that captures a lot of details so let me let me just use an example here California what you have on the left is the generation and then what you have on the right is the capacity and the little dashed line is actually the 2050 demand so you see in California basically you meet the demand on an annual average from resources that are available in the state now that’s a different situation for the northwest where in fact it over supplies and therefore exports similarly for the Great Plains where there’s a lot of wind and therefore there’s a lot of oversupply and so again this color map is the same that’s wind up above that line which is being exported and there’s some deficits in texas and in florida and butted northeast in fact from the offshore wind they actually have a surplus and they actually export and i’ll show you that on a couple of the visuals going forward so this is a really important map because it kind of shows where we’re regionally is the supply being met by relatively local resources or where the trade flows if I could use that term between regions and I deviated from the normal renewable electricity futures talk to put this slide in because I think it’s really important this comes from work that we did for what’s called the clean energy ministerial which is a group of 23 energy ministers that get together and last year they asked us to do a study to do a comparative international case studies of the best practices for integrating variable renewables and this was kind of the visual map that came out of it and they’re kind of five key steps involved in that which is everything from expanding diverse Geographic resources enlarging balancing area authorities developing rules for flexibility of markets etc etc leading the Public Engagement particularly for transmission build out as well as market rules and things like that and I put this in here as a reflection of what I’m going to now talk about which is really about what does this flexible system mean if you’re ever going to to move

toward a very very high renewable electricity penetration scenario and this kind of gives the roadmap that’s a whole separate set of publications but I think it’s important to keep these in mind and so this is kind of alkali division but you know just a bullet list of a lot of those options for flexibility everything from demand response and in fact we’ll highlight some of those demand response aspects in the study there’s increased flexible generation particularly from the fossil resources I’m going to show you some detail of what that looks like that raises a lot of a lot of hands talking to current utility operators in the u.s less so if you talk to them in Ontario just depends on what standard and what’s practices storage and of course coordinating power markets etc so this is a really interesting set of results and their two extremes on here one is a peek set of peak demand days and this is now hourly calculations from alkyl of july three days in july actually four days in july and an off-peak spring day so in a peak scenario in the baseline everything kind of runs as traditionally was designed you have baseload plants lots of base load coal that just run flat out hydro which ramps then natural gas to floats of natural gas and then you’ve got a little bit of renewables on top of it and this is the load requirement well if the load so the load requirement that we had is the black line and then we introduce some demand response some load shifting capability in the in the scenarios and that goes from about 15 or 18 16 gigawatts to possibly as high as 30 to 50 going forward that may be actually a conservative Aereo and set of assumptions but in this eighty percent renewable scenario you still have i’ll call it some some base load like plants but then you build up a very different stack of generation profile still meeting the overall demand as it goes forward and then you get this little slice of grey at the top which is actually what we call into the study curtailment was one of the interesting assumptions that we talked about for a long period of time which is what do you do with excess electricity if you have it well in this study we chose to just call it curtailment we didn’t put it back in the system we didn’t recycle it which would be obviously the economic optimal way of thinking about it here we just said it’s curtailed it’s gotta cost of curtailment rather than thinking about it as a as a zero-cost resource so this all looks i’ll call it relatively manageable when you convert to an off-peak time it looks very very different and so here you see lots of curtailment you see virtually no coal being used and then a fair amount of cycling and on and off cycles of almost all of the resources and so it’s an operationally very different challenge but technically it’s still feasible to do that though and again those are the eighty percent iti scenarios salad would remind me how much tell 54 clock 515 five minutes past five okay okay all right I’ll keep going you then had let me contrast that I’m going to contrast this three times here that’s the incremental technology improvement scenario this is the evolutionary technology improvement scenario and because those are difficult graphs the big piece here is that you see a lot more CSP in both of these and more solar but also a fair amount of more curtailment and what does that mean it you know it basically means that you’ve got more resources available then you can use in particularly those off-peak days and therefore some more opportunity I think to use those electrons in different ways you also if you’re a markets person would recognize that those would be hours in the market where you would have incredibly negative pricing and the current pricing policy regime probably wouldn’t hold up on an economic feasibility standpoint so again a technical feasibility study not an economic one in that sense and then we’ve got the coentrao as well and here you see again the the upper curve is fairly regular the lower one has a significant amount more of photovoltaics

in it almost no concentrating solar power in it because you can’t get the power from the desert southwest to the demand centers and then also some more offshore wind but I’ll show you that in some more detail each of these solve if I can use that term over 8760 they meet the reliability requirements of today’s national electricity Reliability Council going forward on an hourly basis what this also means is that the flexibility requirements in a in our standard scenario on fossil generators are are very different and again I mentioned Ontario ramps their coal coal plants of when they have them there they’re phasing them out eighty to ninety percent of a ramp rate of ramping of full capacity that’s very different than the way our coal plants have been designed and are operated it’s just a different paradigm of how you would go about doing things but there are two things here so this shows gas combustion turbines combine cycles and coal in the baseline scenario and then in the eighty percent unconstrained transmission scenario and you just see a lot more variability you also see a lot less energy of course because most of the energy is coming from the from the renewable resources but this looks like a headache to most traditional coal plant operators not to some of the other ones who’ve been in in other areas even even the combined cycle ramping may be very fast for I’ll call it old vintage combined cycle fleets but not current Vista JH ones that are being sold in this in the system today and combustion turbines are pretty used to that kind of situation this shows just a highlight of house concentrating solar power with with thermal energy storage actually dispatches so this is a load curve this is the remaining load curve after the other variable renewables are subtracted from it and this shows then if you actually have CSP with thermal storage how it actually will dispatch into the peaks because it’s economically optimal to do that as well as technically feasible to do that as well just another key it’s a small highlight of things a couple couple tidbits on then constrained scenarios so what we what we wanted to do is ask the question have we been overly optimistic in in thinking about an economically optimal situation where you could actually build transmission and that may not be the right question but the question would be if you could not build transmission because transmission tends to be one expensive but to very slow in terms of building out because of all of the interstate rules all of the constituents that are involved in it could you still meet eighty percent and it turns out you can and so in our baseline scenario it’s actually an enormous amount of new transmission it’s a hundred and 110 to 190 million megawatt miles it’s about double our current transmission system and people might want to say well geez that seems like it’s an excessive build out what if we constrain to that so in a situation what we did was we essentially limited any new transmission to existing corridors not new ones and no interconnection ties so er cot for example stays isolated and then we reran the scenarios and said well can you actually meet 80% scenario the eighty percent requirement and the answer actually is yes and the mapping and I’m not going to spend a lot of time on this looks moderately the same but but fairly different you remember the large wind in the Great Plains that reduce by about a factor of 2 Florida increases Texas is now fully sufficient because it’s required to be but there are enough resources there and California is slightly deficient from its state level but it’s connected to the WAC and things like that but the interesting part for us was could you constrain it and still meet the requirement the answer is yes what do you trade off you trade off local resources for far away resources and that’s a that’s a trade space that people just need to be aware of and think through in terms of how much Peavy goes on a roof how much local wind or local biomass is developed vs for example wind in the Great Plains and then a transmission line across those states so this is a complicated set of comparisons but I wanted to put it up and just spend a little bit of time so you can see some of the trade spaces that we we found in those constrained scenarios so either constraining transmission or constraining flexibility what does that mean that means that we

essentially reduced Oh increase the flexibility requirements on the system for dealing with the amount of renewable generation so that means the capacity value of renewables was decreased the reliability requirements was actually increased quite substantially as well as the flexibility of those fossil of fuel generators was decreased so you can see in there that we had more operating reserve requirements for example in that system in terms of total capacity because each each reserve plant contributed less to the overall system so you can see some of this trade space quite a bit and in particular when you constrain transmission mean this went from 120 to about 30 million megawatt miles you can see comparable losses in reduction in losses of T&E etc there is quite a storage new storage being built goes from about 20 gigawatt state to about 120 to 150 gigawatts so there is new technology being built going going forward all of that is today’s commercially available technologies no new technology innovations and in fact in all these studies we didn’t we didn’t postulate new technology we looked at today’s commercial fleet and we said okay what does incremental technology improvement get you or evolutionary technology improvement gets you but we didn’t put in for example enhanced geothermal systems ocean energy floating offshore wind and things like that it was a very conservative set of Technology assumptions as we did that this shows that about the same graph that I showed you before but it it shows you the impact of those technology assumptions in terms of where does it go so you can see with no technology improvements ESP actually doesn’t actually contribute significantly to the to the system in 2050 but with evolutionary technology it really does contribute quite a bit so there’s a lot of technology improvement opportunity in the CSP CSP world and the other ones are a bit more mixed we pretty much use all of the geothermal we have available and almost all the biomass available in any of those technology scenarios as well a couple of the outputs just to think through it so one would one would anticipate that we see concomitant reductions in greenhouse gas emissions from the power sector we do see that about 80% either on a burner tip level or a life cycle emissions level we also see reductions in water demand and use in the power sector which also could be an increasingly important aspect to think about as we go forward and then we know we’re always asked well doesn’t that use a lot of a lot of land and so we looked at those metrics as well both for call it full fully disturbed lands or just marginally disturbed lands and you can see some of the comparisons here relative to golf courses corn production and major roadways so again it it you know here’s numbers I think that’s up to you know whoever wants to read the study and think about it what does that mean for either your locality or for the country relative to the trade spaces of having the greenhouse gas reductions the water reductions etc economically we also looked at the impact on direct costs on electricity prices so these are average annual electricity prices in the in this different scenarios and what you see here is a comparison against a number of different studies that came out from either the EPA or the energy information administration relative to the range across all of the renewable electricity future studies which is the kind of light blue swath in the back and you can see you know that our projections are projections but calculations are in the same ballpark and in fact potentially lower in the in the outer years again this depends upon the energy technology cost assumptions much more than on the fossil fuel assumptions as well what we didn’t include in these was the cost of the energy efficiency investments we of course didn’t look at household expenditures because we didn’t do a full economic economy-wide model so these are direct expenditures only including all the transmission upgrades or new builds that would be in there as well so it you know it looks fairly reasonable relative to the other analyses at that time and would need to be looked at in of course much more detail going forward we also looked at the sensitivity of the different prices relative to different input assumptions and here the kind of the key thing to look at at least as I think about it is what’s the sensitivity in the price relative to technology

improvement constraining the transmission or any of the other constraints demand or fossil fuel prices and of course because fossil fuel doesn’t contribute a lot the sensitivities relatively low demand drives up price as you would expect because you’d have to build more supply and a lot of that supply has low value but the most interesting part particularly for those sitting in a research university is in fact the enormous potential impact of technology innovation and it really is i think motivation for us to just work harder particularly in today’s low gas price environment we looked at the manufacturing capability this just shows kind of an annual build-out of what the what the annual installations might look like again on an economic optimization not on a business perspective somewhere between you know 20 and maybe as high as 40 or 50 gigawatts that’s not really unsurmountable compared to both us installations as they have them but also worldwide installations and worldwide production capacity as they have been so kind of the takeaway where was you know industry probably could respond to that and would actually probably like that you know to be able to really grow and and work that well the other piece that I do want to just spend a second on is is really this impact of high demand and I think it’s really important because those based assumptions of really implementing energy efficiency tend to get lost in the messages because it’s a it’s a it’s a study of renewable electricity futures but as the National Academies have done and many studies out of here and elsewhere the you know that the energy intensity of the US economy is enormous compared to most other countries and while California leads the pack and has done for decades the rest of the country has a long way to go to really catch up on this and this just shows the impact on the requirements for the different technologies going forward and we would have to build a lot more supply and deal with a lot more of the flexibility integration if we were not to curb our increase in demand and I think that that just it it’s worth thinking through what that really means particularly given overall national consumption levels so key results therefore here I’m not going to read through them I think the key thing is that we found in a technical assessment that up to eighty percent renewable scenarios are technically feasible on hourly basis meeting today’s reliability constraints and requirements and a number of different constraint scenarios and sensitivity scenarios also indicated feasibility at the national level for needing those kind of renewable electricity futures such a future would require more flexibility more flexibility both in the operation of the system but also going forward in the policy and market design elements of that system itself thinking of it the abundance and diversity of the u.s resources really allows us multiple different pathways we didn’t decide on a given pathway we wanted to say is are there multiple pathways and I think the answer to that is absolutely yes given a resource base and that the incremental cost is is actually comparable to at least other analyses of decarbonisation of the power sector that were done at that particular time and actually might be might be the same for those that are done today which are not enormous my term sorry not at not a term that we used in the study they’re real between about half a percent and one-point-two percent per year going forward so they’re not a large leap and that’s on the power price not on the total cost of power because if you look at reduced demand times the quantity of the price going up the actual expenditures may or may not be flat or decreasing depending upon what your own individual uses etc where are we going from now we have a lot more rigorous reliability announcement to be doing sub hourly looking at frequency response looking at contingency to black 22 outages and things like that we have a lot of work to look at the the market flexibility issues and the policy side of that equation which will be going forward so we’re in discussions with do we now about what that might look like and then of course continued work on the on the nology innovation side which is you know one of the most important factors on the cost of realizing such a future so let me skip through that and what I really

want to do is while we do questions show a couple of visuals if I might and I hope I can get these up oh okay so let’s see if I can get the next one up sorry we’re working on the web here so what we have here is oh I can’t get the graph down on the right anybody really fluid with a with a different resolutions as it’s a approached all right i’ll let you i’ll let you see the western part of the u.s. this is the build-out of capacity expansion from 2010 through 2050 and you can see the growth of the the resources being built out in the western part sorry i can’t resize it to show you the whole country and then you see the graph on the right hand side and let me see if i can get you the next graph these are all up on the website and see if I oh boy okay you’re going to see a bit this is actually a visual of the hourly dispatch of the electricity system for the country and what you see in the in the moving graph is the dispatch by our built up again in that stack of its fossil and nuclear built at the bottom and then bio bio geo hydro etc and if you can focus your slice of your visual at least on the bottom you can see the sun rise in the east by the solar coming up the bright yellow spots in in florida and then actually setting and then rising as it goes to the west and it’s a lot more impactful if I could show you without that big graph in the middle of it but I apologize for the for the logistics pieces of it but I encourage you to go to the website and see those it’s NRL gov / re futures so thank you for your attention thank you for inviting me out and I hope that that captured the highlights ok questions um ok how about you Charlie a lot of valuable insight I’m curious about the Scorch energy capacity it has about 110 hundred fifty watts of power capacity how long is that discharge couple hours so um that’s good question that I don’t have the answer to off top my head we can dig it out of the actual energy supplied by it and you know off the top of my head I think that it’s less than let me put it this way it’s a it’s less than about eight hours per day on average total but you know we can look at the detail of it and get you that so the trade space between storage and demand response when we took on this study now three years ago was a fairly broad unknown you know we restricted storage to really well understood well characterized storage technologies there’s no advanced batteries in here things like that this is kaze pumped hydro things like that and demand response at that point in time really was was a pre big unknown there’s a new technical feasibility study of demand response for the country coming out soon by do ii and i think that you know our estimates in here are actually fairly low relative to what that will look like but the storage is called upon just a little bit also because it’s expensive but it’s needed okay great yeah how about you back over there you mentioned you’re looking now at sub hourly issues and I’m wondering how that demand responses I think being viewer you know as having opportunities in the sub hourly balancing of the grid are you looking at that in your ferret work um we are yes so part of the ongoing work is to update at least a subset of these scenarios with current technical assessments of capability and then we will do some a fair amount of the sub hourly work at smaller geographic regions because of just computational limitations and so in those we will use the most up-to-date demand response scenarios or technology assessments that we can for those areas okay um more

questions any more students yeah nope ok alright yeah ok oh I was just curious of what sort of happy value did you end up with for the higher noble penetrations so capacity value changes as a function of the penetration as I tried to describe earlier so the the 02 teens percents of variable renewables all have you know depending on the technology say twenty to thirty percent capacity value and then that follows a functional form that we’ve developed and analyzed in quite some detail to the point where the the enth one and being greater than thirty or forty or fifty percent has almost no capacity value added to the system is in what Ranger um that transition occurs between twenty and forty percent but it depends on the specifics of the system the geographic diversity with within a given balancing area Authority and again whether or not for example CSP has storage built onto it or not so I wouldn’t look for a you know an easy answer to that but you know look at it in quite some detail okay um I think did you have a question up here second row yeah yeah did you did you take any consideration the potential increase demand theft maybe kind of out of the ordinary fan I’ve seen promote personal experiences we’ve become a more electronic society I’ve seen homes with hit a local computers multiple TV screens big huge kitchen it just seems that the residential side is are the homes are being built with a much larger capacity for demand I’m just wondering is there a possibility that dr. Mann they actually grow at a faster pace than we’re currently looking at so in the baseline scenario that we developed the residential sector actually pulls to a free flattened demand growth profile between about 20-30 and 2050 maybe as early as twenty twenty-five same with the commercial sector so that has an enormous number of behavior maybe in purchasing assumptions in it in the baseline scenario the the high demand scenario did not assume that flattening and that continues that essentially EIA reference levels of about 11 one-point-two percent per year so that might be able to capture that again assuming efficiencies are captured in other sectors as well so it would have to play out okay one more question yeah okay we’ll go with you yes dat percentage I wrong lever by the a days percent so the scenarios we looked at were between 30 and 90 and we fixed those that was a constraint on the modeling scenario so he said by 2050 or in 2050 the energy from from the in the power system had to be x percent 30 or 90 percent for the whole country so not for each state or each balancing area but for the whole country on average and so it’s it’s a modeling parameter that it’s constrained okay sorry I think we better wrap up now but but it will be up here if you have any more questions and anyway thanks for the wonderful talk and and please join us for the energy social you for more please visit us at stanford.edu