 Thank you very much for this opportunity to present some of our research here today. My name is Fatih Yunal, and I'm a data scientist for the fossil fuel non-proliferation treaty initiative, and I'll be talking about machine learning classification of fossil fuel supply site policies. I would like to acknowledge my collaborators, Frederick Daly, Shashi Batlow, and Tobias Brosch, and I just realized I forgot to put Peter Newell's name on there too, sorry. So, understanding how fossil fuel supply site policies such as bans, moratoria, limitations emerge in public and private sectors is of course critical to ensure timely and effective climate actions. However, as you might know, as the volume of text-based information on fossil fuel supply site policies increases, identifying, categorizing, monitoring set policies has become increasingly difficult. My co-authors and I believe that natural language processing, or NLP, can be used to assist in monitoring these large and growing bodies of text. And so the aim of our study, our project, was to classify policy documents by policy category. For example, new documents pertaining to supply site restriction policies would be automatically labeled as such, while as restriction, while as adaptation policies, for example, would be automatically labeled accordingly. Ideally, human annotators would monitor new policy developments around the globe, but as we all know, human annotators are expensive and there are more pressing, other more pressing tasks present requiring human intelligence. Instead, in our project, we train an artificial neural network, or an ANN, an NLP model to automatically classify new documents to assist researchers, policymakers, and representatives of civil society in their policy tracking and monitoring workflow. As I said, we use an artificial neural network to train a state-of-the-art AI system called Tensilflow, developed in 2015 by Joffrey Hinton and colleagues in Google's brain team. ANN has achieved remarkable results in a variety of natural language processing tasks, including document classification. Overall, our explorative case study shows that using machine learning for policy text classification achieves high accuracy in identifying relevant documents and has thus the potential to assist in policy research and advocacy efforts. A shameless plug before I get to the ANN part. For our case study, we use a text corpus, which is really available online from the fossil fuel nonproliferation initiatives database. The database contains expert annotated policy text referencing fossil fuel supply-side policies around the globe. Here you can see the landing page of the online application. The app visualizes the current state of fossil fuel supply-side policy development and currently contains over 1,500 commitments, including moratorial limitations, divestment pledges, etc. On the top of the application, you can see some key statistics, and on the bottom right, the data can be filtered by different policy types and levels. On a different tab of the application, one can select an individual country and is provided with a profile presenting key statistics. And here you can see the download page of the app, where the database can be filtered by country policy type, etc. And it is freely available for download. So this is our data source. Here you can see three examples of supply-side policies, namely a ban on the left side, a moratorium in the middle, and a limitation on the right side. So these are the types of text that we use to train our artificial neural network. As contrast categories, we use UK policies on climate change adaptation from an existing database. But what is machine learning and how does it work? Of course, 10 minutes time is not nearly enough to explain this, but just to give you a broad idea about it. Since the aim of our case study is to identify supply-side references from policy text, we make use of a distinct class of machine learning algorithms within the general machine learning literature, artificial neural networks, or ANN. Machine learning generally refers to a group of data-driven methods that combines tools from computer science and statistics to learn from data-driven methods. Central to machine learning is the process whereby the algorithm learns some pattern of a data set and subsequently tests its performance on another data set. Here we develop an artificial neural network model, capable able to classify textual passages pertaining to fossil fuel supply-side policies. And what we want to achieve is the neural network to automatically distinguish supply-side policies from similar but distinct text categories such as climate change adaptation policies. These text sorts are usually found within the same relevant text corpora such as government databases. So to this end, our model receives text as input and should be able to produce a label to classify the input as either fossil fuel supply-side, climate change adaptation, or non-climate change related policies with a certain probability. To produce such a model, we first selected an appropriate corpus as I detailed in a previous slide and we worked with 900 policy documents in total, divided this set of documents into an 80% training set, a 10% validation set, and a 10% hold-out set to evaluate the models. Our key skill metric is the overall accuracy score with 100% indicating a perfect classifier, 0% indicating total confusion. As you can see in this figure, the neural network yielded an overall accuracy of 83%, which means 83% of the predictions were reliable. So to summarize, how much time do I have? Yes, to summarize, in this project we explored the use of machine learning methods for the study of fossil fuel supply-side policies. We argued that machine learning can be exploited to assist in identifying, tracking, and monitoring supply-side policies and our results confirmed that machine learning is indeed useful to this task. More specifically, 83% accuracy is quite a remarkable result, given that the policy text that we used shares some significant overlap with each other. But the ML also fared very well in comparison to a more rigorous and realistic benchmark, which is in comparison to human expert annotators showing substantial agreement with those. In conclusion, ML is really useful for policy research. So what is the advantage of having artificial intelligence assist in policy research, you might ask? And what are further use cases and future directions? So the obvious advantage of having a semi-automated software that is able to automatically classify new documents is that it saves time and resources, plain and simple. It provides, thus because this process is speeded up, it provides almost real-time tracking of new documents, which in traditional research context will be much harder to achieve. The technology used to create such a software is freely available. Thus, there is no additional cost making this tool very useful for research and advocacy purposes. What are the potential use cases? I have 30 seconds left. I'm happy to talk about potential use cases and future directions in the Q&A. But one example is to use ANN in multilingual context. So we just worked with English text, but the model could be retrained in any other text as well. Another use case would be to expand geographical coverage, because one could use news articles referencing new policies and train an artificial neural network for that purpose. Thank you very much. So what I'm going to present here is a multi-model assessment and exploration with integrated assessment models that we did almost a couple of years ago by now with various co-authors among which Steve Pied, that may be here in the room, and other two modeling themes to explore what would be the implication of worldwide supply-side policies by which we mean essentially extraction ban. So traditionally, as most of you I assume might know, the carbon tax has been regarded. So demand-side policies and especially the carbon tax has been regarded as the first best solution to the climate externality, and that stems from economics, basically. And that's the main reason why in almost all virtually all the modeling exercises that have a global scope, so the IAM scenarios reviewed by the AIPCC, the carbon tax and demand-side policies are the main tool. And for the same reason, the carbon tax is the main instrument that the literature on international climate mitigation and cooperation focuses on, the idea being that it should be easier and more efficient to agree on a single number, which would be the global carbon tax. Now, there is a lot of truth in that, but there are also challenges because the real world is different than the modeling world. So we kind of challenge that assumption and we recognize clearly all of you that are here know this, that there is the potential of using supply-side action and in particular extraction of fossil fuel as a climate policy instrument. So we wanted to see in a modeling exercise what would happen if we did use those kind of instruments within global mitigation scenarios. So yeah, that's the scope. The models we use are the WICH model, which we maintain and develop at CMCC and then remind the PIC model, TMUCL from UCL and Prometheus from ATEMS. So yeah, in particular we want to assess two main research questions. The first one is how far and how fast extraction ban can go as a substitute of carbon pricing. So we want to take the models, we want to put extraction ban there and we want to see what happens basically in terms of climate mitigation pathways. In order to do that, we design some scenarios. We have of course a reference scenarios, which is current and stated policy. This goes in global and multi-regional integrated assessment models. Then we have one scenario in which we ban only coal to see what happens when you ban only one fossil fuel. And then two scenarios in which all fossil fuels are banned, so-called oil and gas, but at different speeds. In one scenario we design an exogenous narrative in which we try to mimic that international cooperation is hard. And so we have basically a gradual phase down, a gradual regional phase down of each hydrocarbon and then the dark red scenario in which actually we ban everything together. So we simulate a perfect operation, which would be similar, a kind of counterfactual of what a global carbon tax would look like or wouldn't look like most likely. So this is kind of an idealized scenario in terms of cooperation. And then the second research question is how do carbon pricing or demand-side policies interact with supply-side policy in the context of two-degree scenarios? So here we have basically other two scenarios, what we call the DMD, which is a classical standard well below two-degree that is very similar to many of the IPCC scenarios that you're used to. And then on top of that we add also the same supply-side policies, the same extraction ban that we have in the sub-scenarios of the fragmented narrative. I'm not going to go into the detail of how we built the scenarios, but just to say that we take into account that different fuels are more difficult to ban because they are more important for the energy transition and that different fossil fuels are more cardamom intensive. So we ban first coal and then oil and gas. Also we keep into account the existence of residual production here. If you look at the red line, the red line is the exogenous trajectory that we impose when constraining global production. The lines there are just to show that historical precedents just to look at how ambitious these bans are in terms of historical precedent. Clearly it's different because we do it for climate mitigation. All those scenarios where all those, the production there follows the different logic especially for the oil crisis in 1973 and 1979. So I would say the main result here, so again this is fairly standard. Our reference scenario is actually fairly optimistic. We extrapolate a current and stated policy and we see sort of declining CO2 in the second part of the century. The first thing that is fairly evident is that banning godly coal, when I say ban, I say it just for the sake of simplicity, but actually it's a strong reduction. We allow a residual production. It helps reducing emissions, but it's not enough. So the story is more complex than just coal, and that of course we kind of knew already. And then what we can see looking at the red lines versus the black and the green lines is that supply side policies, the way we design them, can actually bridge the current level of effort with a two-degree scenario. And actually if the action is coordinated, the emission trajectory kind of looks like it's compatible with a two-degree scenario until around mid-century, 2050. What you see then is a divergence and this is one, I think, one of the key results. And you see that divergence and you see these are the residual emissions allowed by our scenario design. But if you look at the primary energy supply in the scenarios, it's actually very similar. So the green line and the red lines have fairly similar production level in terms of fossil fuel. The big difference there is CCS and negative emission. So carbon pricing incentivizes a wider basket of technological options in the power sector and in the industry sector with respect to just cutting off the fossil fuel. Which means that supply side policies alone can substitute basically carbon pricing until mid-century but then they fail to phase in key technologies such as, well, what are assessed to be key technologies such as CCS and negative emission like bioenergy with carbon capture storage or direct capture. Of course, the amount of residual emission is dependent by our scenario design. So that this is around 10 gigaton per year in 2100. That's compatible with the residual emission range in the IPCC, but that's just one number. It could be five gigaton. The important part is that you cannot get to net zero with supply side policies alone unless you believe that you can basically completely phase out for fossil fuels from the energy system, which as we understand now, the energy system is not the case even if there is, I would say, some debate now about that. On the other end, another interesting result is that instead, if you compare the two green lines so with and so the carbon budget scenario with and without supply side policy, you see that the reliance, if you add the supply side policy, and of course if these supply side policies are relevant, meaning that they reduce demand with respect to fossil fuel demand with respect to what you would have in the in the demand side only scenarios, what you get is a reduction in these CCS and negative emissions. So you still have them, but they are reduced, which is a positive policy prescription. I would say because I just said that negative emissions and CCS are important to reach net zero, but the consensus that is developing around these technologies is that since they are risky, the less you have them, the less you rely on them, the better it is. So then we actually did a much broader analysis, but I don't have the time for that. I will just say that what we found is that supply side policies as substitutes of carbon pricing have advantages. For example, yet they can have positive effect outside the coalition that implements them because they increase the international market price of fossil fuel, but they don't allow to reach net zero. Instead, if we use together the two policies and we balance them together in a very careful way, we have a lot of co-benefits among which we'll use less negative emissions. We keep the revenue side to producer, which is kind of debatable if it's good or bad, but if you want, we can discuss them more. And we reach the same carbon budget with a lower carbon tax. So the bottom line is there is a lot of space to explore extraction bond together with carbon prices or other demand side policies, but they must be, I would say, coherent in terms of ambition. So they must be both there and they must be coherent in terms of relative ambition. So yeah, thanks. So my name is Nicholas and today I will be presenting on behalf of my co-authors, Dr. Paivi Luala and Dr. Philippe LeBiong, unfortunately he couldn't be with us, but they're here in spirit. This presentation will essentially be two, the summaries of two research projects I've been involved with. The first one being a fossil fuel cuts database. So there might be a bit of overlap with Fatih's presentation. I'm essentially a pre-modern interpretation of the database. Looks like automation has cost me my job. And the second will be understanding the determinants at a national level of adopting supply side constraints. Just before we start, as a terminology matter, a supply side fossil fuel cut, which is a word you'll hear me say about a thousand times in the next 10 minutes, is essentially any measure or instrument used by a civil society or by a government to constrain the exploration or production of fossil fuel reserves. So I'll very often refer to them as fossil fuel cuts. Let's begin. Here we are. So the fossil fuel cuts database essentially was the first database that catalogued all initiatives that sought to constrain fossil fuel production. So fossil fuel cuts pretty much. The first case goes back as far back as 1905. However, our research is really focused on 1988 onwards. This is what our publications have all focused on. We have seven different instruments that we have inventoried, which I will be discussing in two slides and about two slides. So these are instruments that can be used by civil society, organizations or governments to stop, to keep fossil fuels in the ground directly, preferably indirectly if necessary. And there have been updates since 2018, which is when it was first created. I updated about twice a year between lectures when there's time. So try to keep it pretty fresh. So what have we found through this database? We have found 1,930 cases since 1988. If we go back to this first case in 1905, we have 1,970 cases. And we have noticed that over half of our countries have actually adopted at least one fossil fuel cut in its history, whether this is blockade or whatever. We'll look into that later. Unfortunately, we do notice there's some very strong regional or geographical patterns. The United States alone accounts for pretty much a third of all fossil fuel cuts that have been adopted since 1988. And then the following countries with our Australia, the United Kingdom, Canada and Nigeria. But if you take the top 10 most adopting countries, eight of them aren't the OECD. So we do recognize that fossil fuel cuts are occurring primarily in western OECD nations, which we'll have some interesting implications later. In terms of the temporal distribution, so here on this very basic graph, you see that blockades long dominated the fossil fuel cuts game until about 2012, after which they skyrocketed, they plummeted quite the opposite. The same year divestments took off and keep dominating the game today. We see around the late 2010s that moratoria slash bands kind of picked up at the same time as, what is it, litigations? Yeah, litigations. However, unfortunately, moratoria and bands have kind of tapered off in the last five, six years, whereas litigations are still going quite strong. And the stragglers of this whole story are the subsidy phase outs, emissions trading schemes, and supply side carbon taxes, which have been in the background, they've been very regular, quite infrequent, which is a shame, but that's something we should focus on. Next. Sorry. So that was essentially the fossil fuel cuts database and the findings. The next one is about the determinants of supply side initiatives. So essentially, this research was developing a conceptual framework for factors affecting the likelihood of having active constraints for national governments. So looking at 124 countries with fossil fuel reserves between the years 2006 and 2019, we have identified several factors that contribute to the likelihood or not of having an active measure at any given time that has been implemented by a national government. I do want to stress this is only for national governments and not subnational governments or civil society organizations. That is for future research, ideally. In terms of what we have found, well, sorry, yeah, so for the conceptual framework, we've looked at four main factors, which were then divided into a series of independent variables. So you see the four main factors, the fossil fuel sector, economic factors, institutional factors, and climate change factors. And you can see the various independent variables we looked at. And the seven instruments I highlighted earlier are the dependent variables. In terms of our findings, so we've noticed that a lot of this might seem a bit obvious. For instance, countries with an active civil society that promotes constraint measures is more likely to have an active measure. Makes sense. On the flip side, anocracy, so imperfect democracies or totalitarian democracies, are less likely to have active measures when it comes to moratoria. There's not sufficient evidence for other types of instruments. Similarly, countries in current or post-conflict contexts are also less likely to have active measures, but this is not the case for subsidy phaseouts. We have also noted some weaker evidence. Please do, I emphasize the word weaker evidence that increasing levels of democracy also lead to increasing likelihood of having an active measure. And very interestingly, increased levels of corruption also indicate higher likelihood of having an active measure. In terms of our economic factors, we note that richer countries are more likely to have active measures, and this is more specifically for moratoria, specifically I guess. And on the flip side, countries with higher debt rates are less likely to have active measures, though this is also more tentative results. We have also noted that foreign direct investments does not seem to have an impact either way. False of your factors, countries with a higher dependence on false of your revenue, less likely to have active measures, and the same goes for active OPEC countries or countries who are part of OPEC concerning moratoria specifically, which will have some very interesting implications in terms of a coalition building, which I'll discuss at the very end. We have also noted that the size of fossil fuel reserves and the scale of production of fossil fuels does not have an impact either way, which also is very interesting finding in terms of coalition building. But again, that's for the end. And last but not least, climate change factors. So countries that are vulnerable to climate change impacts are less likely to have an active measure, understandably, I guess. And though this is not a climate change factor, we've also found that countries with bigger populations are also more likely to have active measures. So in terms of what this actually means, the aggregate of all these findings, fossil fuel cuts are rising, which is fantastic news in many ways. However, they're rising in a very unequal manner. As noted earlier, it's mostly Western OECD countries who are doing most of the fossil fuel cuts. In fact, North Africa and the Middle East have adopted virtually no fossil fuel cuts. So there's quite a discrepancy here. And national governments are total laggards. Most fossil fuel cuts, about 75% of them actually come from civil society organizations. The rest tend to come from subnational governments, and then the national governments tend to just pick up the crumbs. So this is something we should definitely be targeting in terms of coalition building. Secondly, fossil fuel reserves and production don't strongly discourage the adoption of fossil fuel cuts. This is very encouraging because in terms of coalition building, so for an international managed decline, for which we need coalitions of willing countries to move first and second, this expands the scope of possible partners. We can now start looking at countries that are not traditionally seen as like potentially interested in fossil fuel phase outs. So this, according to our research or to our findings, this would involve oil-rich countries, but who are not dependent on their oil revenues. I'm sure you can think of a few. In terms of other findings, OPEC's disinterest in moratoria, for instance, also can introduce this notion of instrument-specific coalition building. So for instance, the fantastic work at the Beyond Oil and Gas Alliance, for instance, could also distribute its entities into different instruments, meaning that you would have a coalition for nations interested in divestment, coalitions for those carbon taxes and so on. So that nations can then choose to pursue coalition activities in relation to the characteristics that align with the characteristics of their nation. So for instance, OPEC countries probably would not join a coalition built on moratoria, but perhaps they'd be more interested in divestments. And last, this is a matter that is close to my heart. Reverseability of fossil fuel cuts is a big problem. One of the instruments I highlighted is moratoria. Moratoria, by definition, is temporary. Most of the cases I've seen in the database, once a moratoria expires, the government just lets it expire and drilling, if it's economically feasible, resumes, which defeats the whole purpose of declining emissions if it's just a temporary solution. We note, I note particularly Australia's carbon tax, which was introduced in 2012, which is the supply side carbon tax, which was axed just two years later under the Tony Abbott government. So I'm hoping that international institutions or international initiatives to keep fossil fuels on the ground, such as the BOGA or the Fossil Fuel At Your Proliferation Treaty, will also take this aspect of fossil fuel policy into their work, and that we can actually keep fossil fuels in the ground today and tomorrow rather than just today. Thank you very much. Before presenting my research, I just want to take a moment to reflect on the rich and deep history of this hallowed ground we're on. Long before this was a place of higher learning, one of the oldest institutions in the world, it was actually just a humble place where farmers came with their oxen and other animals to cross the river. And that's where the name comes from, oxford, which comes from the old English, oxnaforda. So just, yeah, now that Salinas is out of the way, I can show you my research. So again, just I'm doing climate tests. So we're going to be comparing new fossil fuel extraction or related infrastructure with climate objectives, national or global. So I have a suite of tests based on existing research but synthesizing it. First is global and national infrastructure committed emissions from existing infrastructure, then cost optimal energy transitions at a global scale, then being scaled down to my test subject, which is going to be Canadian gas extraction, then national energy transitions that follow the global average under 1.5. Sorry, that was fair share under 1.5. The fourth is the global average under 1.5. The fifth are using national decarbonization models, and the sixth is for federal emission reduction targets. So I'll go through these individually. Climate test one, again, we're going to look at carbon budgets for 1.5 and 1.75 degrees and compare this to fossil fuel infrastructure at a global and national scale. So that's the test subject and benchmark is the carbon budgets. Here are the latest carbon budgets for 1.5 and 1.75. So as a proxy for well below two degrees. And you can see that committed emissions are already already exceed 1.75 except at a low probability at a 50% probability threshold of exceedance and definitely exceed 1.5. Then at the national scale, we can look at committed emissions for Canada and compare them to various decarbonization pathways and the cumulative emissions that would occur under these pathways. So I use SERP or SERF, which is the Climate Equity Reference Framework as a fair share pathway. Then the EMRG is from Mark Jacquard's lab. It's like a political scene is like a politically pragmatic and cost optimal model with very little carbon tax, then deep decarbonization pathways project by Battai and colleagues, which is heavier on carbon tax. Both are low oil price futures. Then there's Environment Canada and Climate Change, which is very proprietary and much more opaque. But you can try to ask them questions about their assumptions. Then there is SR15, we're still using this, the LED model, low energy demand model. Then there are the federal targets and some scenarios that I've extrapolated. So here we have Canada's committed emissions, and you can see that they already exceed all the cumulative emissions from these various scenarios, which means that even if we didn't build new infrastructure in Canada, we'd have to decommission it early in order to meet any of these fairly not very ambitious for the most part scenarios. So for the global, we also could look at supply sides. So no new analysis here, but it just helpful to frame it in terms of climate tests for policy use, strategic assessments, communications reasons. So here's the recent study from Kelly Trout and Greg Muddit, which shows that we're all the existing reserves, if extracted and burnt, would exceed a 1.5 budget and maybe even exceed a two degree budget. So another something you could also call a climate test. And of course, the well known result from the NCE model that we don't need new oil and gas fields if global demand aligns with 1.5 degrees, another version of a climate test. So for the remaining climate tests, which are specifically for Canadian gas extraction, again, I'm just kind of trying to formalize climate tests here. And now I'm going to be applying them to a specific sector. We're going to look at development in Canadian gas extraction for certain projects and for extraction on the industry for the industry as a whole. Then we're going to look at the benchmarks for different for the remaining tests, then we're going to compare them. And that's what makes the test. So first, we need the forecasted extraction. So here I'm using the Canada Energy Regulator forecast. So it's pretty industry favorable, but it's still from the natural resource department of the government. And we're going to be using their evolving scenario, which does see some sort of peak and decline. But still a lot of gas, of course. So for the first one, we're going to compare this to the Wellesby results, Wellesby and colleagues from nature. And you can see here that under a cost-optimal energy transition, where everybody around the world is just using stuff along a least cost curve, Canada has very little left. And we'd already be over extracting this year if we continued along this evolving scenario. And now for the remaining tests, I'm going to be comparing emissions upstream emissions to upstream emissions. So I use life cycle analysis to estimate the emissions from the evolving scenario. And you can see here, and panel sees the cumulative emissions from these. And the fugitive emissions are what cause this sensitivity. And probably people here are familiar with there's a large discrepancy between what the government uses, what industry consultant use, and what is known in the scientific literature to be much, much better information. So 2.7% the orange is close to is the best estimate from Alvarez and colleagues from the science from a few years ago. That's for the North American weighted average 3.7 is for unconventional alone. And so Canada's production is trending towards fully fracked gas. So it's somewhere, it's going to be somewhere between 2.7 and 3.7, very comp with with high confidence. And now we're going to compare this to the decarbonization scenarios now scale down to the gas sector. And again, the cumulative emissions. So finally, we can put this all together and finish the climate tests. So a fair share scenario has virtually no gas extraction. The global average has all right, well, so this is again, it's not necessarily an indication of of the gas sector itself. But the the rate of decarbonization of the country implied by implied by leaving enough room for development for other countries, according to this, according to these parameters, this model from this project suggests that we need to we need to reduce emissions under to be aligned in a fair way for 1.5 extremely quickly. I think it was zero by I forget what it was, but you can maybe I'll go back later. But you know, but definitely zero within within years, not decades. And that implies domestic domestic reductions, tantamount to that. So we're talking about phasing out all domestic emissions within a matter of years. Is that is that all right? Just to leave more room for other countries. And then the global average, which is not necessarily fair, but it still aligns with the 1.5 degree trajectory allows for some more. And then these remaining ones aren't aligned with with the Paris goals, they're just what we think is economically feasible without and politically viable, tenable, but with without any consideration to global climate goals. So we've got the Canadian decarbonization pathways, which are more aligned with a two to three degree world in an in an inequitable way. And the federal targets, these are the old ones, but this if these don't pass the test, then more stringent targets also don't pass the test. And you can see here the discrepancy we've got, you know, so at least twice as much or three times as much three times as much emissions being planned to you know, under under this forecast under this evolving scenario, compared to even modest modest decarbonization pathways. So we're just saying this is heavily gases heavily constrained, even if we're not going to do enough to meet the Paris agreement, but still decarbonize in a in our at our own pace. So again, just to sum up, you know, the the results of these tests clearly indicate that there's just no room for new gas in Canada under any of these under any of these perspectives. But not everyone really gets this yet. So we're still talking about exporting energy to Europe. You know, this is a recent a recent meeting between the German Chancellor of Schultz and our Prime Minister, Justin Trudeau. And there's still a lot of discussion of people pushing really hard for Canadian energy interests. So we need to move beyond this. And we know that forecasted gas extraction extraction any way you slice it is clearly incompatible with our national goals, which are paltry still better, but not but not aligned with Paris and international climate goals. So New Island G is definitely not on the table if we're serious about climate. And we need like it's just just clear that we need a rapid phase down of gas production and a plan to do this, because any any decarbonization scenario highly constrains gas. And and so if we were actually serious about even doing the bare minimum to meet Paris, we need to wind down gas in a matter of years. And thank you for your time. I was a bit confused about the supply versus demand side ones, which makes a big difference in Canada where the extraction emissions are a tiny fraction of the the total carbon footprint. And in the last the last figure you had before the fail, fail, fail one, the the y-axis says upstream GSU emissions. That's right. Then the boxes on the right are saying production. This is either only I'm comparing upstream to upstream. If you wanted to compare full life cycle emissions with downstream emissions, you could. But I'm looking at the constraints of domestic emissions on this. So yeah, so so it's just comparing like to like. But but but it wouldn't you know, you could you could you could look at this from the same way. But I think you would come to a similar conclusion. You could look at it with with total upstream and downstream emissions. But when when I thought it made more sense when you're looking at decarbonization models, which are only for domestic emissions to look at emissions occurring in Canada alone and compare that to yeah to that. But but if you wanted to look at the global impact of gas, you'd have to you'd have to look at downstream emissions as well. Thanks. Any more clarification questions? If not, we can move to open discussion. We have 20 minutes. Thanks for this. I want to make sure I heard correctly. So this is for this is not a specific LNG project. This is all LNG sort of like planned as of now. Yeah, I actually had when I did this GNL was still under consideration. Now it's not but now Cedar is under consideration similar a bit smaller. But yeah, but but roughly the you can see that the projects are a bit more we've used it full capacity than the production forecasts. So even that suggests that there's too much production being proposed for LNG terminals. But yeah, got it. Thanks. Thank you. Thank you. Hi, I'm Natalie Jones from ISD and my question is for Nicholas and my words are very slow to come out today. So you have to bear with me. I was interested in your statement about investment and how this wasn't correlated. I was on the litigation panel this morning where we heard from a speaker about the effects of investment treaties and and how companies have been taking cases under these, which might be really bad for supply side policies. And I was wondering if you looked at membership of treaties at all as part of your analysis, or if you have any thoughts on on on how this might have a bearing on your work. Thanks. Any other questions? Someone at the back. Yeah, Brendan Devlin from the European Commission. European Commission has been given the mandate to create a moratorium on oil, gas and coal production in the Arctic. And we're wondering how to do it. And I was wondering if any of you had any advice for us. I'm writing my thesis right now about moratorium, moratorium bans in the Arctic. Thank you. Well, you've come to the right session and he is looking for an internship. I wish it was so easy. And from that, I wanted to ask another question. But how would you go about on your collective understanding of what you just told us, all of you? How would you go about creating a moratorium? I can see the end point just to to anticipate that question about how do you stop a moratorium being a short period of time? There we have the mandate to create a legal framework, which has high downside costs if you break it. So that's the end point. But how do you get the discussion going to get there? We've had the intermediate problem of Russia, which has taken out a lot of of Arctic production as a result of the war. But when we come back to it, how do you create a negotiating strategy to get to a moratorium that will that will end up with a legal regime, which is effective? But how do you get there first step? And then the question is that the second question I had is to all of the panelists, you all take the net zero, the IEA net zero scenario. And one of you mentioned it, that there will be a concentration of suppliers in the end game for hydrocarbons on low cost producers. Why do you assume that it will be low cost producers? For example, the European Commission has adopted a position of in our methane strategy saying it will be the lowest greenhouse gas suppliers greenhouse gas emissions in the supply chain, not in the upstream in the supply chain, just to go to your question on what is in the upstream. So we are creating indexes indices of emissions from Canada, United States, North Africa and such like that's one option. And we're also looking at other options, which is to privilege countries which have lower than average national income on a per capita basis. So why do we assume that it will be the low cost producers? Indeed, we are taking the opposite assumption that it will not be the low cost producers who are the last suppliers on the block. It'll be the countries with the lowest emissions times the countries with the greatest need to have income. And what would that do to your models and outcomes? Thank you. I think someone did signal that they wanted to ask the question, but if it's okay, I'm going to let the panelists answer first and we'll add more questions. So Nicholas, did you want to answer Natalie's question first and then we can open it up? Right, right. So for the FDI, I also do want to stress this could also be a methodological limitation of the study. For instance, we were looking at FDI in general. We were trying to target fossil fuel FDI, like fossil fuel specific FDI. However, there was lack of data for the most part. So I can't argue that this specific targeting of investments won't have an impact or won't increase the likelihood of increasing the likelihood of having an active measure. Quite enough though. However, yeah, actually that was it. I think it really could be a methodological impact. I think we were discussing in the discussion section, we were also discussing the possibility of perhaps tying funds for green energy development and so on in like coalition agreements and so on with countries that are looking to transition and don't have the means. So in this case, we do implicitly accept that there is a likelihood. However, the data that we have used does not prove this yet, but this just means we need further research. So it's a bit of a cop out. I'm sorry. Okay, and then the second question. I think I'll let the other panelists answer first. How to impose moratoria on optic bands, kind of touching on the political feasibility and durability dimension if you can as well. Anyone like to go first? You know, I would rather answer the other question because I'm not a policy expert, but for the modeling, I think it's a very interesting point. So if anyone has. I could give a pretty, pretty cliche answer, I guess. It's a hard question. Nobody has a good answer, but I think it has less to do with policy and more to do with people without a lot of public support and massive campaigns that are pushing for it. It seems to be the only way to get anything done that takes the possibility for big profits off the table. So maybe there could be some big flashy, very popular permanent moratorium or whatever you call it. What is a permanent moratorium? I'm sure there's a term for it. A ban? A ban Arctic oil. I know it's not the same, obviously, as they do, but what from county Washington state did successfully manage to go from repeated moratoriums on expansion of the VP facility there to a ban on expansion of it? And I could connect with people who understand the political dynamics that led to that. So the very short answer is that it's very complicated to model fossil fuel markets. And so using a least cost approach is the easiest and in some cases, depending on the model structure, only way to get away with it. But it's true that some markets, especially maybe for example, for coal, it's a closer representation of reality, a competitive market clearly for oil and also for gas because of the rigidity of the infrastructure, is not necessarily a great assumption. So that's kind of a standard assumption, but it's not a great assumption. It's a necessity that we need to use. Some models are more equipped than others to represent the nuances, but then you have trade-offs. Because if you introduce like rigidities and you try to assess real-world complexities, then you increase the answer something when you try to project these things over 50 years, 100 years. That being said, I think the carbon intensity, it's not easy to represent. They also play chain, but I think it's a great point. So thanks for, I think it could be done and I think it would be a valuable improvement of the models. Can I just ask maybe and ask another question? Someone maybe not on the panel, but maybe Professor Eakins. How does the times model actually deal with carbon intensity? I remember looking this up a while ago, but do these models of cost optimal transitions actually account for this by costing and get in? Because I know that the carbon intensity of the fuels along the supply curve are accounted for in some way. If the data is there, I don't suppose we know what carbon intensity of each field is. Steve or one of the guys who actually does the modeling would be a better person to ask if they're here. But perhaps I could just answer the question from the European Commission there, which is that we did a model run of prioritizing production from low-income countries, which I think is what you were talking about. And it's fine if people price discriminate. In other words, if the European Commission has lots of money and therefore can pay a higher price to those countries, that's fine. But if you do it on a global price basis, in other words, global or market price that favours these low-income producers, these low-income producers, you really screw with low-income countries that are not producers because they're paying a much higher global price. And so that came out very strongly and was quite counterintuitive. I was expecting this to be a scenario that made me feel good. But actually it didn't, because we've never seen a whole bunch of low-income countries, but we screwed a whole bunch even more low-income countries who are demanding consumers of fossil fuels because they won't pay any more. So I think one always has to do that, so. And, yeah, Austin, can you let this stand a minute? Yes, it's been patiently... Yes. Mark, Empanali, Carpentry. Just want to build on the point Professor Ekins has made around price and the importance of price signals. So when we look at supply, obviously low-cost producers are the ones that tend to attract the capital based on price assumptions and demand assumptions. And of the 3,000 gigatons of CO2 that we know about in reserves, 3,000 or 4,000, a thousand gigatons of that is owned by publicly-traded companies and about 40% of what's in the market in any one time comes from listed companies as opposed to state-owned enterprises. So what we need to look very closely at what happens there in my view. So in the work of Carbontracker and the work we're doing with Global Registry, but mostly in Carbontracker, we build cost curves of all coal and gas production together, and that allows you to intersect your supply and demand curves with the SDS scenario, the IA 1.5, the SEP scenarios, and then everything above the highest end of the cost curve. And this all published on our website, you can see all the production that gets killed, and obviously because of oil sands being so expensive, Canada is right at the top end of the cost curve. So we've got huge amounts of production that's planned that will not make it to market. On the rather different point about moratoriums, where we want the moratoriums to be. So in the last 10 years, there's been two and a half thousand coal oil and gas IPOs and second replacements on the New York Stock Exchange and London Stock Exchange, on the European Stock Exchange, that's raised about 700 billion. That's just the equity. If you put that on top, that's another trillion, trillion and a half of that on top of the equity. We think there should be a moratorium on coal oil and gas IPOs. If we can't burn what's already in the public markets, why is the European Commission allowing all these companies to sell through IPOs on bosses regulated by the Commission, they should put a moratorium there. It's insane that all these IPOs are coming to market. At the COP, there was a coal IPO at the COP, during the COP, it was six times oversubscribed. This is where the, if I can put it like this, the Commission should be focusing its attention. Happy to discuss further. Thank you. And Nicholas, you wanted to add something. I just want to talk about my thesis real quick, because it is topical and I spent six months on it. Just regarding the current state of Bantz, unless I'm bored, would you do you want to hear about it? Because it's also possible you're overwhelmed with information. But I know, so there are eight Arctic nations that exist today and five of them have been historical producers and Canada and the United States Bant production are offshore drilling. This is specific to offshore drilling in 2016 and Greenland had a moratorium in 2013, which became a ban in 2021. I'm looking at the resources and the strategies used by anti-extractivists, so those who are opposed to offshore drilling, to enact these bans. And I have come to the conclusion somewhat regrettably that the actions of civil society organizations were actually quite insignificant relative to the price of oil. And that one of the reasons that these moratorium bans could pass is because Arctic exploration companies were no longer interested. So you highlight Norway and Russia, which are the exceptions in this story. They're actually investing even more. Russia is multiplying its investments quite drastically. So I'm giving you a complete answer and saying the only reason these countries could do it is because there was no opposition. There was, by the end, the moratorium bans were just, I'm not going to use the word greenwashing, but that could be strongly implied. Just this notion that it was symbolic politics. And I think the European Union will have to struggle. Well, first of all, you have to work with unclothes, which is a bit of a disaster. I know Greenpeace was pushing, for instance, for an Arctic refuge across the entire area. So perhaps the European Union could put its support behind such an idea. But this will require some pretty dramatic diplomatic disputes with oil producing nations, of which there are only two left. So there is power in numbers. But I do wish you the absolute best of luck in this mission. All right. I think we have time for one or two quick questions. Austin, someone back. Hi, Nadra Ujir, Commonwealth Secretariat. I guess I've got two questions, if you permit me. One is just around better understanding. I think it was the seven approaches or measures you mentioned, Nicholas. One of them was divestments. And I just wanted to better understand in terms of the actual fossil fuel cut in terms of, and just maybe it's my understanding, but divestments really would result in changing of ownership or hands of an asset from one to the other, not necessarily keeping it those resources in the ground per se, so not necessarily resulting in any reduction in production as a result of divestments. So just wanted to better understand how that would work in terms of your methodology and sort of relating as well to the question around scenarios and the challenge around introducing some element of equity into the modeling. And I just wondered in terms of the approaches, we just had a discussion around income being one of the criteria in terms of cumulative contribution from fossil fuel production as a potential measure in terms of, again, looking at that as generally speaking, 75% of our carbon budgets has already been consumed. So looking at that as one of the inputs or factors in the modeling exercises. Thank you. Yes, mine is pretty brief. The divestment one, thank you for pointing that out, is a bit more controversial. I've had quite a few discussions about it. The idea is through divestment you would use capital flow to energy projects, to fossil fuel extraction projects. It's not necessarily as convincing to me than the argument that it's more normative exercise and the idea of divestment is one that portrays, that symbolizes the rejection of organizations, whether private or public, of continued extraction. So it's more of a perhaps a social exercise in fossil fuel cuts rather than a physical exercise. However, well, I will need to research this further. I do believe that reduced access to capital should also, in the long term, reduce the ability to extract. But I will have to read up on this. You don't seem convinced and this means you must have some resources which I would love to hear about. And last question, Daniel, did you want to touch a bit on the equity and you mentioned... Sorry, what was the question? Was that for Daniel or any of the... How do you incorporate equity principles in climate testing, translating global emissions pathways to a specific country or... Yeah, that's a tough question that I've just started to really work on. So this was the analysis that I presented was more from a demand side perspective, looking at what a fair share implies for domestic emissions. But you can also of course look at it, well, and just before I go, if you look at it like that, then you can see that doing our fair share means decarbonizing as fast as possible, which means that emissions have to come down in all sectors and decarbonization models, which do things, if you want to base it on, let's say, a conventional economic perspective, there's no way to bring down emissions very rapidly without highly constraining oil and gas extraction because they're really... It's the low hanging fruit. It's a lot of emissions for very little value in the Canadian economy especially. For example, I think tar sands are about 3% of GDP, oil and gas is about seven or eight in total. So as far as getting back for your buck, you have to constrain oil and gas production, it's just no other way. So very, very rapid reductions which are needed to leave more space for other countries to decarbonize at a relatively slower rate, let's say slower than the global average, requires us to have faster phase outs of oil and gas extraction. You can also look at it from a supply side way and there's a recent report. There's a paper by Steve Pye and colleagues which did another model that I'd like to incorporate into climate tests. There's also the recent report by... Forget the first author but he worked with Kevin Anderson at the Tyndall Center which had kind of a less mathy but more heuristic way of deciding what rates would be equitable for different countries to phase down their production and then did that in a global analysis to show what countries would get to to extract what in a more equitable way. So these are kind of how people are reckoning with this question. If that helps, if that was, I don't know if that answered your question. Yeah. Any burning loss remarks from our panelists? Yeah, just... So in our specific scenarios we opted for a more, I would say, realist perspective so we didn't look into equity. There is a lot of literature on equity. Barden sharing usually is on emissions and it's not on production but I just maybe wanted to point out something that it's pretty obvious I think but so it's true that of course the large part of the problem of emission is due to fossil fuel burning and so if you don't extract fossil fuels you largely solve the climate problem but the climate problem is due to emissions at the end of the day and so that allocating using the supply side action as a climate policy tool still require in a fair way or in a non-fair way or in a cost efficient way still require a relevant level of international cooperation meaning that if Canada reduces production and at the same time does not put together the same a similar amount of the monsite policy to reduce also consumption at the same time what happens is simply you remove yourself from the fossil fuel market but without having any potentially good outcome for internationally for the climate which again it's not rocket science but I think it's important to state that the production reduction is meaningful only if it's coupled with the monsite action so so consumption reduction okay um we only have a 15 minute break now so I'm going to wrap this up thank you for your active participation and let's thank our panelists again