 And now, I will open the floor for your question and remark. So please. Yes, James. Hi, I'm James Fuller from You and Your White. I'm obviously a CG model myself, so I'm looking to go. But let me be that paid devil's advocate. What I see from Chaz's presentation is that he's laid down the gauntlet for the last two presenters there. So Chaz has been saying that, actually, you can't just pick one climate change scenario. You need to be looking at all 56. And even the 56 may not be good enough. It may not be a true reflection of the distribution. So maybe you could reflect a little bit on the motivations. You can mention the world bank. And why you picked that particular scenario and what concerns you have. I worry a little bit. We know once the climate change community there's this worry about false precision, that you do lots of downscaling, but on one scenario. And you miss the fact that it's the variation between the scenarios where the true story is. I worry that if we go down and I'm doing it myself, we go down to the sub-national regions. That's false precision from us economists, particularly if we're not taking in the proper climate change, the full range of the climate change scenarios. And my second issue is this. I saw both of you were talking a lot about variation and the impact as the crop yields had variation in this now, but you're basing this on two scenarios. Where is that variation coming from? Is it year on year variation from the climate projections? In which case I worry that that's weather. That's not climate change. I mean, a lot of the fluctuation year on year that you see in the climate change models out in 2050. That's just an assumption that the future variation repeats itself. So what you're looking at when you show your box and whiskered flux is not climate change variation. That you get from variations across the climate change scenarios, it's weather variations. And that's historical. That's not climate change. I'm James Bresinger from Ifpre. Thanks first of all for these very good presentations. I'm now referring to the two country case studies using a CGE model as well. And I noticed that post studies focus very much on the local effects on climate change, like looking at those countries specifically while not looking very much at the rest of the world. Now, some of our projections, actually, most of the projections from Ifpre and others show that climate change obviously happens worldwide and it affects other countries as well. So the projections show that global food prices are very likely to go up because all countries are affected through lower rain, higher temperature, lower yield. There's only so much land to expand to. So I was wondering, especially in Middle Eastern countries that are net food importing countries, Turkey maybe not to that extent, are we not missing part of the story if we are not also taking the global effects into account both in macroeconomic terms but also on the household level? Other questions? Please. I'm sorry, but I'm not quite sure about the role of development here in the Middle Eastern or in affairs in Finland. Thank you for the presentation. I have one question for Charles regarding these three things, you were talking about the natural runoff stream and not about the actual. So do you know if anybody is doing projections on the actual stream flow? So there's not the population growth and the use of water, increasing use of water by agriculture. So how do you get the effect of both natural and actual runoff of streaming stream for the food security and bringing in the people's life? Other questions? Please. A question for Charles. On your request, driving the range and variation in your GCN scenarios, was it 50 different models or was it 10 models with different assumptions about drivers and variation or parameters in a GCN model? What was driving that range? Do they all have the same assumptions about growth and variation or different assumptions? More questions? And now it's time for the presentation. Can I start with that? OK, James Thomas, thank you very much for Thomas. Yeah, we really have a big problem about understanding or relying on the climate models. There are fixed six of them, yeah. And unfortunately, when you are trying to do some regional study for Turkey, most of the results are mostly not available. You have to find them, feed them into the model. So my belief is that or my hope is that as more data become available for Turkey or for other regions, for the globe, we will be able to incorporate all these information into our models. But at the current stage, first, I don't have the capacity to process all the information available. Second, all the information is not available anyway. So in time, I hope to enlarge my database and incorporate all these scenarios. For this differentiating between weather and climate change, you are right, we should do. And that's why I reported periods, not each year. I showed each year, but what I was focusing on was what happens in 20 or 30 year periods. So that's how I tried to overcome that thing. And for local effects, and you are absolutely right, climate is changing all over the world, and the world prices will significantly change. But with country CG models, that's all what you can do is introduce some price shots, right? World prices should change. At least, world prices should change. For that purpose, Ismail and Vali, we are engaged in a study with GTAP. That will take into account the effects on the whole globe. And we think to feed the results of GTAP model in terms of price change into the national CG models in order to take this face into account. I think Hassan hasn't left me much to say about it. But I think I doubled down what he said on the criticism about the climate data, the yield impacts. In the case of Morocco, basically, initially, we started actually with the Cliron yield chucks that Chaz has supplied us. And we wanted to make a run with that. The problem, at least, from our side, what we have with those from, we had some issues with particular crops, the way they are modeled. We discussed this, which has some key crops, for example, in Morocco that were not basically modeled the correct way, so to speak, by Cliron. And additionally, when we got the data and we investigated the impacts, they were, basically, how can I say, so against conventional wisdom that says that Morocco would be losing drastically, whether or not that's right or wrong, we decided to go with the World Bank data because we thought that was basically the safest route, the less controversial route to go to. I mean, we run some simulation with that data, but what we ended up having is just the impact was just minimal. What we had, basically, is just a continuation of what we've seen in Morocco happening for years now. That is, yes, agriculture in Morocco is susceptible to climate condition. It's highly volatile because there is not a lot of resilience built in in the sector. So that, basically, motivated, basically, our move away from Cliron and opting with the World Bank data. Serial scenarios. Yeah. Yeah, I mean, the scenarios, the World Bank study, as James mentioned, compared to Cliron, they used a limited pool of scenarios, of course. They used only two global scenarios. They used only one GCM, which is they had them GCM, which is, basically, as you said, you're relying, basically, on one GCM to capture, basically, to try to capture the climate uncertainty, which is counterintuitive. At least, if you want to capture the answer, you have to have multiple, basically, projection in the futures that are captured by different GCMs. So on that point, I mean, basically, there is no arguing there. I think on the data issue, it's the same thing. When we started working on this, as Hassan has mentioned, there's a lack of available data. There's only so much you can do. I mean, you have, basically, you want to do the best you can. You want to model everything. You want to make as close as possible. But your data constraint, basically, just cannot allow you to go that far. And we've tried to manage, somehow, what we have to make it work. I think that's it. Thank you, Sush. Come, please. All right, so I first want to just address real quickly what Sush was talking about. There's one distinction. The model that I was presenting was I run, but which isn't an I was supposed to call it. Like, yeah. So it's easy because they have the same thing. I do the same thing. So they're different models. And I apologize. I wish I could have some more time modeling using fly crop. I think we could have gotten better results. At the time, I didn't have a lot of time. And I think that was all what you said. So the question about sort of the devalues is actual. Yeah, that's it. All right, so I kind of hesitated to even bring this up. So the point that I was making was that we're introducing uncertainty, even with the historical data. But the Clyron model is part of, of course, the whole suite of models. And it's supposed to model natural stream flow. And then we have another model that models the irrigation. And actually, Alyssa is an expert in that area. But the irrigation and how much they're releasing from the dams, et cetera. So that's kind of a separate aspect. And it should have been more clear. And then the question about the distribution. So those are each point. 22 for the A1B, 17 for A2, and 17 for B1. And those are the only points there from the English distribution. So there's no, it's not historical, inter-annual variability. It's a mean taken over all of those years. So it's really just the distribution of each GCN specifically. And I don't know exactly what's causing that spread. I think it's very detailed. It's how the GCN's model climates. And there could be lots of things that cause that spread. But they're running the exact same scenarios as far as emissions. Theoretically, they're running the same scenario. Because the A1B is defined in a certain way. And they interpret that in their model. So it is surprising that range is there, I think. Thank you, Charles. Other questions? If I may have just one question, because in the two presentations of this my understanding, we talk about this regional disparities in terms of the impact of climate change. And I wonder if there is something to learn about this regional disparity. Is there any adaptation process that can really learn at the local level? Or does the reflect on somewhat the way the farmers or the actors at the regional library adapt to climate change? It's related just to the characteristics in terms of geography and the culture. Well, for Turkey, I just mentioned that. For example, the eastern parts of the country, the households in the eastern part of the country is not that much affected. The main reason for that, I couldn't go into the details. But the main reason for that was that they are already making very fat agriculture. And they are not very dependent on water anyway. So when the end, of course, the climate change is not that significant in that region. So what I observe from the results from many scenarios is that if a region is not very dependent on water or if a region has a good water infrastructure because the regions with good water infrastructure are not affected too much from the climate change shocks too. So they are not affected too much. But that doesn't mean that if the farmers can adapt by themselves, I'm afraid the CG model cannot answer this question easily. Because I have households, they are affected. If they are not very dependent on factoring comes from irrigated water, they are not affected too much. Or from natural irrigated land, they are not affected too much. That's all I can see from the model. But by common sense or by conventional wisdom, I can say that, OK, if a region has a better infrastructure in irrigation or if a region is not much dependent on water, they are not affected as a region, not as a household. They are not affected too much. So in order to answer fully your question, I mean, we need to go deeper in the model, introduce some kind of adaptation strategies for the households. I mean, in terms of Morocco, it's pretty much, I would say it's the same thing as Hassan has said. We cannot infer, basically, how if your question is specifically asking about how farmers adapt and what's available for them, basically, to become more resilient to climate change, relying on the CG model to answer that kind of question is, at least, the way we've taken it, it's clearly, that's not, it wouldn't shed too much light on that question. But whether or not farmers are adapting, I think historically, I mean, farmers are not done. I mean, they are pretty, how can I say, they notice changes, they react to changes, they adapt to the best of their abilities. But, of course, there are conditions on the grounds, sometimes primarily driven by policy choices sometimes, that distort sometimes incentives for farmers from adopting more effective choices, basically, for adaptation. And I think that's basically what most of the time, it's really the CG model tries basically to capture more of those kind of, those broad policies than going and boiling down from a bottom up structure and see, build up, kind of, analyzing the adaptation strategy of farmers. Thank you, Hassan. In fact, my idea was about the fact that if regional mobility can really, in the future, because of climate change, we can really see more regional mobility or displacement of people, because some regions will be more affected. And if it is really a physical constraint, that means people or farmers are forced, indeed, to feel if they cannot really adapt. Thank you very much. We should really join me to congratulate this excellent presentation. Thank you. Thank you.