 Hi, everyone. So actually, I work at the University of the Austrian Cape with Roy Martins as my supervisor and Mario Santos. And, sorry, that's a mistake. And I will first introduce you to the SKA. We will be one of the largest telescopes, one of the largest telescopes in the world, radio telescopes. And it will help us to understand a lot of the phenomena that we are wondering about, such as dark matter, dark energy, and also the formation of the Ferris C star, and also the epoch Afroanization and many, many other things, including even if they're the aliens in the universe and they're trying to communicate with us. So I'm more interested in the BAO. And the BAO is the Parionic Acoustic Oscillation. And it's measured by the angle subtended by this standard ruler. So the angle is subtended by this ruler, the function of redshift. And we map the angular diameter distance, which is the tangential component. And we could also measure the Hubble parameter by measuring the redshift interval. And actually the BAO is a very good probe for dark energy because, first of all, it's the principle of the BAO coming from the physics that we know. So it's well understood from that concept. And also it's easy to be measured at higher redshift, which makes it complementary to other two probes, such as supernovae. So in this presentation, we'll be talking about the cosmological performance of the SKA H1 Galaxy Survey in measuring the BAO. So to do this type of forecasting, we need to know about the telescopic specifications and the survey specifications and then how many galaxies your telescope will be able to see and also the bias. Then you could do the forecasting. So let's talk about the telescopic specifications. The SKA will be built in two phases, phase one and phase two. So phase one, it will be built in South Africa, which is 254 dishes. And it will be in mid-frequency. And they also will have also dense scores, also mid-frequency. And another 96 dishes will be in Australia. And they will be more like low frequency receivers. And the phase two, it's actually not a specified 100%, but the target is to increase the sensitivity 10 times of the mid- and third frequency. So the one in Australia and the one in South Africa is like a pass finder for the SKA phase two. And we need also the field of view to be 20 times of the phase one. And one of the plans that they're discussing is to make it at the spiral arm. So this is the Karoo in South Africa. And the white things mean there's no populations and the red places where the people are. And so the idea is to make the radio telescope in very empty areas, at least the core. And then this is spiral arms to guarantee more field of view. But the core will be, most of the dishes will be inside the core. And actually, those spiral arms actually goes up to Mauritius in this area. And one of them also goes up to Ethiopia and Ghana in the middle of Africa. So that was the telescope specification. Now for the survey, we need to balance between the sensitivity and the area of the survey. So it's very important. You would like to go wide, and that's great. You get a larger volume axis. That means greater access to Fourier modes. But you don't want to go too wide, because you decrease your sensitivity and you will increase the shot noise. And also the maximum range if we decrease. And one of the most important terms to define the survey specifications is the flux sensitivity, the SRMS, which is the noise associated with the flux measured by the interferometer. So for, so Merkat is a plus as a mid, we'll make the 254 dishes together. Because Merkat will be 64 dishes. And then whatever you add to it, mid to make the total of the phase one. So this is phase one in South Africa. And this is phase one in Australia. And you could see that the sensitivity is not, the noise is associated with the interferometer is high. It's 152 microgenis key and 179 for the Australian one. But the idea is that the SKA2 will go as low as five microgenis key, which is a really low number. And then also the survey area is very important. So SKA phase one can't go for at least for H1 survey. It's the, and to be able to resolve galaxies, we can't go more than 5,000. Not more, but I will explain just now. And for SKA2 is 30,000. So to know which is the best area to choose for the survey, we do the figure of merit. And figure of merit is a way to compare two experiments. Which one is better? You just associate a number, which is the figure of merit to an experiment. You determine it by doing the Fisher matrix, which is 2 by 2. And the elements of our interest here is the dark energy parameter. So it's W not WA. And then you just take the inverse and the determination and the square root, you get the figure of merit. And we did it for two sets of frequency. And you could see here that it's a peak of around 5,000 for this 800 at 1,300. And the other one is good, but it peaks up later at 2,525,000, which means it's more optimized to do it at the area of 5,000 than any other. So this is the one we will adapt with this frequency. So now we want to know how many galaxies our survey will be able to detect. So we do use a simulation, a semi-analytic simulation. And the simulation provides the H1 luminosity and the line profile and also the redshift. And from there, you could calculate the number density. But before we do that, we pick up the galaxies according to the rotation, the curve, the H1 line curve. We at least detect two points of the curve, at least two points. So we know that it's a galaxy. And that means we will ignore the phase-on galaxies because this is the show user. Because if you have two homes, that's guaranteed that it's a galaxy. But if it is one peak, it might be a galaxy and it might be just noise. So we just threw away all the other RFI or maybe it's a galaxy. But we just threw them away. And we make sure to take the galaxies at least has two points detected. And also the width of the line has to be larger than twice the assumed frequency. And the typical line usually is 10 kilometer per second. And the SK8 will easily be able to detect that number. So this is the result from the simulation. And this is, we could see, this is the noise associated with every, we did it for different type of telescopes with different noise, SRMS. And this is like for the SKA2. And this is here for the SKA phase one. And we could see that the noise actually makes it difficult to detect galaxies beyond this point and it affecting our detection. But when you go to, this is a perfect telescope with no noise. You get that this is actually what you could be able to detect if there's no noise associated with the telescope. So this is the points from the simulations. And this is actually, we did fit the points from the simulation and produced equations that now, even if you have a telescope, and this number doesn't correspond to your noise, you could actually use the equations to produce whatever, without going back to the simulation. You could actually just use the equations to produce similar curves for different SRMS. And now for the bias, we extracted the galaxies and then we sorted them according to the red shift and position. And the bias, as many of us know, is just the square root of the power spectrum, the number power spectrum divided by the matter power spectrum. But we choose to do it at k, 0.2 h divided by the megaparic. And we assume that the bias is the scale independent. And the same thing, we get also different bias values from the simulation and for SKA 1 and 2. And we also developed an equations able to produce the bias with respect to red shift for different SRMSs. And now we did the simulation, we discussed the specification for the telescopes and the survey. Then now we could do the forecasting. And we'll consider just a few SRMS values. We will consider 70 and 150 and 200 with Mikrojanski as optimistic reference and worst scenario case. And for SKA 2, it's 3 and 5 and 23 Mikrojanski. So first we start by forecasting for the BAO scale, DA and H. And we use the Fisher matrix. And the Fisher matrix is all actually the forecasting about the Fisher matrix because you put the survey specifications and the telescope specifications in your V survey. And it's already affected our number density here. And the bias will go in the power spectrum. And the two elements, the INJ for us is the BAO scale, so it's DA and H. And after you get your Fisher matrix, you just take the square root of the inverse. And the first element is DA and the second element is H, the error and H. And you could see that this is the error and H. And for SKA 1, it's very expected that it won't do as good at the SKA 2. But we just were forecasting to see how well it will do. And actually, it's not that bad because if we compare the results with BOSS, it's actually comparable how well the SKA 1 will do. And the data could be complementary. And also, the intensity mapping will be done by SKA 1 and it is more promising. For SKA 2, we could see that the error could go down less than 0.5%, which is very good. And the same for the DA. And now you could transfer the error from the DA and H to any parameters you want, and we choose those parameters. And we could see that SKA 1 with these specifications and SKA 2 will do actually the dark energy parameter. It will be less than 0.05. And this would be a magnificent number because you could actually now differentiate between models and you could actually test if the dark energy is a dark energy or it is a modified gravity need to be implemented. Thank you.