 I am from Cape Town, South Africa, and I will be presenting preliminary results of my master's research, which is looking at understanding the link between ENSO and drought over Southern Africa. So why drought? So some of you might not know, but for the last two years, Southern Africa has actually been experiencing one of the worst droughts experienced in over 35 years. So the map over there actually shows the percentage rainfall from our rainy season last year. And as you can see, a majority of the region received less than 65% of its normal rainfall. And several of the countries have been declared disaster zones. And then looking at the bottom middle image, that is actually Cape Town's water supply currently, whether before and after, we're sitting at about 10% of capacity in our dams, and only 2% of that is actually usable. So we have an estimated 60 days left of water. So we're on severe water restrictions at the moment. Other impacts of drought, it impacts agriculture, and this results in economic losses as well as food shortages. And this is particularly prominent in areas such as Southern Africa, which rely heavily on rain-fed agriculture, as well as subsistence farming. And like I've mentioned with our dams, so long-term drought results in water scarcity, which results in people seeking unsafe water sources, and that leads to the spread of waterborne diseases. So to put this into context a bit more, Southern Africa has a highly variable rainfall, ranging from the Namib Desert on the west coast to the more tropical kind of Mozambique side. So the north-eastern region receives the highest rainfall, whilst the southwestern region receives the lowest rainfall amount. And the majority of the region receives summer rainfall, which is our December, January, February period. The synoptic systems, which bring about rainfall, include the shifting of the ITCZ. There's a tropical temperate trough, which lies across the continent. Cut-off lows and mid-latitude cyclones affect the southwestern tip during the winter months. And then there's this Angolan-Kalahari low-pressure belt, which lies across the west coast, bringing the little bit of rainfall, which does fall in that region. So the influence of El Nino Southern Oscillation. So typically, an El Nino event will result in drought conditions. This is due to the shifting of the tropical temperate trough east of the continent, and a descending limb falling over the continent, creating a high pressure. However, there is a northward shift of the mid-latitude cyclone track. However, the storms are generally weaker, and all of that leads to generally drier conditions. On the other hand, El Nino results in wetter conditions, and this is due to the shifting of the tropical temperate trough over the continent again, and there being a convection in that region, and the mid-latitude cyclone track shifts southward again. But the storms are stronger. However, there is high inter-El Nino variability, which has already been discussed. So not all El Nino events result in drought and not all El Nino events result in wet conditions. So for my research, I am basing it off these sea surface temperature anomalies identified by Johnson through self-organizing maps. So Johnson focused on the Pacific Ocean and created these eight ENSO patterns. And then Hull expanded that to include all three tropical oceans. And that creates these eight patterns of ENSO. Hull then went on to explain the precipitation anomalies associated with each of the ENSO patterns over southern Africa. However, they only considered precipitation, and it has been shown that temperature also has a strong influence. And that comes into the standardized precipitation evapotranspiration index, SBEI. So just to clarify, a negative SBEI is drought and positive is wetter conditions. So in the more tropical regions, such as Tanzania, it is seen that precipitation actually has the higher correlation to the ENSO index. But when you get to the more mid-latitude regions, that temperature actually seems to have the higher correlation to the ENSO index. And luckily, SBEI is kind of the best of both. And then the other issue with that study is that they were using observed data and that only has few observations per pattern, so the result isn't particularly robust. So the aim of my research is to investigate the link between the different ENSO patterns and drought over southern Africa using SBEI. So I'm using SBEI to enable me to do multiple ensemble simulations to create a more robust result. So the steps are to evaluate speedy. So I'm using the higher resolution of speedy, the T63. So the first step is to evaluate it and then to impose the SST patterns onto the model to see the influence of ENSO on drought, on the SBEI. Just a question, this version that you're using has the vegetation turned on? Yes, I'm sure, yeah. So for this study, I'm focusing mostly on the evaluation of speedy. So the setup, so like I mentioned, I'm using the T63 resolution of speedy. And then I'm comparing that to the crew observed data, monthly data. So the first step was to re-grid the speedy data and then I evaluated several climate variables for the DJF period from 1970 to 2010. And yeah, I was looking at temperature and precipitation, moisture balance. And then I looked at the capability of speedy to simulate the ENSO SBEI patterns. So going on to temperature, so I've got the crew observed data and then the speedy simulation and then the bias and then with the root mean square error and then temporal correlation and then in brackets is the spatial correlation. So generally speedy seems to have a warm bias over the majority of Southern Africa and particularly pointing out the region along the west coast, which seems to have an influence on precipitation, which I'll get to. So then looking at the moisture balance, unfortunately precipitation isn't too well simulated at this stage. There seems to be a high potential evapotranspiration, which speedy seems to be simulating. And again, that's particularly going on that west coast region, particularly over Angola in the northern section. And this is probably resulting in the low moisture levels, which is the lack of rainfall in that region. But yeah. So the evapotranspiration isn't too badly simulated except for that one region. So that is the moisture balance. Then looking at SBEI, so speedy does seem to, so this is for the DJF period, so speedy does seem to get the pattern of SBEI, just not the range, it's not quite getting the extreme highs or lows. And then when you look at the Taylor diagram, you can see that it really seems to be precipitation, which is pulling the SBEI value down. Then looking at the capability of speedy to simulate the ENSO patterns, so for this I chose, so these are the dates which were provided by Hool of each of the patterns. And then I chose two years from each pattern except for the first one because my simulation just didn't go that far back. And then I created the composites of the SBEI values. So again, it's not too well simulated and this is probably linked to the high evapotranspiration values which is impacting the precipitation. But it is pattern, let me just check. So I think the third pattern is almost like an inverse of the observed. But looking at specific regions, some regions are simulated better than others. So these are the El Niño patterns and then going on to the La Niña patterns. Again, it's getting some regions better than others. And again, the last pattern again seems to almost be an inverse of the observed. So that is the La Niña pattern. So the results aren't too great yet. So just to conclude, that speedy is generally simulating a warm bias over the majority of southern Africa and this is leading to higher evapotranspiration values and that seems to be limiting the efficiency in simulating the moisture balance and the SBEI values. So the correlations are quite weak for the SBEI of the different ENSO patterns. So I didn't actually have time to look at the geopotential height or the pressure fields yet. So that will be the next step to see what teleconnections aren't being simulated properly. And then to possibly look at ways of improving the performance of speedy before I can move on to the sensitivity tests and actually imposing the SST anomalies for my research. So that's it. Thank you. Thank you. Thank you.