 Okay, hello. Good afternoon. I'm Pao De Soto and today I'm going to talk about different kinds of approaches to analyze the transport network in the Roman Iberian Peninsula. First of all, I want to say that these talks haven't been possible without the collaboration of Luce Prignano, Ignacio Moret and Tom Brugans, who shared with me part of their research and we developed some thoughts about what I'm going to talk about today. One important thing is that maybe I make some mistakes explaining some part of their projects, so maybe then later in the discussion slot they can explain better and much more what I'm going to show today. Okay, so what I'm going to talk about today, the idea is that we are going to explain the transport networks. What means the transport network? The idea is that I'm going to talk about physical relationships between towns and such elements and other towns and settlements. So I'm going to talk about roads or in some cases river connections, anything that will be physically detected in archaeology. In this talk I'm going to explain three different cases studies where we have used network analysis to analyze these kinds of transport networks and the idea is to explain if these kinds of analyses can help us to understand better how a Roman territory was organized and how the Romans designed their transportation networks. Very quickly, why work in the Roman Peninsula? The idea is that this geography is an almost closed, which means that we don't have to care about the networks related in the neighborhood, so they are closed. We can analyze it by itself. This size is interesting to make a big analysis, so it's big enough to do this kind of analysis, but it's not so big to make something that it was not affordable. And obviously, because we have a lot of data, we know a lot of cities, population, roads, a lot of archaeological finds, so we can use a lot of data in this analysis. If this is very quick, I'll again, the Iberian Peninsula has a very long Roman occupation, which means that there were the conquests of this territory took several centuries. There were different faces, there were moments that the Romans used wars in other territories and with other tribes. They used packs, so it's a territory that is evolving with the past of the time, so we have different Roman provinces and with several subdivisions of these provinces. Just a quick comment that to do this project, we digitized a lot of Roman roads. In total, we digitized more than 1,000 road segments, which means more than 41,000 kilometers of roads digitized. We took information from a lot of different kinds of sources, but it's very important to thank the collaboration of other archaeologists who share their digitized data directly. This is the result of the digitization of all of the Roman roads in the Iberian Peninsula. As you can see here, we know a lot of main roads, but also we have a dense network of secondary roads and all of the territory is covered, so we don't have big gaps of information. For that reason, this kind of information is good to transform that in a network. How can we use all of this data, all of these road networks to analyze the territorial organization in this area? We can do that in different ways. Today, we're going to talk about connectivity, but if you are interested in transport cost tomorrow, I will explain more things about the transport cost. The idea of the connectivity is, since we understand cities and roads as nodes and connections, we can analyze the connectivity of these kind of networks. This is the first case. This is the case that I've been working more. It's the Mercatori project. The main idea behind this project is that analyzing the connectivity, the most simple connectivity of this network, we can identify the most important or the most significant cities in this territory. The main idea behind that is that when we think in cities like London, Paris or Rome, they always have much more infrastructures than other small towns in faraway places. The most important modification in that case is that not all of the connections have the same weight. They depend on the mean of transport that they represent because it's not the same to have a major road that a secondary road or to have a river or a sea harbour. The idea is that having with this network, we can convert it to make the analysis and this is a map, the resulting map of analyzing this kind of connectivity. You can see here in deep blue the places, the territories that were much more connected and in red the territories with less connections. Then we can use these kind of maps to compare, for example, with the location and the size of the cities in the Roman times. As you can see, most of the biggest cities in Roman times were in places with more communications, with more connections. We can do the same with the cities with more political rank. In this case you can see here the red dots are the province capitals and in pink there are other capitals of secondary territories. All of them are always well located. We also can compare that with other periods like the medieval era and we can compare the change of connectivity between periods. In this case we can see here in blue the territories that won connectivity in medieval times and in red the territories that lose connectivity from Roman times to medieval. It's interesting to see which areas are changing their connectivity. The idea is that using this kind of methodology we can see some interesting patterns about the location of cities and which is more interesting is to find which big cities are not located in well-communicated places or things that happen. So in the future we will analyze why this is happening in Roman times. The second case, it has been developed by Tom Brugam, Pignano and Moret. The idea of this analysis is to find how the network design affects the centrality of settlements. One of the ideas is that obviously not all of the cities or settlements in Roman times were located in the main roads. So what happened when we tried to connect all of these secondary cities located outside the main network with the main network? The idea is that they need to build long connections to allow people to arrive to these main major roads. There was necessary to create a secondary network, important secondary network to allow all of this mobility. The idea is how this affects the values of these networks. They have done several analyses taking into account only the major or the basic network, but also including the cities located less than 50 kilometers outside the network. And then finally taking into account all of these cities in this territory. The idea is that the results that they can explain much better later is that the idea is that you can see a very few changes in these structures. They can also do the analysis of the centrality of all of these towns. What we can see when we look at the highest located cities in connectivity is that there are not much more changes between them. To conclude this part, the idea is that the centrality should be a key factor in the development of cities because all of the most important cities were located there. But there are also important parts of cities that were not located in these major communication roads. When we analyze that, the idea is that thinking in the network, when we include cities that are located less than 50 kilometers, there are not significant changes in the values of the network or in the centrality. But when we include all of these cities, even there are not great changes in centrality, they are almost the same. There are very few changes for people. People needed more time and more effort to move around the networks. The three keys is the equitable efficiency model, also developed by Prinyano Morer and Brugams. The idea of this model is that they assume that the nodes had not only information about the most close networks, but also they have information about all of the nodes of the network. With that, they can do coordinated decisions. If I don't explain it very well, then correct me. The idea is that this model was created under the network using these kind of values. To do that, obviously, they took the geographical distance, so they value that with the shorted pass. From all of the minimum values links of the links, they choose the one that has the less value. Depending on which kind of nodes they took, they made a different, the resulting network is different and the values are different. This is, for example, the empirical real network. But when they took into account all of the nodes, but only showing the resources of the main road, of the links of the main road, it appeared a structure like this. In this case, it's only taking into account all of the connected nodes of the main network. Finally, this is the resulting network using all of the nodes of the network, both the ones that are in the main road, in the main road network, and the ones that are isolated. The idea of that is that all of the isolated nodes affect the creation of this network. When we don't take into account these secondary nodes, the Roman network seems to be not so efficient. But what is interesting to see or to explore is that when we compare the networks using all of the nodes, it seems that the Roman network seems to be more efficient, or seems to be efficient, which means that probably the Romans use or had the idea of optimizing these kind of resources to build their transportation networks. Another thing interesting is that despite all of these network models are different, the value seems to be very similar. So it's interesting. I'm to conclude. There's a few things. Today I just wanted to show you some applications of network analysis applied to transport network. We think that it's a way to explore and to use, because we can take a lot of information from that. The idea is that all of this calculation and all of this analysis help us to better understand how the Romans in this case, Romans, configured this territory. Because the good idea of that is that it seems that the Romans take into account the optimization of resources and the communication between places to design this network. Thank you.