 together with the case that we listened to this morning and Andrea will be present I mean the exposure assessment put in in a more practical way we decided to have now as the time for coffee is fixed and we have the coffee break will be 10 to 4 and now we took now for half an hour and I don't use it the GIS and the first half an hour will be in a theory and then an exercise is not really and it is an exercise in individual base because it was most complicated to organize but it will show you the some examples so again please be free to as you and if you if you need this extra explanation even during the presentation because now is the time to to ask thank you very much Andrea may be introduce yourself a bit thank you so much thank you cargo thank you Eric and organizer for inviting me to this very interesting course and I will I will go on a on the on the argument that the case in this morning has introduced about the exposure assessment of population to environmental stressors and in particular I I will talk about GIS in this in this topic I I'm I'm from Italy I work at the environmental DPA of Emilia Romagna region that is the the poverty in the north of Italy one of the most polluted area in at least in Europe but not only and it's 20 years I work on on matters about dealing with environmental epidemiology at impact assessment and in particular on environmental exposure of population to environmental stressors as I I work in in an agency that deal with the environment environment okay first of all case at the end of these lectures send send us an email with additional information you you asked in at the end of the lesson so I put this information on my slides but as the title said is I referred to the lesson from case he wrote the two websites where ADM as urban the commercial product the commercial dispersion model is a valuable is downloadable which is and the other one is our model is a free but as case said not so user-friendly software that can be downloaded by the website of the EPA and again from case the information you asked for about European papers on satellite data and rule modeling there are several papers these are three very important article that case and colleagues wrote in in this years so I will leave this information on in the in the in the power point five so going on quickly I will introduce the concept of this very very quickly and and and how the GIS GIS can play a role in in exposure assessment through in the first part through a little theory and in the second part with the some example because as a color said because in for for reason of a very short time and and difficulties in organizing a real practical I will show you some example by slides also with the summer screenshot from our map that is the most used the commercial software but with the example of function that you can find also in the free software like could use or grass or other or other software like friend to do to do GIS analysis and there is a short introduction to Gis what role the role the GIS plane exposure assessment some example in particular on proximity buffering some concepts that case already introduce this morning and in particular in exposure assessment studies so GIS what what is what can do a GIS geographic information system starting from a common user of the Google Maps or other other software that can allow us to put the our mouse or our situation in in in the point where we are I try here to center the Leonardo building this building in the in the area starting from this the GIS can elaborate this information there's this photograph there's this kind of information and separate the information in the different strata code layers that can explain us the different kind of information such as a reduced elevation sensors that is a proxy of density population streets and you can see that are all information directly or indirectly related to environmental stressor in particular a pollution so a GIS geographic information system can help us to managing this special data mapping different kind of information starting from environmental stressor or situation related to population and going on to disease to mortality morbidity data to all information about the illness or mortality that are the three big argument chapters that we have to use when we do epidemiological studies and also at impact assessment exercises and moreover there are many many possibility to put this possibilities to work with this data in a spatial way and so modeling this data and with the linkage or integration of other models for to give an example of these three big chapters if you if you see this the slide we can we can see that we can with the GIS we can specialize a variation in an environmental hazards so we can retrieve information we can obtain information about the heterogeneity of this the distribution of this environmental hazard at the same time in the same place in the same area we are studying we can put on the GIS the spatial variation in the distribution of population and at the end also regarding the again with the population distribution information about population that can be very useful to understand where which population and where the population the most susceptible population is located starting from information about age or sex or other other situation related to each individual and combining this information we can we can have a spatial variation in health outcomes and what is the added value of a GIS software GIS system a GIS approach in in our analysis is that common database allows us to answer to a question like who what when why how this population disease an environmental stressor is occur and in in in the way of occurring but a GIS help us to add the information about where so we can talk about location we talk about trends so we can move on not only in the space but in the space during time we can identify patterns there are also some geostatistical modeling and approaches that we not don't I don't show here but like clustering that is an approach that start from a GIS situation a situation of location of pathologies in the space and and also more of course modeling so also the scenarios counterfactual scenarios that Francesco for us to introduce in his lecture about health impact assessment so how the GIS go through the chain that we will to in an environmental health study or environmental health impact assessment starting from the information about land cover roads topography and other information about the situation the environmental situation we can build a pollution map that is a distribution in the space of the environmental stressors is a good a good practice to validate this our modeling our information we we build the in the GIS software with the monitoring data with the real data if they are available this is the this is not only the case of of for example we're modeling that case introduced this morning but also a kind of validation of our approach to to modeling that could be be made without some information that is crucial for for a correct interpretation of the this pollution maps so validation is is a very important aspect not not I want to to underline not a validation of a single model we are not talking about validation of our mod or another distribution model we are talking about validation of our approach in the environmental health chain to assess exposure of population and then the fundamental aspect of where is the population so and in combination with pollution map we can provide an exposure map and with the information about the exposure response relationship we can talk about at risk in different ways from a point of your risk assessment I think but assessment or make an epidemiological studies a study to to give more information to the this relation to this exposure response relationship the source of data is increasing in the in the last year so there are many many information that are specially available and starting from the existing maps going on on routinely collected data satellite data in this in this morning we talk about this this aspect routine monitoring data and also data built for a specific study design or specific study protocol but the key issues is that these data are to be georeferenced and and a location is a fundamental to to define the location of this this data and this is a very simple aspect as a concept of a very crucial aspect when we work on this on this instrument because there are many many many huge availability of software and way to geocode the information but we have to pay a lot of attention to the accuracy and precision of this this processes this morning there was a I think if I remember a question about how to represent the street in the format and the main two main way to store information in a GIS system are vector formats and raster formats and what is the the main difference between the two if we consider an example about a motorway street a big street we can have a file that describe a GIS file GIS structure GIS a number of files that describe this streets in two way with the vector formats include discrete representation of reality is provided there is a table with the location information about each point and a way a file that described the way in in which the these these point are related and are linked another way is a more under understandable and more common format like images that is raster format in which the the the square is divided in in little squares and each each square contain information that can allow us to understand where the the street is about vector data is the main the main way in which the the information are stored in a GIS file and we can have three kind different kinds of vector data that are points lines or polygons and it depends on which on which information we have to store for example we have if we have a GIS system that represent a GIS file that represent the location of our fixed monitoring stations in a in an area we we will have a point file with a table each of this file has a database one of the file that construct the layer of a GIS file is a database commonly is a database in the in the base format and or other format of common form but there is always this information that contain information about each of the points that are stored each line is a point and then we have other information about that allow us to to put this point in in in the space in the GIS system another system is made by lines as I show before in the slide before about the streets and we can we can think to also streams or also disease vectors and the other one is polygons we can think to a shape file that represent the buildings of a city all buildings are represented as polygons on the contrary raster data has a common structure that is a grid with different resolution that allow us to to understand by information all about each square little square what what which and what what what are and where are the information we we are storing and as I said before for each of these five there is an attribute data that is the the database I said before every every shape file that is a file the name of the file in GIS system as a separate file that contain for each location that can be a point or an arc or a big or a polygon information about this this single structure in this example we have information about I think buildings in in an area with the the area of the building the perimeter the name we can store everything we want to to and we are interested too so with this information we can what can we do what we mainly do with these strata these different strata that are overlapping one each other we can make an analysis of overlapping overlay analysis and in in this example we can we can see three different type of analysis a point in polygon a line in polygon and polygon on polygon these are common way to to name the the GIS analysis and there are very simple explanation that these that we are overlapping information information about point in a polygon a line in a polygon and one polygon in another and this this could this explain the results this in in this case at the end we have information we have another shape file in point format with an additional information we have we will have another column in the database information with information about the characteristic of this polygon white instead of instead of black or other information that it contain in this polygon in this in this situation this information is transmitted to a line polygon and in the last the last example we have two different polygon that intersect and we have as a result a new polygon file with four character for different different information another very used commonly used analysis is in in GIS file in GIS analysis in environmental epidemiology is a proximity analysis and this is an example about distances and buffering distancing distances is related to the capability of the capability of GIS to calculate distance between points distance between a point and and the line or a point in the polygon in in in a very simple way and this is we if we if we think to our analysis you can you can think to a point as a for example and an address of people and we can calculate distance to environmental stress or this could be a street and you can calculate which person are close to a roadway or a major major road and or isolate people that are closest identifying the threshold and this is possible with the simple simple analysis in GIS another situation is buffering if we have it's typical when we have here not a receptor like human people but an environmental stressors like chimney or or a point of measurement of pollution and in this case we are interested in capturing people or or our receptors that are in a circle with a predefined radius and so we 90% of times we can we use this this approach with when we have people here and this approach when we have an environmental stressor here case a talk a talk already already talk about the the first law of geography that everything is relating to everything else but near things are more related that distance things and this is the base the basis of this all these approaches he show also show this this slide so because of time I go quickly on this this regarding a pollution approaches he talk about different possibility to model special data when we have different point of measurement and depending on what are what are our goal we can use one instead of another and this is an open debate of which approaches better in related of the situation of our study for example this this situation that is the nearest model of the foranoi tessellation that unify all points that are most close to an initial point than than all others it could be useful when we are dealing about I don't know electromagnetic fields and the situation of exposure to a source of electromagnetic fields but if we are that is a very fixed situation not related to other points but if we have or for many point of view another useful application is this one is related to the closest hospital where to go when we have an emergency situation and so the point is very very rare respect in respect of this this map but you can imagine that this is a very useful information to choose the the most closest hospital to go and this information can be linked also to information about streets and to to find the most not the most close in in in in a sense but in a but most close related to the availability of streets but in other situations like air pollution other approaches are most used and in the slides in the slide I collect briefly collect the most common approaches to air pollution modeling most of them cases have already presented starting from proximity models that are very simple that are based on distance based on only a geographical situation a geographical information without any information about also without any information about the quantity of pollutant but in in a vermetapidemiology they are very common very many many studies that use the simple information with the important results for example the the resident in proximity of important industrial industry or in a round a very high polluted a very high traffic road there are many evidences epidemiological evidence on this based on this simple indicator these are there are other models that the case already discussed there is an increasing of difficulty complexity and also performance and the the last one is a source apportionment that is an aspect that it's important for for our work because there are statistical models that allow us to comprehend consider the contribution of one single emission sources at the is a statistical approach user that the user GIS to locate information but maybe is a statistical method that applicable when the tail pollution measure and chemical profiles of its source is available but it's very very important also from a point of view of policy makers that because it can provide information about the impact of a single source like traffic or heating or any other agricultural activities or any other approach perhaps I have to move quick it's fine it's time to stop two minutes okay okay I go quickly for example an example different two different approaches proximity and modeling when we have a point source pollution like an incinerator in industry it's very very useful to have information about the dispassion model of this plant of this emission sources because if you can see here the difference between considering these population the red one are population are address of population the if you if we consider distant bands we have these population that is equal exposed to the industry instead if we have the information about the dispassion model we can see that population that is here is very very polluted is leaving a very polluted area in respect of the population leaving here at the same distance of the plant so the choice of the approach depends on the source of pollution and the availability of data but my opinion is a personal opinion but in in case of the point source single point sources of pollution a dispassion model is crucial to have a good approach to exposure assessment the last things I want to talk with before the the break is a residential exposure we don't discuss this but is under our our thing but a residential exposure is a crucial point a very critical critical point because in our studies in 90 percent of study we use residential exposure as the point of exposure of population we know that we move especially we especially we move in during our working age and but there are many service that are surprising for me demonstrate that also people in working age spend more than 60 percent of his time at the address location so if we have to choose one location the residential exposure is the best and we have to be very careful in geocoding our residents because if we go to on the internet of the Google maps and we put an address the software provide us a point but this point is is linked to a sort of probability of success of match of except match and this is very important because if we not have information we don't have the correct information we can put a point of address here instead of here and is a real example when we have no information about the real number of the address or so on yes questions you have a question but questions in the room? I've got a question from a she was asking in all the lectures for the moment we haven't mentioned the health impact of climate change that is highly affected by air pollution and many other variables and just focused on air pollution is it possible to assess the overall impact of climate change on human health GIS or any other mean? yes okay okay I've talked about the GIS in using for environmental stress one of environmental stress is of course also temperature and matter related on climate change so yes there is the possibility to use this information and when when we are interested and when we are in the situation in which an environmental stress related to temperature is specially distributed in an area so this is a case in which the GIS could help to define different levels of exposure to this environmental stress is someone using the GIS here yes please how many people you want to comment easy difficult what do you use the personal address or my address of people in fact I am a PhD student in human sensing and GIS and I think it's usage in human health and what we were talking about is a lot and I was curious about the modeling which you use and from my experience I was just working in a project which shows the impact of different weather parameters in one disease pediculosis although we found that there is no relationship between pediculosis and atmospheric factors but I think that it's a good way to represent the different factors the relation of different factors in health and it's going to be continued yes representation is a key aspect of GIS I have talked about the situation in which we consider an environmental stressor as a potential influencer of our distribution of health outcomes but yes also the easiest aspect the first aspect of specializing health outcomes is very important as different applications such as clustering or other situations that could be very important yes open access I often use ARCMAP that is not an open access software in the second part we will see some example on ARCMAP and we use also QGIS software that is I think the most close also as a philosophy as a structure to ARCMAP I think that these are two main software we use there is an increasing but I'm not an expert of ARC colleagues of mine are expert of ARC and there is an increasing availability of libraries in ARC software that are obviously free that can manage in a very good way GIS data, geographic data ARCGIS ARCMAP is revolutionary okay so we can go for the coffee break so we take 10 minutes coffee break so we are back at 12