 Good morning, good afternoon, and good evening to everyone. I'm Riccardo Verney and I work at the IAEA in the annual energy statistics team. Today I will present the electricity and heat module followed by some exercises. So firstly, I will talk about some key electricity trends, some key concepts in energy and electricity statistics. And of course we'll have some time at the end for questions. So let's start together with some key electricity trends. So as you can see in this graph, this graph represents the evolution of world electricity production between 1974 when the IAEA was founded and 2020. Over this period, world electricity production increased basically every year with the exception of 2009 following the economic crisis and 2020 due to the COVID-19 pandemic. Overall, this resulted in a fourfold increase in the electricity production from roughly 6,000 terawatt hour in 1974 to about 27,000 terawatt hour in 2020. So now I have a little quiz for you. So let's go to our quits platform, Menti. So the question is, what is the main fuel used for electricity generation in the world in 2020? So if you go to menti.com, you can use the code that you can see, so 74955713. And you can answer this question. So I wait for your answers, okay. I start seeing some answers for coal, oil, hydro, okay. Basically we covered all the fuels. We have answers for all the fuels, but we have a winner, which is coal. And coal is the correct answer. So let's go back to the presentation. If we have a look to the graph of over this 46-year period, the mix of fuels used to generate electricity changed. For instance, if we look at the share of electricity from oil, we had some answers for oil in Menti, but as you can see, oil has fallen from about 25% of electricity generation share to less than 3% in 2020. And in a similar way, also the share of hydro decreased from 23% to around 17% due to the fact that many of the suitable large-scale sites have already been dammed, of course, and by contracts, we can see the share of production for natural gas, which is the light blue one, increased significantly, rising from 12% to almost 24% of generation. So it doubled in this period. And there is also a noticeable increase in the production from a nuclear power, the yellow one. It's shaded in yellow and its share increased from 4% in 74 to roughly 13% in 2020. So during this time, you can notice the answer we gave together, you gave. So you can notice that coal still remains the main fuel used for electricity generation, and which is share changing very little. And the main reason is its relative cost advantage over other fuels. However, also on the mission front, there is some good news, because if you look at the share of electricity output from solar and wind shaded in orange and light green, a second from the top of the graph, you will see that even if the output of these renewable sources is still a relatively small share of the total output, their share has been constantly increasing in recent years. As production costs have fallen, and also countries have begun to implement more environmentally friendly policies. So in summary, to summarize, compared with 1974, the electricity mix has a lower share of oil and hydro, an higher share of natural gas, nuclear, solar and wind, and with the share of coal, which remains a constant. Now let's have a closer look at data for electricity generation in 2020. So as mentioned before, coal remains the dominant fuel, providing 35% of electricity generation, followed by natural gas, which provides 24%. And we have to consider that in total, almost two thirds of all electricity generation comes from combustible fuels. So coal, oil, natural gas, biofuels and waste, with the remaining third produced by non-emitting sources such as nuclear, hydro, solar, wind and geothermal. So if we go to have a look to the world electricity production by region, we can compare OECD countries and non-OECD countries. And as you can see, electricity production in non-OECD countries has grown much faster than in OECD countries over the past 20 years. So in particular, production in non-OECD countries surpassed OECD one in 2010, around 2020. So since then, the electricity production in non-OECD countries has 20 to increase, while production in OECD countries has reached a plateau. And in 2020, the share of production of non-OECD countries is around 60%, compared to the one that was in 1974, which was around 30%. Then, if we desegregate these non-OECD countries production by region, we can see a very interesting topic. So much of this growth in recent years has been driven by China and other countries in Asia. Now have a look to the electricity consumption by sector. And firstly, you can notice that consumption has increased four times between 74 and 2020, growing from 5,000 TWh to almost 20,000 TWh. This is not surprising at all, because if you remember a few slides back, we saw that also production increased fourfold. So of course, it makes sense that consumption and production increased with the same trend. However, the numbers are slightly different, because if you remember, the production was 27,000 TWh, while here the consumption is 23,000 TWh. So what is the reason for this gap? The reason are transmission and distribution losses, which are around 7% of the production and energy industry on use, which is around 9% of the production. And looking at the consumption by sector, industry is the largest consuming sector, followed by residential. However, we can also hear distinguish between OECD countries and non-OECD countries, because this is our word graph. But if we have a closer look to the distinction between OECD and non-OECD countries, we can see that for OECD countries, three sectors, which are industry, residential and commercial and public services, consume roughly the same amount of electricity demand, so around 30%. By contrast, analyzing non-OECD countries, industry represents, I would say, almost half of the total electricity demand. So there are many reasons for these differences, which are mainly the structure of their respective economies, income levels. But however, also as we said before for the electricity production, this trend also different from country to country. So this was the trend modules. Now let's move into some key concepts in the reporting of electricity and heat statistics. Firstly, let's have a look at the source of electricity. Electricity can be produced both as primary and secondary energy. Primary electricity is obtained from natural sources, such as hydro, solar PV, wind and tidal and wave energy. Secondary electricity is obtained from the heat generated by burning combustible fuels, from nuclear fission, from geothermal heat and from solar thermal heat. Also heat can be produced as primary and secondary energy. Primary heat is obtained again from natural sources. Natural sources are solar thermal, geothermal and nuclear, while secondary heat is obtained by burning combustible fuels and from transforming electricity into heat in electric boilers or by capturing heat from low temperature sources using heat pumps. Now let's make a distinction between two different kinds of producers. This is by far one of the most important concepts of energy statistics because electricity and heat producers can be divided into two broad categories, which are main activity producers and auto producers. So main activity producers generate electricity and heat as their primary activity. So the main purpose of the facility is to generate electricity or heat. For instance, a nuclear power plant is a perfect example of a main activity producer. On the other side for auto producers, auto producers generate electricity or heat only as a secondary activity. So let's make an example. For example, a large chemical site may have a proper power plant to generate electricity and heat, to use them in the production process. And some of these electricity or heat may be in excess and be sold. However, the main purpose of the plant of the facility is to produce chemicals, not to produce electricity or heat. So this is why it's called auto producer. And to conclude, let me highlight that it's important to note that the distinction between main activity producers and auto producers is based on the activity, on the primary activity of the facility, not on whether they are public or private companies. So we can distinguish between main activity producers and auto producers, but we can also distinguish plants in three categories. There are electricity only plants, which as the name suggests, only produce electricity. We have heat only plants, which again only produce heat. And we have also combined heat and power plants or CHP plants, which can generate both electricity and heat in a combined process. Now let's have a look at some reporting convention, because we made this distinction between main activity producers and auto producers and between electricity only, heat only, and CHP because we have different reporting convention according to every kind of different plant. So for main activity producer, the convention is quite straightforward. So all production of electricity and heat must be reported. However, for auto producers, there are some specific conventions. So for auto producers, electricity only, all the production is reported. For auto producers, heat only, the sold heat is reported. And for CHP auto producers, again, all the electricity production is reported and only the sold heat is reported. So let's make an example, because I think it's way more clear than the theory. If we have an auto producer, heat only auto producer producing 100 units of heat, it sells 30 units. We have to report only 30 units of heat in the electricity questionnaire. And in a similar way, only 30% of the fuel inputs would be reported in the questionnaire. So in this way, the efficiency is not disturbed, because otherwise, I would have a big input without just a partial output. The remaining 70% of the fuel inputs associated to this heat would be reported in the appropriate questionnaire of the specific fuel. But the heat not sold is not reported anywhere. So this is really, really important. And another important concept is the distinction between gross and net production, because we have gross production that refers to the total output of electricity or heat generated in a plant before it's used. But not all of these is used for productive purposes outside the power plant. Some of the electricity and the heat generated is used on site at the power plant itself for lighting, heating, support, plant operation, et cetera. And this is called on use. So the remaining production that is left over after subtracting the on use is called net production. So remember that for auto producers, production only refers to the heat sold. So if we put together this concept and the concept of on use, we can sit together another reporting convention, which is really important. Because we just said that for auto producer, we only report the amount of heat sold. So not the total heat production. If we reason in this way, we cannot report the amount of heat used for on use. So we need to make an assumption on this. Again, as before, for main activity producer and auto producer is quite simple. Since net production is the difference between gross production and gross production subtracting on use, and we have net production. But for auto producer, for the electricity only, no issues is the same. But for heat only and CHP, we need to make an assumption. And the assumption is that gross is equal to net. So we don't have any on use. This is basically because for auto producers, it's really difficult to distinguish between the heat used for on use and the heat used to support the plant operation. So this is really important. I hope it was clear just to recap all the convention that we saw together. For heat produced by auto producer, there are different reporting conventions. We only report the heat sold. And we assume that the gross is equal to the net. For main producers and auto producer electricity, there are no particular conventions. So let's change a little bit the topic because we have covered, I would say, the tricky part of the presentation, which are the reporting conventions. So let's move on to have a look at the electricity and heat supply chain and some of the data that we collect in our questionnaire. So on the side of the supply, we collect fuel inputs to the electricity and heat generation. We collect gross production, on use, and net production. And we also collect data on a pump at storage hydro, electric boilers, and heat pumps. And finally, we also collect data on speaking always on the supply side. We also collect data on trade. So putting together all this data, there are all the data on the left side of the slide, we have the supply side of the electricity balance. Now, in between supply and consumption, the electricity travels along cables, so we have losses. And this data are also collected in the questioners. So finally, on the right side of the slide, we have the final consumption, where we collect data on consumption across the various sectors, such as industry, transport, residential, agriculture, et cetera. So in theory, the difference between the supply side and the final consumption side should be just the losses. However, in reality, since we collect data from the supply side and from the final consumption side, which are different, we also have some statistical differences. And to conclude, we also collect data such as peak loads and capacity data, which can be really useful for electricity analysis. So to examine in a little bit of losses, a little more in detail on the magnitude of the losses that can be expected and on why there are these losses, electricity travels through cables and transformers, and some energy is lost on the way. Much of this is lost in the form of heat, as the electricity current flowing through the cables raises the temperature. And this energy is lost and it's basically dissipated in the surroundings and it reduced the amount of energy that can reach the final destination. So to give you an estimation, losses can be expected to be in the range between 5% and 15% of the total production. And of course, quite obviously, if we have older grids, we can expect higher losses or we have more distributed grid, we have higher losses. Also we have higher losses when there is a high rate of bypassing of the net so basically theft of electricity. And let's make a comparison between the production and the final consumption, because as we said before, the different should be the losses. But we also have the own use of the energy industry. So as we can see in this slide, the numbers are the same numbers we analyzed before in the previous slides, if you remember. So we have the production, which is around 27,000 terawatt hours. And we have the final consumption, which is around 23,000 terawatt hours. So what is the difference? We said before that the difference was losses and energy industry on use. This is correct, but just to give you an idea of the magnitude, in total about 16% of the total gross production, which is quite a sizable amount, is lost or used out of the final consumption sector. OK, and now I will talk about some really interesting definition, which are efficiencies and capacity factors. And these are really useful because they can be really useful for statisticians in order to check the completeness and correctness of the data. So efficiency, what is efficiency? Efficiency is calculated as the total gross energy produced by a plant divided by the energy content of the fuel used to produce it. So basically the output divided the input. And of course, this efficiency must be lower than 100% because you cannot create energy. And the expected efficiency may vary depending on the fuel and the technology used. For example, if we take combustible fuels, combined cycle gas turbines are expected to have a higher efficiency than cold power plants. Combined cycle can reach 50% of efficiency. Efficiency must be calculated in energy units and you must use the same units for the inputs and for the output. So please, if you use terajoule for the output, you have to use terajoule also for the input and the same for kilowatt hour. And also please note that for the input, it is calculated on an NCV basis. So NCV has to be used. And this is particularly important for natural gas because natural gas usually it's reported in a gross calorific value. So efficiency can be a useful check for statistics because we can use our knowledge of expected or historic efficiencies in order to check if the reported input and outputs make sense. Regarding traits, unlike other fuels, we have some differences in the convention for reporting because electricity and heat are reported on the basis of border cross. So not origin and destination. So also transit is included. So for example, if there is an export of electricity from Portugal from Portugal to France through Spain, this has to be reported in this way. Portugal reports export to Spain. Spain reports imports from Portugal and exports to France. And France reports imports from Spain. So even if the trade is from Portugal to France, Portugal and France don't report any trade with each other. So this really differs from the convention used for other fuels. So please keep in mind. And there is one more key concept in electricity statistics that is really useful. That is the difference between energy and power. So power is the rate at which the energy is used. So we can say that power is simply energy divided by time. And while power is measured in what? Energy is measured in joule. So one watt is equal to one joule per second. And I would like to focus just on the important distinction between watt hour and watt, because watt refers to power while watt hours refers to energy. And this is the last definition I would like to share with you. I was speaking before about the capacity factor. So in addition to production, consumption, and trade data, the electricity questionnaire also captures data on power plant capacities, in specifically, speaking, net maximum capacities, which are the maximum power output that a power plant can produce on the last day of the year. So on the 31st of December of the reporting year. And these data are really useful for analysts, because they can be used for checking the data as we can compare actual reported production values with the maximum potential production to check if the data makes sense. So the capacity factor is defined as the effort production divided the maximum potential production. Again, it has to be lower than 100%, because a plant cannot produce more than the maximum potential production. And the maximum potential production is defined as the capacity of the power plant multiplied for the time. And if you are considering a capacity factor of one year for the time, you have to consider 8,760 hours, which are the hours of one year. And also, this capacity factor also like efficiency, it depends on the technology. For instance, if we take nuclear plants, they are expensive to build. And therefore, they are typically around as much as possible. So they will have a high capacity factor around 90%, while solar PV, which does not run at night, basically. And also, it's also weather dependent, because if it's raining, it's not working, they will have a lower capacity factor. So as statisticians, in the same way we used efficiencies before, we can use both historical and expected capacity factor as a check on our data. So finally, some comments on data collection and sources of data before we move on to exercises. As we, with all the statistics, data collection has costs in terms of time, of course, in terms of money, but also in terms of opportunity costs. Because please remember, it should be kept in mind that not having the proper information available may lead to higher cost in decision making. In terms of specific challenges for electricity statistics, we have to mention the liberalization of the market that had an impact, of course, because previously, when there were monopolies, all the electricity data could be gathered from one company, but now data needs to be gathered from many different companies. Some big, some small. So in addition also, these companies, there is competition between these companies and competition leads companies to be reluctant to share their data and also to confidentiality issues. So this can impact a lot data coverage and completeness. And of course, similar to other fuels, resource limitation in statistical offices and high staff turnover can cause data quality to suffer. In terms of data collection, the objective is to have detailed and reliable data on the different parts of the production and consumption. And for it, we need a mature and sustained energy statistics system. So to facilitate this system, we should establish a legal basis, establish a proper reporting mechanism in the form of questionaries, which have to be as user-friendly as possible, a network of contacts on a grid tent table. And we also need to establish proper dissemination mechanism to allocate proper resources to collect and process the data and to review methodologies and processes in order to anticipate and adapt to change in the energy situation. And since we are talking about data collection, what are the types of data collection? Firstly, there are surveys. So countries can administer surveys with energy suppliers or end users. And energy companies include power producers, transmission and distribution system operators, market operators, or electricity exchange. And on serving energy companies, we can gather information on the amount of electricity and heat they produce and eventually sold in the market. These data are usually quite easy to be tracked because companies, for the companies, they are essential for their accounting processes. Surveys on enterprises, on the other hand, provide the advantages of obtaining detailed information on the amount of electricity and heat consumed in a particular sector. So in particular, surveying enterprises is helpful in capturing electricity production by auto producers. And to conclude, household surveys, to conclude the surveys, household surveys provide comprehensive information on the type of specific end use of electricity in the household. For example, space and water heating, lighting, cooking, et cetera. And even if the data gatherer are extremely comprehensive, this surveys has some issues because they usually present high response burden. And they are also time consuming and normally expensive to administer. A second source of data that we use are administrative data. So these are data collected primarily for non-statistical purposes, usually for implementing programs and policies and are adopted for producing statistics. So to make an example, energy regulators have information on the type and capacity of issued licenses for power plants or transmission system operators have information on the amount of electricity passing through their network or again, custom offices, et cetera. We can also gather data via direct measurement. So there are some variables that are better measured rather than survey. For instance, quick example, the energy consumption can be collected by smart meters. They can provide up-to-date information. And finally, estimation and modeling, allow data to be quantified when actual measurements are not available. So they provide a quick results and can act as a substitute to reduce the frequency of expensive surveys. But its result is high dependence on the accuracy of the model and on the input data. And an example of this is the, for example, for solar PV plants, if the actual electricity production is missing, if we have the solar PV capacity and we have a model solar radiation value of the location, we can use them to obtain an estimate of the electricity generated by solar PV plants.