 Today we are having the first meeting of the community of practice related to the earth observation for water management topic under the digital water program. And today's topic is understanding the different types and applications of EO data. So this meeting will be recorded and made available on demand on the IW website. And after this meeting, you will receive a link with this, you will receive an email with this link and other materials from today's meeting. Also, the chat box, you will find it below. And in there, you can interact with each other and you can also just, you know, ask general questions. Be sure to use the chat box to share your experiences with us. So sorry. So IW's work on earth observation for water management is a result of successful partnerships and collaborations. We are a part of the consortium of European of the European Project Primwater, which looks at EO technologies for better water management. In Primwater, sorry, in Primwater, IW is responsible for communication, dissemination, sorry, and exploitation of Primwater products and end users engagement. Primwater is also represented in the steering group of this community of practice. We also have a memorandum of understanding with geo aqua watch, which aims to enable water professionals to access and share information on the application of EO information and technologies for improved water management. Both IWA and Primwater are represented in the aqua watch steering committee and working group one, which is focused on outreach and user engagement. Primwater and aqua watch also collaborate in promoting their activities within their respective networks. Next slide, please. So today we are going to be hearing from three speakers who will be giving their perspective on the topic of today's meeting, which is understanding different types and applications of EO data. We'll be hearing about next level of water quality monitoring, but from IWA has from EOMAP. We will be hearing about the Primwater operational platform from Evangelos Poromas, from MVIS, and we'll be hearing about the new E-Rail for satellite based water monitoring from Christian Totrop of DHI. Following that, we will have a question and answer with all of these speakers, moderated by Samuel Alagrida from IWA. Then we will have some hopefully active and interesting insightful breakout room discussions, which will include all of the attendees today. Then after that, we will have a short summary of the discussion from these breakout rooms, and then we will conclude with some closing remarks. So just a bit about this community of practice. One of the aims is to make sure that we bring together experts who are using IRF observation technologies within their day-to-day and their research across the water sector. This community of practice is connected to the IWA's digital water program, which in itself provides a platform for others who are working on digital water to come together and share their experiences and to promote leadership, as well as confidence in the overall digital transformation of the water sector. Next slide. So for this COP, our expected outcomes are just to ensure that the IWA membership and the wider public have an understanding and an awareness of the opportunities available for using IRF observation technology in water management, both quality and quantity. Also, it is our hope that IWA is seen as a platform where you can find all of this knowledge and you could also connect with others who are working on these, who are working on similar projects. And finally, we hope that this community of practice will encourage experts, both IWM members and non-members to come together and work together and network and hopefully share more information and probably help us to work better with EO. And today, for our meeting, our objectives, mainly just to discover the types of earth observation, the types and application of earth observation data in water resources, water quality and urban water management, to understand your experiences and challenges as practitioners in using EO products and services. And hopefully to discuss possible outputs of this community of practice, what you can contribute to this community of practice. And now I'll hand over to my colleague, Samuela, to introduce the speakers. Thank you so much, Erin, and welcome everyone. This is Dr. Samuela Guida, and I'm the Strategic Programs and Engagement Manager at the International Water Association. So it's my pleasure to introduce to you Heva Haas. Over the past 15 years, Heva has led and managed projects and teams at international institutions such as the European Commission and in the private sector, which allowed her to build international network and bring satellite-derived services to the market. Heva is the head of strategic accounts at EOMAP, which is the leading service provider of global satellite-derived aquatic information. As a head of strategic accounts, she demonstrates the value of maritime and inland water solutions to commercial and governmental clients. And she believes that the best reward is to get feedback that the earth observation solutions have increased customer's efficiency, save costs, and lower their project risks. So, Heva, the floor is yours. Thank you so much, Samuela, for this kind introduction and a warm welcome to everybody who has joined. I hope you can see my presentation now. We can. Yes. Thank you. Yeah, Samuela introduced my person perfectly. One thing I'd like to mention is besides that I'm the head of strategic accounts, I'm also very enthusiastic about earth observation. And my aim of today is that within the next 10 minutes, I can share that excitement and, yeah, that you also become enthusiastic about earth observation or even more enthusiastic, because having you today in this COP meeting already shows that you have an interest, of course, in that topic. I will tell you more about next level water quality monitoring and start with an introduction about EUMAP, because I think it's important to know us. We are a high-tech company for satellite-based aquatic information products and services, and we are serving to engineering companies and agencies worldwide since 2006. Me and my colleagues are based in the headquarter in Munich or south of Munich in Germany, but we also have motivated colleagues and offices in Australia, the US, Dubai and Indonesia, as well as a colleague and teams in Brazil. So what are our products and services? We are providing satellite-derived symmetry, water quality monitoring information, shoreline and erosion mapping, as well as sea floor characterization and bentic habitat maps. This is all delivered to you via apps or APIs for a very seamless and online product access, but you can also receive that information online and, let's say, as a desktop software for mapping and monitoring aquatic environments. Some facts and figures. We are working on more than 100 projects per year, and we've seen in the past year already a 30% growth. That means aquatic earth observation seems to be quite useful for several sectors, and we, of course, hope to see that number is increasing and that we find more enthusiastic earth observation users in the next few weeks, months and years. So we are serving 35 plus countries and we're using more than 20 satellite sensors. I hope you find yourself in one of those fields. These are the application fields of aquatic earth observation and satellite data is useful for drinking water monitoring, habitat classifications for environmental authorities, algae blue monitoring. It also helps governmental agencies to understand the water quality of any lakes that they might have to survey, may it be on a small district level or even country level. Satellite data can also be transformed into land water elevation models, and it is very helpful for the hydropower sector when they need to know the, let's say, having sediment analytics. That is a very strong economic cost factor. If you have too much sediment in the reservoir, you should know that in advance and react timely. Satellites also help in maritime navigation. They support infrastructure mapping and the offshore blue energy as well as the traditional oil and gas sector. So you see many useful application fields, and I mentioned it before. We are working with more than 20 sensors. I'd just like to name a few. The most important ones for us are definitely the European sentinels. These are freely available data. That is 300 meter resolution daily. That's the Sentinel-3 up to 10 meter resolution with the Sentinel-2. That has a repetition time of five days. And if you want to have more temporarily higher coverage and also spatially higher resolution, then we're using commercial satellite missions such as world view or planet. I've already mentioned our activity fields. So depending on the application, we are using the fitting sensor to serve and make a product and service tailor to your needs. Today, I'd like to look and dive deeper into the water quality monitoring. Water quality monitoring or Earth observation-based water quality monitoring provides near real-time information on water quality, including a lot of parameters that you can already see here in this schematic water body. We can look at the turbidity. We can estimate the chlorophyll A. We can detect the surface water temperature and the total suspended matter as well as the harmful algae blooms amongst other cyanobacteria. The nice thing is that we are using a unique physics-based algorithm. And that means it's harmonized over different water bodies globally. That means we can go wherever on the globe and have a look at the water body there and we'll get good results even if we do not have ground data. Local survey data, ground data is excellent, but we would not necessarily need it. And also it's sensor agnostic. That means we can use any type of the sensors that I have just mentioned, also depending on the application. Let me give you a quick journey here to some of the applications worldwide. I'd like to take you first to New York City, where we work together with Xylem, one of the biggest water company corporates globally. And we serve the New York City Port Authority here. When you have a look at the graph, you see the orange line. These are observational observations from a sample site, from a ground sensor. And you see that here, mainly, let's say, you have, if it goes well, two, maybe three observations per year. So this is the orange line. If you look at the blue line, this is the observation of the same parameter, again, chlorophyll, that can be obtained from satellite data. We have a very high temporal frequency and you see that very different timelines or, let's say, time series are available here. So let's have a closer look to this kind of portal, what the customer sees. We have, and I already mentioned that, the gridded EO product here. You see the chlorophyll data. This is of 2019, sometime in October. And you see that the whole bay and the area and even the river is covered because satellite data, of course, provides you with a holistic view. In comparison to that, you see the blue dots and these are the local sample sites. And obviously, they might measure very detailed, but with the satellite, we are, yeah, with a very high accuracy. We have very comparable results and we can have them at any point that you see in the colored map. We are building so-called virtual stations and at these virtual stations, we can extract the time series that I have just shown you before and, yeah, get much more information out as we can do with traditional means. Let me go to another place also in the US, Lake Elsinore and Canyon Lake in California, where we're looking at the drinking water quality. This is the satellite image. And this is the image that shows, again, the chlorophyll values. We already see that the lake on the left hand side has higher values of chlorophyll and what that means. And if you transform that into a harmful algae bloom indicator, we see that the lake on the left hand side, there is a very likely occurrence of a harmful algae bloom. This can be detected and you even see where the concentrations are higher. Such information is important for the authorities to act appropriately. And what it means to act appropriately, I'm going with you to an example now in Germany, to the Mandichösee. This is a bathing water, so fortunately, it's not a drinking water lake, but sadly enough on the 21st of July, a dog that drank from the water has passed away. And on the next day, there was a bathing water prohibition then issued. If we look back in time, so if there would have been a satellite-based warning, which at that time was not available, we could have seen this happening, because we're using planet data and also the sentinel data. So we're mixing these or putting these together to a very dense time series. And you really see how the, in this case, cyanobacteria hub event is building up. And yeah, we would have been able with satellite Earth observation data to already warn on the 20th of July. So we could have been, yeah, maybe saving a dog's life. And this is, of course, just an example how effectively satellite data is in place compared to maybe traditional means, because if the sampling only happens on the 21st or 22nd, yeah, this is unfortunately too late. So continuous monitoring of the water quality is key for, yeah, the, I'm coming to the key features actually with this, because yeah, key features and the cost benefit of the water quality monitoring. Well, we see, and this is important to know, we can look at water bodies with different spatial and optical properties. We can look at marine and inland waters. The products I've already mentioned, but just again to say we look at chlorophyll A, cyanobacteria, the second-depth temperature, turbidity, and sedum, but also other parameters. This data is then directly delivered to an online data portal. And apart from having that data ready to analyze it, it can also support the ground measurements to allocate the survey sites in the right way. Besides the, let's say the data and the maps, we're also able to provide, of course, monthly reports with the data analytics inside. And the nice thing is this comes customized to your requirements. So you receive reports that fit to your reporting needs. Who are our clients? We are serving federal, state, environmental agencies, water utilities, and local consultancies. And of course, cost is a factor. And I'd like to mention that in terms of analytics and also of the sensor installation, we can state that earth observation is, in average, 10 times cheaper, more cost effective than the traditional means. Furthermore, the health security environmental risk is reduced due to less monitoring campaigns, obviously. And also the costs for boats and other equipment can be significantly reduced because you can reduce the number of times of the surveys. So let me have this overview just to say, what is the general benefit of earth observation for water management? We've already seen the large area overview. And I hope I could show you some examples that this continues resolution optimization. So we are working from 300 meters currently down to half a meter spatial resolution with some of the commercial data. Then it's combinable with other data. This is also very nice to get a holistic view. And it's directly digitized. And very important, most importantly, it's non-disputable. This information is very objective. What do we have in terms of time horizons? We can look back in time for more than 30 years. We are starting our time series at EOMAP from 1984. And in present, we can monitor the current status even in real time and issue alerts, which we have seen before is often very crucial. And in the future, I think we will also hear from that. We can also support forecasting with satellite data. I've already mentioned it because this is what is next and what is new. We will go to more and more spatial and also temporal resolutions up to half a meter and several times daily for certain water quality parameters. We're monitoring at different levels, global to local. And we use increasingly artificial intelligence to improve our algorithms and services, especially for the automated quality controls, for example, filtering out cloud shadows or ship detections. Let's say last slide and maybe the most important information for everybody who is hopefully already convinced, but like the last slide, it is needed to convince you to use Earth observation even more frequently. Well, it will accelerate operation decisions. If you're a water utilities operator, a water utility manager, a regulator or a policy maker, we can serve you all with ready to use data. You will have better and more cost efficient information compared to the current standard monitoring approaches. And yeah, with this, you can assure supply stability. For example, if you're managing aquacultures or drinking water reservoirs. What is also nice, as I said before, we can work globally. We can visit any place on the globe. Despite the pandemic situations, we can go there and have a look. And also nice to mention, maybe, is that it's a carbon neutral survey method. Yeah, I hope you are excited about Earth observation as well. And if you have further questions, I'm around in this session. And yeah, you can also contact my colleagues and me at WeCare at EUMAP.com. Thank you very much. Thank you. Thank you so much, Yvain. You have definitely convinced me. So that's good. Thank you. Thank you again. And looking forward to the discussion later. But now we move to the second presentation of today from Evangelos Romas from Mbis. So Evangelos is a researcher in the field of hydraulic and hydroecological modeling. He's the head of the research and development unit at Mbis with significant experience in 3D hydrodynamic and water quality models for service water bodies, process simulation, automated calibration and data assimilation techniques using satellite imagery and in situ monitoring database. In the recent EU funded project like Speso, Primewater and IFOS, he has contributed to the architectural design of the development of an operational platform for real time forecasting of water quality characteristics in inland and coastal water bodies. So Evangelos has sent a presentation, a recording of the presentation. So let me see if I'm sharing my sound. Hello, I'm Evangelos Romas from Mbis and I'm going to make a short presentation of Primewater's operational platform. The platform is available through a web browser either through the Primewater's website or directly through the second link. This is the initial page of the platform with the four operational case studies of Primewater in Europe, United States and Australia. The Primewater currently offers four services, which are the EU based monitoring system, the hydrological forecasting service and the water quality forecasting service, which is offered both by process based and data driven models. I will now jump to Melbourne Western Water Treatment Plant and I will start with the EU monitoring service, which provides operational EU based water quality products from Latsat and Sentinel missions. On the map here, we are able to see the latest available satellite image for our case study, which is for today, obtained just a few hours ago. The product shown here is chlorophyll A concentration at a resolution of 10x10 square meters. And by clicking on the map, we can read chlorophyll concentration at various points of interest. You may notice a gap here in the center of the map, which as you can see in the true color image is due to the cloud coverage. Apart from chlorophyll A, we are also able to quantify turbidity, total suspended matter, total absorption, taking this depth, and also a harmful algae bloom indicator, which classifies the probability of Hubek system based on the presence of thick organic pigments. Also, we are able to provide surface water temperature from Latsat. However, this product is not available in this case study due to the low resolution of Latsat at 30x30 square meters and the small size of the ponds, which causes a strong interference with the land boundary. However, surface water temperature is available on all the other case studies. In the calendar here, we can see the dates with available overpasses of satellites, and we can go back in time until 2015. Also, there is a useful tool for exploring the temporal variation of a parameter in a selected area. For example, in the graph here, we can see the historical turbidity values for the last three years at the selected point of interest. Currently, these EO-based water quality products are produced by EOMUPS processing algorithms. We also aspire to include algorithms from other remote sensing partners of prime water, for example, the C-Cyrus processing algorithms, and we also aspire to include hyperspectral water quality products from Prisma and Desi's missions, although these products will not be available rationally, but only for historical periods. The next functionality of the platform is the hydrological forecasting service. I will now jump to Lake Hume and select the hydrology tab. This service is provided by SMHI and the World Wide Hype Model Setup. In the map, we can see Lake Hume and all its upstream hydrological catchments. These two contribute directly in the lake, and if I click in one of them, I can view the hydrological forecast for 10 days ahead regarding river discharges entering the reservoir. In the graph here, we can see hintcast and forecasted values of river discharge for today and the following 10 days. Apart from river discharges, the hydrological model provides also forecasts of nutrients, nitrogen and phosphorus and suspended sediments, and those variables are a very important input for the water quality modeling that is performed inside the reservoir. The hydrological model offers both a deterministic and probabilistic short-term forecast. If I switch to the probabilistic forecast, then we still have 10 days forecast, but this time we simulate a hydrological ensemble of 51 individual model trajectories, as you may see in the graph. So apart from the average ensemble value, we can have an estimation of possible rates of values for each day, and we can calculate some useful statistics such as percentiles, which is a valuable input for quantifying and dealing with forecast uncertainty. There is also a third hydrological setup, which is available only in the European case study, which is a seasonal hydrological forecast, where we can have forecasts for up to seven months ahead, averaged on a weekly basis. Moving further down to the modeling chain, we have a reservoir water quality forecasting provided by Envis. This is the Mularge reservoir in Sardinia, where we have used DelphiD Suite to set up a three-dimensional hydrodynamic model coupled to an ecological model. The hydrodynamic model is forced with meteorological and hydrological input and provides us the circulation pattern inside the reservoir and water temperature for the next seven days. The map currently displays water temperature for the top layer on a grid of 100 by 100 square meters and 20 vertical layers. By clicking on the map, we can read temperatures at each cell. We can also visualize the water velocities with these nice animated arrow traces, and of course I can move this time slider up to seven days ahead in the future, in six hours interval. Actually there is a nice tool that visualizes the evolution of the parameter in the forecasting period. For example, in the graph, we are now observing the deurnal variation of temperature in the topmost layer. As I said, we are using a three-dimensional model, so apart from the top layer, we are also able to get information for any depth we would like. For example, here, in the middle layer of the reservoir, the temperature variation is not so pronounced. Values are almost identical for the forecasting period. The vertical profile of each forecasted parameter can also be viewed in a separate graph for any selected point. For example, here the model indicates that there is a thermal stratification at the selected point of the reservoir. The parameters of the water quality model include chlorophyll A, nutrients, phosphorus, and nitrogen, suspended sediment, and dissolved oxygen. Here I have selected chlorophyll A, and the map and the graphs are now showing chlorophyll concentration. And finally, there is a third tool that allows us to view cross-section of any simulated parameter along the two main directions of the reservoir. Both in the hydrodynamic and water quality models, we are incorporating data simulation techniques that are using satellite-based products, as well as in situ measurements, to correct the model state and avoid error accumulation. So, with the process-based models, we are able to expand the availability formation that we have for our reservoirs up to seven days ahead in the future. And compared to Earth observation, we can also have information at deeper layers and information about non-optically active parameters like nutrients. Apart from the process-based models, MVC is also employing machine learning models for forecasting chlorophyll and algae bloom events. So, I will now move to our last cake study, which is Lake Harsha. Data-driven models are trained for specific areas of interest, presented as triangles in the map, using meteorological forcing, forecasted nutrient fluxes from the hydrological model, and are compared against EO-based chlorophyll values, or even in situ measurements. Since in Lake Harsha we had some nice historical time series with a large number of chlorophyll values and cyanobacteria cell counts, we are currently using three types of machine learning algorithms, Gaussian process regression, random forest, and random under-assembling boosting. I will stay in the GPR model, which predicts chlorophyll concentration only in the surface layer this time. By clicking on a point of interest, we are able to read the chlorophyll values at the current time step. Again, there is a graph that presents the time variation of the forecasted parameter. The dust line are the hand-casted values, the continuous line are the chlorophyll values for today and the next 10 days ahead. One nice feature of the machine learning algorithms is that they can provide us with some confidence levels. So, for an 80% confidence level, we can see that the predicted values of chlorophyll concentration increase from 4 micrograms per liter to almost 20, and even higher for the 90% confidence interval. I will now jump to the random under-assembling boosting algorithm to present you a different type of product we are using for bloom forecasting. So, this algorithm has been trained to predict cyanobacteria cell counts, but not as a scalar quantity, but as a probability of exceedance of a specific threshold specified by WHO. And we have made this choice because we have seen that the model has a stronger skill in predicting if cyanobacteria concentrations are going to be over or under the threshold of 100,000 cells per ml, rather than in quantifying exact values of cell decities. So, in the graph, we can see that the model predicts that for the selected point of interest and the entire forecasting over a horizon of 10 days, we are having a no-hub situation with a high confidence level. While if I click to this point, then we see that we still have a no-hub situation, but this time with a lower model confidence. Thank you so much, Evangelos, for that presentation and for showing the possibility of the prime water operational platform. We now move to the third and final presentation of today, Dr. Christian Todrup. Christian is a leading a health observation scientist at the DHI and has close to 20 years of experience in the geographic information systems and the geospatial analysis for addressing pertinent environmental and water resources issues. Christian is currently assisting UNIP with the maintenance of the freshwater ecosystem Explorer Geospatial Data Platform and the Data Science for STG 6.6.1 Progress Reporting. He is also the project manager at the Blob Wetland Africa, Wetland Africa Project, and the Water Surface Water Dynamics, which both have a key focus on mainstreaming health observation in support of water and environmental management and for reporting and acting in response to the global water agenda. Christian, over to you and thank you for being here. Thank you very much for this introduction and I'm very pleased to be here. So, as mentioned, I am working at the DHI, which is an independent private and not-for-profit organization based in Copenhagen, Denmark. Our mission is to try and unlock solutions to water challenges by providing access to real-time data and emerging technologies, relying on almost 50 years of research in the field of water environments, and we are trying to make it globally accessible to our data and software, and this is definitely a nice opportunity to present a little bit about what we are doing in the domain of earth observation. But maybe first a brief background on the water challenge. We all know that our water resources is affected by climate change as well as there are increasing demands for water, for food production, energy, and water to meet a growing population. It almost gets itself done that they don't need to monitor our water resources, both at national, regional, and global level to understand their changes, their vulnerability, and to provide information for sustainable management decisions. Tragically, you can say that we have seen a steady decline in in-situ hydrological monitoring throughout the work, and now there's for sure an increasing awareness that earth observation has the potential to park it till this gap. I think this was already shown by Eva, a few slides back that what is the reasoning for using earth observation. And first and foremost, it's a continuous data acquisition, so we can monitor the earth's surface on a regular basis. It's actually quite nice that it provides access to a historical archive, so you don't need to start your monitoring. At time one, you can actually access plus four years of data to look at historical changes. And then maybe of increasingly importance is the multi-scale in terms of very spatial resolution, but also multi-center capabilities. So there's an ever-increasing fleet of satellites being launched which has different capabilities. And by combining these different sensors, we can say a lot about different environmental parameters and processes come down to local, to river basin, to global scale. There are so many sensors now that it can be difficult to keep track, but there's one important development since around 2015 with the European Copernicus Program. And I want to highlight this as a specific, you would say, paradigm shift because it's really providing the global community access to global data on a free and open data policy. And with this, you can see here, a family of satellites which is providing users with unprecedented capacity to monitor the earth. And especially, I can say the focus here will be on the three first here, the Sentinel-1, Sentinel-2, and Sentinel-3 satellites and how they can be applied to improve our border resource monitoring. The global aspect and the global data is important because at the heart of a number of global agendas, you really find water, whether it's in climate action and private agreement, it's the Sendai framework for disaster risk reduction or the sustainable development agenda. Then there's a need to have access to water data to support the implementation and monitoring progress of these global agendas. And there's actually already quite a few global EO products out there, and many of you may be familiar with the Global Surface Water Extend Explorer, which is a unique benchmark product, you would say, providing access to global data on surface water dynamics for the past 35 years. It is, however, also constrained in a number of ways. It's only using one sensor type, the Landsat system, which again is unique in terms of its historical record, but it's actually not, you would say, state of the art in terms of what you can monitor today. The data is not made available operational and it's only looking at water extent and there's many more options to monitor border resources behind that. In particular, what we have been working at lately is to look beyond that, you would say, single sensor approach where you are using optical data alone or SAR data alone to monitor water resources by combining the two technologies to provide a much more consistent in both time and space. Here are an example from a mapping of mainland China. And of course, it's important to mention also that the new data being made available provide us with a pentapyte of data that needs to be analyzed. So a lot of what we can do today is also being facilitated by development on IPT infrastructures, which allows us to compute these maps at scale in the cloud. So we can do this at scale, but as you can see on the left figure here, actually providing quite detailed information down to the level of individual water bodies. And maybe to further spread some of the benefits of moving from, you would say, the traditional 30-meter global product to a 10-meter product which you can get from the Sentinel satellites. We can capture many more details and actually some studies have shown that there's quite a significant part of the global water bodies which you could say exist within the spatial resolving power of that 30-meter product. So by moving to 10-meter, we are capturing many more water bodies and we can track their changes and they could have quite significant importance or impacts on local livelihoods, but also some ecological relevance. So this is quite important that we can capture more detail. We can also be more consistent in terms of capturing seasonal changes and again relying only on an optical data model. We will in places actually come into situations where there's too many clouds for a specific time period that you would actually have data gaps and it's trying to illustrate here on this temporal development curve shown in this graph and where these red squares show you some months where the global surface of water explorer has not been able to detect water of any significance but where the new dual sensor model can actually capture data because it integrates SAR data which you will know is intensive to clouds. And then we are also available to produce this in near real time and in the end, I mean, if it gives you more time near more accurate area statistics on surface water dynamics. But it's not only about water extent, we can also look at water levels from space using a technology known as radar altimetry. Again, this is not a new technology, it goes back to I think the first altimetry missions was up in the 1970s. But again, there has been some significant developments which means we can do much more today than we could do just a few years back. And what has happened is that symmetry technology especially in the sense it's not an image as you know from a traditional satellite image is being captured at points which in the early missions were spread quite significantly apart up to several hundred kilometers and why only larger water bodies was captured. But with the new missions, the spacing is becoming much less and we can capture many more water bodies and it gives us enough opportunity to look at, for example, changes in reservoir water levels and if we combine it actually with the surface water extent, we can say something about storage changes in reservoir so we can quantify water by looking at water levels and water extent together. We can also use the altimetry missions to look at water levels here from a study in the Sambesi river basin and you can see here we have tried to match up the satellite derived water levels with the station based measurement and see we get quite a nice match up to centimeter, this is centimeter accuracy. And this is of course a huge significance in many of the world's ongoing based basins. And actually in many cases, I mean it's not really the water level but it's rather the discharge which is the important indicator for our users and in that case you can inform hydrological hydraulic models in these on-gauge basins with the satellite altimetry data to produce discharge outputs and give you a strong or better monitoring capacity in previously unmonitored basins. So just trying to wrap this up then, the ability to observe the dynamics of water resources over time has been significantly improved in recent years and it's supporting a number of activities from drought mitigation, irrigation management, planning of infrastructure investment. And the EO data is really plugging in a gap here and it provides us with these new opportunities to provide timely information for making more informed decisions. And as I also mentioned when we combine this with the advanced and technical infrastructures for big data analysis we can both do this at scale as shown but it's also now becoming within the VRM or within the you know realistic chance of countries to actually adopt the technology and do some of this monitoring themselves and not just rely on open access global data sets. And this is just a final slide here to advocate for our pure application project funded by the European Space Agency and where we are actually aiming at trying to empower national and regional stakeholders with not only the data but also the tools so they can independently monitor the water resources and report for the national applications but especially also in response to the global water agenda. Thank you so much, Christian, for your contribution to the discussion. So we now move into 10 to 15 minutes Q&A and I see Eva has already replied to many of the questions and so we are actually for this Q&A we are joined by Apostolis Tsimas that Apostol if you can please come on camera. So Apostolis is the managing director of MVIS and he has deep knowledge of water sector with particular focus on water infrastructure planning development and management working closely with public and private sector water bodies and currently he is the co-coordinator of the Prime Water European project and the IFORS project funded by Innovation Norway. So Apostolis, I actually started with you and I have a question about at which special resolutions you provide your hydrological and water quality forecasts. Right, hello Samella, hello all. It will be a special resolution that we streamline in the platforms that you have the demo actually showed. In terms of the hydrological forecasts referred to catchment levels so broadly to give an indication global hydrological forecasting services run on approximately 1000 square kilometers average catchment levels. In our showcases at Prime Water we have downscaled this information so the models there and the forecast the hydrological models and forecasts run approximately 250 square kilometer areas of catchments. In terms of water quality forecasting within the lake domain there of course this depends also on the size of the water body but in the showcases that you have been demonstrating in the video we're talking about a special resolution of 100 meter cells and more or less 15 to 20 layers on the vertical let's say axis. Okay okay thank you thank you so much for that and Eva I'll go back to you with a question about how do you assure that every user gets the right level of information that is directly useful for their respective application if you can clarify that. Thanks Amuila, yeah it's important because we are serving such a wide range of users from very technical persons. We've seen also in the chat very detailed questions up to a policy maker that just wants to receive maybe a number on his desk and our answer to that is modular solutions so we are providing different modules that build on each other. You can for example receive the eolytics suite that means you really have hands on the data all in a nice interface of course so it's already very user friendly but you can look at the type of sensor data that you like like original data you can dive a bit more into the technical part or you take the EO app and that's like the name already says it's like an app you receive here more the the final product or even the reports can be an app or a portal what I've shown before so it's modular and yeah as a second part just to complete that we also streamline our water quality products into three service lines baseline monitoring and alert so within this all the parameters are actually yeah can be classified and these three service lines really serve the different kinds of users and we have a good customer care so that's how we complete the picture. Good, thank you so much and Christian a question for you, is it possible to elaborate on how health observation is currently being used to support the STGs and the global water agenda? A big question to ask. Yeah I can short cut it a little bit because it's especially the sustainable development goals have been involved so let's focus on that part of it but as you may know it's quite a huge system and indicator framework that the UN system has built and it does really require countries to look behind you know traditional statistical data and especially for the water goals earth observation has proved to be quite valuable and then the UN system has been through a process where they tried to gather information on indicator 661 with the change on water related ecosystems from the countries again because the STGs is really owned by the member countries so they wanted to collect this data at the country level but they found out that a lot of countries don't really have that information at hand so they had to shift strategy and they were then looking towards these global earth observation products and were gathering the information from there both in terms of the surface water extent as I showed but also on changes and there was actually a sub indicator on water quality and also looking at changes in mangrove as a particular water related ecosystem and now you would say they have worked at the global level and the next process is actually to hand they expect that ownership of the data to the countries and try to empower countries to do this mapping themselves so going from the global data to local analysis and local processing and production of the indicator. Thank you, thank you and I just saw a question in the chat from Anthony Kilbreed so maybe I can ask this to all of you so do you see earth observation data for water quality monitoring replacing standard monitoring approaches or do you see them complementing each other in a dual approach? I don't know Eva or Apostoli, so Christian. I mean I can start taking that and we have quickly answered already in the chat but definitely we don't want to replace the ground sensors they have a value obviously what we would like to do is to have the best output for the customer or the user and that means giving him as much information as possible so we see it as a hybrid approach we use earth observation in addition to the ground sensor but we would not necessarily need it so earth observation gives you already a holistic picture and yeah so hybrid. I can only echo that it's not about replacing but about working with the data we have and the benefits from what the individual data type brings so yeah. Apostoli, any anything to add? All data have has their own value they bring their own value so I think here the goal or the aim should be to extract the value from all those data streams available and transform them into the interaction information so no data is just to to abolish so yes certainly hybrid approach here is what we should look for. Perfect okay so we are running a bit out of time so I have to conclude the the Q&A but I thank you all for your contribution and our attendees for the questions so we'll now go into the breakout rooms and let me share my screen once again so the breakout rooms will have moderators and the rapport tour you will be automatically assigned to this so you don't have to do anything but we'll ask you to to please keep your video on during the discussion so it will be a lively discussion and I'll ask Erin to put in the chat a link to go to the Google jump board when you can add your contribution so Erin if you can please do that so how we intended these breakout rooms you will have two different parts in the first part we will ask you to share with us your experiences on projects or initiatives that are related to today's topic which is understanding different types of application of heart observation data and then we will move to your contribution so what do you do you have any suggestion or what you can contribute to the community of practice how can we further share information and if you want to volunteer for some of these actions so I think I'll stop sharing now and please note that I will take a couple of minutes to go into the different breakout rooms but you should be able to join them all so we'll see you in the breakout rooms welcome back everyone I hope you had nice and interactive discussions I think this final part of the meeting is just for us to summarize what we discussed during the breakout rooms so I'll give the word to Apostolis and maybe you can summarize what was discussed in breakout room one right okay so we had a quick quick chat I think I think what is important to mention here and I believe we can keep that also as part of the mandate for these community of practice is the need of better awareness of better sharing knowledge and education on how to actually exploit these very strong technologies about observations and related services so I think this really seems to be a very important point and also a nice and strong let's say topic to consider in our community of practice how to achieve that thank you thank you Apostolis and Eva yeah so we had a very small group a lot of dedicated ladies besides Bibi Laura and me it was Karin Schenk our head of your maps water quality department and a special guest let's say our participant Adele and yeah she told us that she had hands on on flood forecasting she works with data Sentinel and Landsat to do that but also rainfall data so definitely some projects and initiatives that she was involved in and we heard a bit about opportunities because she is from yeah Caribbean let's say from yeah what is important we talked about some products that could be used there and in terms of the activities that could be done in the COP and about the contribution I found we had a very nice discussion that went a bit beyond classical white paper or block which we of course touched upon but it was very important for us or yeah Adele mentioned and I liked it a lot it's like technical and social events so meet and allow people to get hands on so it's more towards trainings so it's really concrete sessions that people see the products in action and yeah maybe also do some joint projects and yes show how products solve specific problems and learn from each other so I found this was a very nice thing and yeah in these days it's not so easy to meet but times will change and I think the virtual world also allows us now to gather more people in a room than we would have maybe done before so yeah actual training sessions was our outcome and I'd like to thank again for the nice discussions that we had in our thanks thank you thank you so much Eva and yeah I do like the idea of of meeting people especially after three years close at home so Christian over to you yes thank you yeah I think also we had to have good discussions and I think one one of the key things that like that I took was was the mentioning again you said there was events and training was mentioned by the other the importance of capacity building which I think we can all agree on I agree to but but also my own experience and some of the participants in the group could also say I mean it's very hard thing also and it is good thinking because you need to meet people where they are and then sometimes you know it's a big challenge to to build capacity in an advanced technology and finding that right level so you don't necessarily need a full phd to to work with the data so so that was an interesting point and then yeah moving forward I think that was a good good suggestions that we try to build or the copies would you know engage in trying to build a best practice from a global perspective not taking a Eurocentric or US-centric approach but really trying to to to look across and find a best practice for water quality monitoring I think was specifically the one highlighted thank you thank you so much and there on on on my side we too had a very little group but I think the discussion was very interesting we talked about the importance to consider user needs in in a heart observation technology but also and this then we connected it to the importance of making heart observation products user-friendly so that we built that trust and that confidence and that easiness on the on the products and we also talk about the importance of involving policymakers which is something that we are considering for example in in in prime water and the on the community of practice we touch upon of course white papers and a webinar with the experts that can share their own experience but also the importance of bringing people together and we talked about the possibility of having a new selectors dedicated to this community of practice so that people know and can share this information because as we have seen for the meeting this morning and this meeting now we have we have many projects and we need to create our community so there should be a way so that people can talk and interact so I think with that we have concluded a bit the summary of the of the breakout rooms and I just wanted to conclude this meeting by giving you an idea of how to keep the conversation going on IWA and network projects and I'll put the link in the chat you'll find a page on this community of practice and there is the link to a prime water european project and geo aqua watch initiative which are linked to this community of practice and a bit of information about the sitting group of this group of this community of practice but also how to get involved if you click on this link here you will be able to go to a survey and you can write down your your name and how you can contribute so that we will include you in our main enlist and our initiative so do join also our IWA connect group which is the platform that we would like to use for this community of practice to share these experiences so I think with this we have concluded the meeting I would like to once again to thank all of you for joining our moderators our presenters our rapporteurs and you of course all the participants so thank you again and do keep in touch with us and Erin do you have anything to add well nothing besides thank you very much for joining us this was the first time we had a community of practice meeting and I think that it went really well so yeah we're looking forward to a lot more in the future