 Welcome to this section of tutorial 7 on handling microwave data from satellites. So, till now we were discussing about few satellites relevant to hydrology and in particular we have been learning more about the GRACE mission that is gravity recovery and climate experiment and we learnt about the fundamental principle behind satellite ultimate tree and then we saw how to visualize data from JSON2 and CERL satellite. Now here CERL is a joint mission of Indian space research organization that is ISRO and CNES that is French Government Space Agency. So till now we have visualized the ultimate tree data using maps and also we have tried to extract the variables in Python. So now let me try to take your attention to the surface water and ocean topography mission that is SWAT mission. So just a disclaimer, this part of my lecture shall touch upon the research component carried out in radar ultimate tree for the benefit of those interested to perform research in this area. At ultimate tree they provide an indispensable means of complementary information for hydrological analysis and river monitoring and the mission of SWAT to be launched in the year 2022 aims to overcome existing limitations of satellite ultimate tree missions to provide a two dimensional data of water surface elevations for rivers wider than 50 meters. Jointly developed by NASA and CNES in partnership with Canadian and UK space agency the KA band interferometer on board SWAT is expected to provide spatially continuous two dimensional water elevations inside each of its wide swaths. So the SWAT data is proposed to be made available as level 2 products that consists of geo-located water pixels with a water mask, estimated water surface elevations and their associated uncertainty. So in front of you you can see a sample of the data that is expected from SWAT. So it is typically going to be like point cloud of water surface elevations. Now the other kind of products that you get are raster based product which is obtained by aggregating the pixel cloud river reach products and also something known as the node products here you can see the node products. So node products here it refers to a point or a node that is located at every 200 meter along the river centre line that consists of averaged height width over that section of river centred at the node, okay. Now let me take your attention to the open source code for SWAT hydrology toolbox, again this is more aligned towards research in radar altimetry for the benefit of those who are interested to carry out research in this field. So you know the procurement of SWAT derived water level data is a prerequisite in hydrology. So the remaining part of my talk shall focus extensively on the key steps that are being followed for the generation of SWAT point cloud information over a specific case study basin in India. So again what you see in front of you is the open source code for the SWAT hydrology toolbox and the CNES SWAT large scale hydrology toolbox is made publicly available by CNES and the aim is that it helps you generate proxy SWAT water surface elevations while accounting for noises you know such as the random noise or the dark water effects, uncertainty due to satellite positions, tropospheric errors, geolocation errors and so on. So all in all to summarize this particular simulator that is the CNES simulator is sophisticated enough for hydrology error budget studies that is why we are including this as part of the tutorial 7. Now let us try to understand about the CNES large scale simulator. So what I will do is I will walk you through how we have used CNES large scale simulator for generating the time series of synthetic SWAT data for our study region. So for representation purposes as an example I am trying to show you the Mahanadi river basin in India and how we have used the CNES large scale simulator to generate the point cloud information, what kind of information, what are level information over the Mahanadi river basin in India, alright. So in front of you you can see three steps outlined as 1, 2 and 3, they are nothing but the crucial steps to configure and run the SWAT simulator. Now it is worthwhile to note here that to generate the synthetic SWAT data true water surface elevation is corrupted with SWAT measurement errors and spatial sampling in accordance with SWAT orbital configuration. Now before we analyze the steps one by one we have used the Google Earth engine and the Sentinel SAR data to generate the time series of river geometry shape files. I will not go into more details about which algorithm or technique was used to generate the water mask. The inputs are a range of dates for which we need time series and region of interest and this was developed in-house by my team. Once you run the code you get the results in the format that can be exported as a polygon shape file which can be imported in QGI's platform. So once we have the results and once we import it into the QGI's platform this is how it is going to look like, okay. So here you see a sample data set for 2016 Jan 6th over the Mahanadi basin. We can check the coordinate system here. So the CNES SWAT simulator works with the WGS84 coordinate system. So our coordinates are also projected to WGS84. Now next important thing is the height that is the reference height attribute. For this particular example we have used the gauge location height since reference height, okay. You can see the labels and the river flags. So basically we have the time series of polygon shape files, okay. So once we have the time series of polygon shape files, the next step is to create an orbit file, why? Because it shows the past plan of SWAT over our study region which is Mahanadi river basin. So let me show you the commands we have used in the command prompt. Firstly we have to activate the simulator by this command. So I am going to type conda activate SWAT environment ENV. Now once we have the SWAT simulator we have to generate the orbit files and for that we have a document having the SWAT machine specific parameters. Specifically you can see the bounding area for which data needs to be simulated. The bounding box details are shown here and you also see the past plan parameters and the simulation start time and stop time, okay. So we are going to activate the past plan as a yes as we need to simulate it for a time series. So past plan activated as yes and we have a parameter orbit file using which we can run the simulator to generate the orbit files, okay. So we have a code for that what I will do is I will directly try to use this. So once we use the commands this is what you will typically see on your screen because the simulator is taking time to generate the past plan for our region of interest which is Mahanadi river basin, okay. Now once we have the same ready we can also visualize the same in the output folder. So typically it takes a few minutes to complete the process. For example for my case as I have used one month of data sets you know it has taken almost 5 minutes and you can see open the orbits and check the past plans, okay and on to the orbit folder we see the past plans. The details are available here. The cycle orbit, mission time, day of year, simulation start, stop all the details you can see here that is the past plan. Also we get to simulate the SWAT science orbit for these and as I mentioned earlier this is for representation purposes wherein we have used one month of data which you can modify according to your screens. So one month of data is being shown here you can modify according to your need. So now that we have generated the orbit files the next step is to run the simulator. So let us see how to run the simulator for which we have a command which I am going to use directly. At any point of time feel free to pause the screen so that you get a better view of these commands, okay. Now once you type in the command now the simulator has begun processing the synthetic SWAT data and the parameter SSIMP. So here you need to provide the shape file, .shp file, orbit file and the output folder where you see the details are visible here. You can see the orbit parameters and the simulation parameters, reference height. So here name of the attribute is height, okay, alright. So it has taken me nearly 14 minutes and 19 seconds to complete this process. So at the end of which we will get the simulator data. So it does take some time as I mentioned earlier for representation purposes what you see here is the time corresponding to one month of data. For our study region simulation of one month of data from SWAT CNES simulator has been completed and now that it has run successfully I am eager to see how the results look like because I need to further use these for hydrological applications. So let us try to visualize the results in QGIS. Now few results are already uploaded in the interest of saving time, okay. Let me open QGIS. So the polygon shape file is given as input with the help of pass plan we get the footprint shown here is the footprint over Mahanadi river basin. Now we can even zoom closer to the footprint to see how the data looks like. So let us try panning and then zooming. Let us try to visualize the point cloud information which is what you see now. This is the output from the simulator. This is how it is going to look like. So we can go to the attribute table and check the information which contains latitude, longitude, the water surface elevations and other components. These informations can be used as input to the river orbs because it processes the point clouds and we get something like filtered points with less noise, you know. We can again open the attribute table and check the information present here. So just to make a note, another information we get is known as the node based products. So for example, say you need to assimilate the water surface elevation data for which you cannot use this dense point cloud information because then the process is going to become too tedious, too complicated. For that we use the node based products which looks like this. So here the point clouds have been converted into node based that is averaged products what you see here wherein you get information at every 200 meter interval. So they are averaged out and we can even check the attribute table again. Many values are null because it just simulates the data for which we have the footprints. We may not have simulated data over all the regions, wherever there is footprint the data is being shown. So let us zoom out, this is how the node based product looks like. So in this particular section we learnt about the SWAT mission that is proposed to be launched in 2022 year and we understood how to run the SWAT simulator data to get synthetic SWAT water level point cloud as observed by SWAT upon its launch. And again this short section was intended to give you a glimpse on the research aspects being conducted in radar altimetry. Let me hope that you found this section useful and I shall meet you in the next lecture. Thank you.