 remote sensing and GIS for rural development, NPTEL course, this is week three lecture two. In this week, we have been specifically looking at the Indian government data sets, open source data sets that can be used for getting data for rural development. In the last class, we had an overview of Boone's website portal and we looked at some random data and how to download and link the data for rural development. In today's lecture, let's look at the specific data that we use to take from Boone for water resources. So water is the theme here and we would like to look at the Boone Open data archive. Yesterday or in the last class, I had mentioned how to pick different tabs in the dashboard. And one of the tabs was the thematic products which are stored in the open data archive. So what do you see in open data archive is satellite and sensors that you could see here, theme products. So basically some data is aggregated as thematic products and some other data is kept as program under program and projects. So we will now look at ISRO's Boone website to understand the different sectors and spaces that are looking upon water. But before that, I will also touch upon a link within the Boone website that has applications on water and very, very limited focused applications. So let us open the portal. So we have Boone's website here and what we see is so we have the Boone and RSC open data archive. But before that, let me type the Boone and RSC website and show you that we do have, if you scroll down, the application sectors. In the application sectors, we have water. So in this, there are very specific applications that have been developed using GIS data. One is Wallama Tree, SATA, IVP, and Maharashtra WRDS, Telangana WRIS. It was moved so you can just click here to go back. And then you have the National Hierology Project, WBIS. Let us click the National Hierology Project to understand about the project and then where the data comes in. Okay, so you could see, read about the drought index there, created a decision support system. You can lower it and then see flood early warning systems that have been created for the basin. Okay, so this here, they have the Godavari basin or Tappi. So across India, only two basins have been created. And that also clarifies the need for this course that all areas cannot be mapped at once. It requires a lot of capacity and the students taking this course will be able to bring that capacity. Okay, so let us look at Godavari basin. For example, you have this Godavari basin and you have the observation data points. So you can see here, these are the observation data points. Let us zoom in to see more. So you have these two payroll observation points. And those are used to look at the rivers discharge and look at where the water is going to be above the flood level. Okay, so you can do an inundation simulation and remove these for now. And then you could see historic simulations, selected date, let us say some hours. And you could see that these are the where the inundation happens. Okay, inundation is the level increases and it comes out as floods. So this is a very focused application. And the latest data you have is August 9, 2022. So there is a bit of delay, but still it's a good effort. And a lot of understanding can be taken from this image. Like mostly here, you could see that the downstream locations are mostly getting flooded. Okay, so you can see here that the flood is here. So all these areas could be given a warning to, you know, go to different areas during floods. Evacuate. Okay, so this is one example, there are multiple other examples for the application sector. I'll do the Maharashtra WRDS, where it gives you resource support for managing the water. So these are the dams that are in the Maharashtra state. And for example, if I click water bodies, I can click that. And all the water bodies are now coming up. So initially there was no water bodies. But now water bodies are coming up. Okay. And then you have a date to see what is the fraction. And you have 22 up to 2022. But let's do a view. And then you can have a number of water bodies fraction where the water bodies are present and the fraction, et cetera. So you could see that the land use land cover is also done for two years. So now if you see the land use land cover is done for two years for Maharashtra, one is from 2006 and another is 2012. So definitely it is at least 10 years lag. So how would you use the 10 year old data as a question? And for that, we need to make our own land use land cover maps either unsupervised or supervised classification based on our research needs. Okay. So let us jump into the open data archive. I hope it has come up. So I will say again, we'll be doing two important phases. Let's go back to the slides and then show you what we will be showcasing. We will be doing, so now we have looked at the applications. Now we'll get into the open source data improvement for water. I have opened the open data archive. And thematic products, we'll be looking at the land and terrain OCM surface water layer products. Okay. So I'm going back to my portal. Okay. So under theme and products. So satellites, you have multiple satellites, but not readily, they are marked as water. Water needs some processing of the data. So you have a multi spectral reflectance curves, or you have hyperspectral images. And from these images, you construct the water bodies. You understand where the water bodies are present in across India at rural scales. So for that, we'll go to theme and products. And we will say either of these. So we have ocean and physical, we have land and vegetation. And then we have land and terrain. So land and terrain includes the water bodies inside the land. Land and vegetation is mostly on the crops and the land. So it's mostly led by crops, whereas ocean and physical includes the ocean and physical products. So mostly in the sea and oceans. Since we are more focused on the rural entities, the rural entities are present mostly inside the land. So that's it. Let us take land and terrain. And if I click this products, you will see multiple products that are kept. Okay. So what we would need to do is we would need to select the one which has relationships to water. So you have snow albedo. Snow cover is converted to water availability in the Himalayan region, because there's a lot of snow. And the snow is volume is then converted into a depth of water, and which goes into the Ganges and other rivers. So that is for that. And mostly for your river discharge, but we'll go to the OCM, OCM surface water layer two products. It's a 2D product, which means it will give you the XY dimensions across space. It will not give you the Z, which is the depth. That is tricky. It's harder to estimate. But for now we will only look at the 2D surface. So it says ocean two water bodies, version number one. So version is how many iterations they have, how many updates they have. So it still is in the version one. It is derived from the ocean color monitor OCM two sensor. And then it is in the visible and NIR bands. So the spectrum that is being analyzed is the visible spectrum and some NIR bands with normalize in the such as NDBEI, NDWI, etc, etc, which are used to extract the water bodies. So based on the reflectance of the water and land into the satellite, they can estimate if it is coming from land, if it is coming from water. Okay, again, this requires you to take the basics of remote sensing. There are a lot of NPTEL courses, I have already mentioned that. Here's slightly advanced, we're looking at for rural development. So then they have used some other knowledge maps through classifications and some categorization of the data. So let's look at the surface water product. Please give it some time to load. And if you could come down, it is still 2014, eight years old document, but it has been used well for the product. Okay, so you have water and no water is zero. Pure water is 200 mixed water is 100. And then you have a spatial resolution of 360 meters. So each pixel is 360 meters. There's no definite amount of temporal resolution. But I think since it's a driven model, you do get it every lag days of one or two days. Okay. And then it shows you how the data has been processed. So first the product comes from the sensor, there's some rectification, atmospheric correction, and then angle normalization because if the satellite is tilted, then you have an elongated image. So all these angles are normalized, atmospheric errors are corrected, the cloud cover everything is corrected, ortho rectification is done. And then cloud shadow masking is taken out the cloud shadow is also casting some images, distortion in the images, that is also removed. And then water bodies are extracted using a combination of NDWI, NDVI, brightness. So all these are indices, indices, which are created by a combination of the bands. So remote sensing tells that color, white color is not white, it is in different bands, with GR, right? So same thing when a pulse, a light pulse gets reflected, it can come in multiple bands. And depending on your sensor's acceptance or absorption, you will collect data. Okay. So here we are using a visible spectrum and NIR spectrum. So you will get some with GR plus NIR infrared. And that is enough to make this NDVI and NDWI. This search for NDWI or NDVI, you will get a lot of materials on how it is calculated. Sometimes I add it here, sometimes I don't, but that is fine. And then you could see here that our raw image is then converted into color composite, where coloring is given blue for water. Okay. And then you have this entire India map extracted with pure water as blue, mixed water as pink color, water pixel under cloud is black, mostly the hilly regions and then background is white. So these are the water bodies based on the reflectance for different dates. Okay. So that is done. But most importantly, you should know that it is multiple bands. So it's like, for example, green minus NIR by green plus NIR. So this is a fraction. And then the fraction converts into an index, right? So that index is used here for calculating the water bodies. That's all this document says. You will learn all of this in the basics of remote sensing again. Okay. So here you could select a year. So you could see that 2016 is the latest. So why I am asking you to learn these from Booghan and other things is so that you could construct the latest data for your research. Here, they've already done it for a time period. And you know that they've used OceanSat 2, which has a near infrared and visible spectrum. So there are multiple data that we will come across in the next lectures. You can construct the same image that I am going to show for a much, much newer date. Here it's only 2016. Okay. And only two months. If it is 2015, you have all the months. So even 2016, only two months are available. Let's take 2015. And then let's say the peak August. Okay. So every third day, as I said, has been given in August. So August 3, 4, 7, 8, 9, 10. So there are some lives and delays. If I see click view, you can see the data now being mapped. So the blue is again the water bodies. The pink is the mixed color and black is no water and incomplete data or water pixel under club. So basically, which is masked. There are some like that. And we know that during August, we get some good floods in Maharashtra. Okay. So you can see that it is there. So let's remove the layer. And you will see an updated layer now. And you can do a view again. So you'll see all these blue, blue colors coming up. Let's select another month. We do know that in December month, it was really flooding in some areas. You see all the floods have come up, excess water. So the surface water getting there. Okay. These actually show that there has been a lot of inundation and water in pure water because of mixing with soil, eroded soil and other resources. So this is very, very important. You could download this data. I will show you how to log in. So if you click log in, it will ask for your username and password. Okay. I already have a username setup, but you can also do a new user or for God password. What you should what you should be knowing is that these passwords are important. And you can go through the new user setup. As I said, you'll have to give the username, your organization, if your government, private sector or what, organization details, first name, last name, gender, address, pin code and what you're going to use it for, focus. You do you want to subscribe for the records? Yes or no. You will get an email link for your acceptance. So you can definitely go and see, but username is there, your email, you could see and then. So let me type in my login so that we can quickly see if it is eligible to log in. So sometimes you have to update your login. Okay. So make sure that you have an updated email. If you forget your login, you can always get the password back by sending a request link. Okay. So I'm going to get the links that are needed for this password. And yeah. In the next class, I will log in and come. I don't want to share my credentials online. So let me log in for the next class and then show you how to extract these data. Okay. So this is how you would log in and collect data for your exercise. And again, you can refresh it to another year, let's say 2013, all months list products, then all the months will come that the data is available. Only now in December is available. It's fine. And then we can do the view. You could have the sea inundation, the water bodies coming up. And then you could do the download button. You need to log in. As I said, it will ask you for the login. But more importantly, you could look at the metadata. So where was this data collected? When was it collected? And then where was it sampled? What was the number of edition? If you have any questions, you could send an email or call them. And then the coverage, the area coverage, those kinds of things. Okay. So any electrification done, what electrification was done? And then it says eight spectral bands, VNIR band with a 60 meter resolution and swath of 142, 1420 kilometers. Number of bands is only one used for this study. And what are the values? 0, 100, 250. The zero is just the background, which is white color, whereas mixed water is pink, pure water is blue, and water pixel under cloud is kept as black. You don't see that here. Okay. But you could see definitely, let's say, near the power engine and stuff. So what do you see here is pink color on the sides. Okay. I hope you could see that in this water bodies, which is blue. And then let me remove it just for the sake of it. You could see this is a water body in a normal map. So the base map is a Bing map or a Google map, which is just a normal map, which is colored, taking on the land use land cover for a particular year. But what has happened is when you add the satellite data, which predicts the water boundary. So sometimes there is overlap and goes beyond. So this is where you can calculate the area of the water body when you download. There is an exercise that we will do to show you how to estimate this water body area and then take the value out. And then you have the pink color. The pink color resembles the mixed water. So which means it's not blue exactly in color. It could be muddy, sandy or brown water. It's because of water mixing and then runoff that comes on the streets. So you could see that on the boundaries, it is always impure. But in the center of the lake, it is still okay. This is the same anomaly for your water bodies also in terms of the rivers. Sometimes this is also considered blue because the boundaries kept like this. So you will see some blue color there. And Bangladesh is also monitored. Okay. So we have come across a layer where readily you could take the data. I would request you to look at what is NDBI and NDWI. And then you will understand how these maps are made for data. So moving on, we will go to the next product. So we have looked at the applications. We have had open data, theme products, land and terrain products have been looked at. Now we will go to program projects, terrestrial science and hydrological products. Here we will have many, many different products. Okay. So we'll have to go back to program and projects. So it resets itself. And in the program and products, national information system for climate environmental studies, MICES. And under that, there is a hydrological terrestrial science. So terrestrial is land, atmospheric is above the land in situ data is the observation data, cryospheric is the cold regions, whereas ocean sciences is there in the ocean. So cryospheric is mostly on the Himalayan regions. Let's say terrestrial. And then here you will see a lot of products. Select a product. So select a project, select the group. In the group, we select terrestrial. And in the terrestrial, what do you want to see? So there is a lot of land products and water products. So for example, we have crop land data, snow, albedo, water bodies fraction, forest fire, forest fires are more land. There's a hydrological products. That's what we're going to see. And then in Indian soil database, land degradation, soil, surface soil moisture, the soil I'll keep for the next class, but we will look at the hydrological products. If you click hydrological products, you will be opening up a new website. So it goes here with some technical documentation, which gives you what is the pixel level. And you could see it is 16.5 kilometers. It's pretty big, but it's good for some applications. Let's look at the hydrological fluxes, spatial resolutions 0.115. So you could see here, on 11th Jan, what are the different hydrological parameters that has been measured? So if you remember in the lecture two, we looked at the water balance for a rural village, which includes the precipitation, surface runoff, storage, and then evapotranspiration, which is the water taken out of the system due to plant and evaporation, and all the other things. So what has happened is, for a particular date, you can look at the data products. So here you don't have data for 2023. Jan, the latest you have is 11th. And then you have three products you can take. Surface runoff, evapotranspiration, and surface soil moisture. There is no rainfall. So rainfall you have to estimate separately. This is only the products that are taken. On the left hand side, you can also look at forecast surface runoff, accumulative surface runoff, climate indices is precipitation index, sanitized runoff index, and then fluxes. So rainfall, you can get it from here for a particular date. So you can see a particular date for a particular year you can take. So let's say 2022, you can take four week, 33 I have selected. It does take some time. And it tells you if there is excess runoff. So basically, from an average, it has a 20-30 year average, and above the average or below the average is the, or 60 years they have taken. So 60 years average and above or below the water level, is it there or not? This is what the product, this product is. There are some other products which looks at only 20 to 30 years, but this product looks at 60 years. And for some reason, it doesn't pick up. So let's do a different year. Now it comes up. Now you can see as basins. And in the basins is the water level for 2017, 25th July, June to July 1st. In that week, is it excess or normal? So deficient, these Ganga region has been deficient. Whereas others have been in excess. Scanty is very, very less rainfall below deficient. And then no rain is black. So this is for the rainfall. Again, we will come back to just the hydrological products. Then we'll take the spatial resolution. Let us take 0.05 degree, which is much, much smaller. Doesn't load because of the internet speed. Let's do again hydrological products. And then we will do 0.15. It regularly opens. Sometimes there is a lag on the, so I'm showing it live for the class. So that is why we cannot record it. So I'm just doing it while I'm connecting the lecture. There is some times it doesn't work. So just be patient with the Bowen website. You have these products. So you can select a different date. Let's select 1st of Jan or 5th of Jan. And then you could see that the surface runoff millimeters per day is there. There is soil moisture available. And then where is the evapotranspiration happening? So there is less runoff here and almost no runoff in the other areas. In that day, so in that day there has been only when there's a rain, there's a runoff. Don't think that all the river is flowing. Why is the runoff not accounted for? That is discharge. But there is rainfall and the rainfall pushes the water in. So when is the big monsoon season? At least in Maharashtra, it is in the June. So let's take June, July, July kind of mid-July. And you could see here now, all the areas green in most part. And western girls where the monsoon is picking up, that is in blue color. There's more surface runoff because there's rainfall and the water brings out the discharge. And then you have, because there's water, there is a positive soil moisture. And you could see the soil moisture also increasing in most parts, blue color. This is like light blue. This color is not black. It's blue, which means that it is above and increasing. And then evapotranspiration, this is driven by the plants, crops, and then trees. So you could see that most of these western girls in other regions is very, very high, which means there is a lot of water that is taken up by the plants and transpired. So you could also download these products. As I said, it is not at real time or near real time scenarios. However, you can learn on how to use this data and then use to do the data. Like use the tools that we're going to teach this class to do the data. And here is the water balance components and how they have created it. So I just clicked on the technical document and you can read it. So as an exercise, when you discuss these in your classwork, in your research work, I recommend you to go through these kind of materials because it gives you a good reading document. And here they're given how they calibrate the model. So the model they're using is VIC variable infiltration capacity model. So I think this is good for introducing the remote sensing data for water from Google resources. That is it for the Google data products. With this, I conclude today's lecture. Again, I am putting the video tutorials for these products. Feel free to go and look at these products for your research. Thank you.